http://www.hdgamedb.com/amazon/history.aspx

sales history graphs now go back to 30 and 45 days instead of just 14 and 7 days.

Gives a new perspective on the magnitude of the Blu-ray and HD DVD surges.One thing the new 45 day trends on the top 10 25 50 100 charts clearly show is that HD DVD sales have not been dropping as the PC mag chart implied, but that they have remained relatively constant and that Blu-ray has only recently been playing at HD DVD sustained level.

THe important thing about HD DVDs level staying high during this period of no new titles is that it implies that most sales are coming from new player owners buying into the existing releases. That implies that the new owners are buying titles at similar attach rates as the older first adopter owners, which may be a more positive sign for HD DVD in the long run than any short term spike in sales.

That leaves open the possibility that a large base of satisfied HD DVD owners with similar attach rate tendencies will buy the new releases at a good pace when they arrive.

But that is the exact situation that exists. Sales of both formats are moved outside of the ranges that are being looked at.Some. For the top 100 discs on each, it is only sales of the lowest discs that aren't counted. While we can take those into account, your example was pretty extreme as far as having the 6th rank disc sell the same number of copies as the 50th or 100th ranked disc.
Top 10,000 or something similar would be what I consider a format average, which is what you questioned me about earlier. Basically all BD vs. all HD-DVD, even though # of titles are different. It leads to a bogus comparison.If you count all discs for both formats, then concentrating sales amongst a small number of them would hurt the average. If you disagree, again you are welcome to come up with a scenario where concentrating them helps when looking at the overall average. In this case you can't use the technique of moving discs outside the number you are looking at since all discs need to be counted, so that they won't help.
Yes, I still believe my statements are true. I looked at 9,000 BD disc sales and 10,000 HD DVD disc sales. I hypothetically spread the HD-DVD sales over 100 titles, with two big sellers, and I spread the BD sales over 50 titles with 5 big sellers. If you look at the Top 10, Top 25, or Top 50, BD sales rank will appear to be better, however, HD-DVD is selling more discs. So sales rank does not show the format selling the best.Again concentrating a large number of sales to 5 discs instead of 2 discs is not what got you the higher ranking for Blu-ray, it was moving lots of sales outside of the range you were looking at. If your statement is true that concentrating them more amongst 5-10 discs helps the average for the top 100, then you should be able to come up with an example that doesn't use moving sales outside of that range in order to lower the average for one format.

Basically, if you want to know if your statement about concentrating them to the top 5-10 titles will help the top 100 rankings, you need to keep other things pretty even, instead of using something else to help the rankings and making it look like concentrating them to the top 5 is what made the difference. Try just concentrating to 2 titles for HD DVD and 5 for Blu-ray and spread the rest to 48 for HD DVD and 45 for Blu-ray and see if that helps the average for the top 50 for Blu-ray. If concentrating helps, then it should. Or spread them amongs 98 and 95 and look at the average for the top 100.

--Darin

There's a thread going in the HD-DVD Software forum that lists release dates in Canada for some pretty heavy hitters like The Matrix and sequels being released on April 10th.

If those dates hold up and BR releases continue to get delayed for lack of BD-J support on everything but the PS3 then I would expect that there could be a pretty big sales surge for HD-DVD in the next couple of months.

On the other hand Casino and some of the other BD release exclusives are really going to hurt for the HD-DVD camp.

If those dates hold up and BR releases continue to get delayed for lack of BD-J support on everything but the PS3 then I would expect that there could be a pretty big sales surge for HD-DVD in the next couple of months. I agree. HD DVD getting the Matrix series exclusively for a while would help it quite a bit.

--Darin

Some. For the top 100 discs on each, it is only sales of the lowest discs that aren't counted. While we can take those into account, your example was pretty extreme as far as having the 6th rank disc sell the same number of copies as the 50th or 100th ranked disc.
If you count all discs for both formats, then concentrating sales amoungst a small number of them would hurt the average. If you disagree, again you are welcome to come up with a scenario where concentrating them helps when looking at the overall average. In this case you can't use the technique of moving discs outside the number you are looking at so that they won't help.
Again concentrating a large number of sales to 5 discs instead of 2 discs is not what got you the higher ranking for Blu-ray, it was moving lots of sales outside of the range you were looking at. If your statement is true that concentrating them more amongst 5-10 discs helps the average for the top 100, then you should be able to come up with an example that doesn't use moving sales outside of that range in order to lower the average for one format.

--Darin

I don't think the example was extreme at all, especiall considiring the small volumes actually in question. I bet Amazon isn't selling more than 1000 discs of each format a week. I exaggerated the numbers to 9000 and 10000 so it would be easier to understand in the simple example. The problem gets worse with 900 and 1000 discs. The example still works if you want to apply a more realistic gradient (but still have 10% of the titles get 50% of the sales).

Can you please show me an Amazon tracking site that shows an "overall" sales rank for each format? I don't know of one. They all chop it at around the Top 100. So why are you trying to force my arguement to cover a scenario that does not exist? Yes, my argument may not hold up when placed against the scrutiny of something that does not exist. I can't say for sure, because that situation DOES NOT EXIST. What I'm trying to say is that mathematically, the two Amazon sales rank tracking site that I know of, may be extremely misleading, because I have found numerous situations that are possible (and perhaps even likely) to exist that would result in one format being reported to have a higher ranking, while the same format actually sells fewer discs than the competing format.

I have already provided one very simple example, just to show it possible. If you can refute that the example, as given, wouldn't result in a higher Top 10 sales rank average for BD, but actually has HD-DVD selling more discs overall, then please point out my error. Also, please stop trying to confine the mathematical model to a situation THAT DOES NOT EXIST. There is no "overall". Also, sales may often fall outside of the broadest range, which is Top 100. To do it your way, you would have to assume that all sales fall within the Top 100, which is completely contrary to my arguement. The grouping of a few titles with a high number of sales is what gives the leading sales rank format a grossly higher average sales rank. The other format having a broader title spread (going outside the Top X in question) is partially how it achieves superior # of discs sold, despite a lower average sales rank.

Finally, for everyone, why we are roughly on the topic, how do the two web tracking sites handle the Top 100 charts if there are fewer than 100 titles that have a sales greater than 0? Basically, what is the lowest possible sales rank? If it is say, 50,000, what happens if items 10,000-50,000 all have a sale of 0 units. What if say, 30 HD-DVD's sold 0 units and some of them had 40k-50k sales ranks, just based on alphabetical listings. On the BD side, a similar situation exists, but they were in the 20k-30k range alphabetically. Does BD come out ahead just based on a alphabetical tie breaker? (Again, I am exaggerating for examples sake).

The hd game db site has a page that you can get the sales rank of all or most all of a formats disk rankings.

I have already provided one very simple example, just to show it possible.We already discussed that having sales outside of the range being considered is a bad thing for an average, for a given number of sales and this is really what was the biggest factor with your scenario. But here is one of your claims:
However, when looking at averages, like top 100, Top 50, Top 10, etc., you are better off having all your sales concentrated into 5-10 discs, than to have them spread out. It will make your average sales rank higher, but you could actually be selling fewer copies of the higher average format.Can you backup that concentrating sales into 5-10 discs helps the top 100 average? In the example scenario you used concentrating sales to 5 discs instead of 2 discs hurt one side, but that was masked by moving more sales outside of the range being looked at. If you want to see if concentrating sales to 5-10 discs helps or hurts then you need to make other things equal, which you didn't even come close to doing. If you want to make some sales fall outside the top 100 range then go ahead, but don't do way more for one side than the other to make it look like the concentrating of sales to 5 discs was a good thing.

Seriously, do you really not understand that your example didn't show that concentrating to less than 10 is a good thing for the top 50 or 100 average?

If you said that concentrating sales to 100 discs helps the top 100 average and to the top 50 discs helps the top 50 average, etc., then that would have been true.

--Darin

The hd game db site has a page that you can get the sales rank of all or most all of a formats disk rankings.

Can you link it? I can't find any page that gives an overall rank by format. Just Top 10, Top 100, etc. No all BD/ All HD page, with sales rank comparisons. They don't include that analysis, because there is no way to make it apples to apples.

Can you link it? I can't find any page that gives an overall rank by format. Just Top 10, Top 100, etc. No all BD/ All HD page, with sales rank comparisons. They don't include that analysis, because there is no way to make it apples to apples.I know the eproductwars site has links at the top for "All Blu-ray" and "All HD DVD". Not sure if that is what Kosty meant.

Those are here:

http://www.eproductwars.com/dvd/hd-dvd.cfm

http://www.eproductwars.com/dvd/blu-ray.cfm

--Darin

The link goes to the main page.

http://www.hdgamedb.com/amazon/history.aspx

I'll desrcibe how to get to it.

1) Click on the HD DVD /Blu-ray Bestsellers tab on the left

2) There is a box (top 10) drop the all sales rank data drop list to show "All Sales rank data (400 movies) or 50 100 250 (it defaults to top 10)

3) Click on the today past 7 days or past 14 days pointers for any title

4) Revel in the fact you can get sales data charts for "Winged Migration" presales on Blu-ray currently 400 in the list with an Amazon sales rank of #110,200 and somehow Amazon is finding a way to seperate it form The Grudge presales currently ranked #110563 :)

We already discussed that having sales outside of the range being considered is a bad thing for an average, for a given number of sales and this is really what was the biggest factor with your scenario. But here is one of your claims:
Can you backup that concentrating sales into 5-10 discs helps the top 100 average? In the example scenario you used concentrating sales to 5 discs instead of 2 discs hurt one side, but that was masked by moving more sales outside of the range being looked at. If you want to see if concentrating sales to 5-10 discs helps or hurts then you need to make other things equal, which you didn't even come close to doing. If you want to make some sales fall outside the top 100 range then go ahead, but don't do way more for one side than the other to make it look like the concentrating of sales to 5 discs was a good thing.

Seriously, do you really not understand that your example didn't show that concentrating to less than 10 is a good thing for the top 50 or 100 average?

If you said that concentrating sales to 100 discs helps the top 100 average and to the top 50 discs helps the top 50 average, etc., then that would have been true.

--Darin

I don't think you understand what i mean by concentrating sales. I'm not saying have one BD disc get all the sales and a rank of 1, and have the next 9 get zero sales, with tied ranks for 49992-50000, with a Top 10 sales rank average of 49996.5.

I'm saying sales have to be concentrated enough on 4-5 discs to get them high into the sales rank (double digits or so, so that when people look at the Top 10, they see many BD discs compared to HD-DVD). Then the rest of the sales are concetrated on 50 or so discs, while the other 110 BD discs go basically unpurchased. Then if HD-DVD discs are spread out (not concentrated) over 150 discs (with some conentration on 1-2 big titles), then when you look at Top 50, Top 25, Top 10 BD may have a higher average sales rank, but HD-DVD actually sold more number of discs across their entire product offering.

The sites should add a Top X vs Top X comparison, where X represents the smaller number of discs available on the two formats. This might make it easier to catch such an error. As is, the error becomes more possible as the number of released titles increases.

The link goes to the main page.

http://www.hdgamedb.com/amazon/history.aspx

I'll desrcibe how to get to it.

1) Click on the HD DVD /Blu-ray Bestsellers tab on the left

2) There is a box (top 10) drop the all sales rank data drop list to show "All Sales rank data (400 movies) or 50 100 250 (it defaults to top 10)

3) Click on the today past 7 days or past 14 days pointers for any title

4) Revel in the fact you can get sales data charts for "Winged Migration" presales on Blu-ray currently 400 in the list with an Amazon sales rank of #110,200 and somehow Amazon is finding a way to seperate it form The Grudge presales currently ranked #110563 :)

And did you read the note at the bottom?

"NOTE: These average sales rank values are for informational purposes only. Since the number of discs on each side can vary, no conclusion can be drawn if the values are compared.

For average sales rank values that can be compared, visit the HD DVD vs. Blu-ray page."

You can't compare this average because it is invalid to do so. This is why this data is not charted. It isn't apples to apples.

I think you guys are sniffing each others tail. The current volumes are sensitive to small variations in there impact on the charts.

The charts are fine as they are, they will give better data as the sales volumes increase

Thats the same wit all the indicators we have, NPD, Nielson, DVD Empire, Amazon fromeprocut wars or hdgamedb. They all are sensitive now to any promotions or small change in sales volume.

That sensitivity will be reduced over time as the volumes increase.

There probably nothing wrong with any of these methodologies. Its just that its too early to tell because the volumes ae still fairly small.

Time will tell us more.

I know the eproductwars site has links at the top for "All Blu-ray" and "All HD DVD". Not sure if that is what Kosty meant.

Those are here:

http://www.eproductwars.com/dvd/hd-dvd.cfm

http://www.eproductwars.com/dvd/blu-ray.cfm

--Darin

Well, they only list available movies (including pre-orders) and don't do any type of analysis or charting of rankings. It also isn't entirely up-to-date and number available for order differs by 1 when compared to the main page.

And did you read the note at the bottom?

"NOTE: These average sales rank values are for informational purposes only. Since the number of discs on each side can vary, no conclusion can be drawn if the values are compared.

For average sales rank values that can be compared, visit the HD DVD vs. Blu-ray page."

You can't compare this average because it is invalid to do so. This is why this data is not charted. It isn't apples to apples. Thats for the studio summary chart on the bottom.

The method I described will show the Amazon sales ranks for all the titles. that is the same as in the HD DVD versus Blu-ray page.

PS I will check witht the site designer, but I think its how I describe it as each title can open up its own little Amazon sales graph with the same numbers.

I think you guys are sniffing each others tail. The current volumes are sensitive to small variations in there impact on the charts.

The charts are fine as they are, they will give better data as the sales volumes increase

Thats the same wit all the indicators we have, NPD, Nielson, DVD Empire, Amazon fromeprocut wars or hdgamedb. They all are sensitive now to any promotions or small change in sales volume.

That sensitivity will be reduced over time as the volumes increase.

There probably nothing wrong with any of these methodologies. Its just that its too early to tell because the volumes ae still fairly small.

Time will tell us more.

this isn't a problem of volume, but distribution. As you see in my example, I use a volume of 19,000 discs, which is probably 10X more than currently (just my guess) for Amazon.

The reason I brought it up was 1) Plaz seems to show HD-DVD selling more discs over a given window, but BD still has a higher dailey sales rank over the same window. How? Is it lag? Or something else? and 2) The Top 10 list at DVD Empire showed a pretty varied list of discs for HD-DVD (HDScape Aquarium, music, and movie discs), while the BD list was basically newer movie releases.

This lead me to look at Amazon rankings and the effect we might see if 1) BD buyers mainly limited their purchases to newer movie releases and 2) HD-DVD buyers bought from the entire spectrum of available titles, with a few new releses getting alot of focus. Under several scenarios, I found that if this were the case, the BD sales ranking could average higher Top 10 through Top 100, but HD-DVD could actually be selling as many or more discs than BD.

Just the fact that this possibility exists should really makes us think twice about gauging the success of either format off of Amazon rankings.

I agree that HD DVD is showing deeper buying of catalog titles.

To me that is a strength as it shows that newer HD DVD player owners are buying the entire catalog and that older owners are still buying something even without new releases.

Blu-ray is much more dependent on hit sales so far. IMHO that's what it is designed to do.

Can you link it? I can't find any page that gives an overall rank by format. Just Top 10, Top 100, etc. No all BD/ All HD page, with sales rank comparisons. They don't include that analysis, because there is no way to make it apples to apples.The analysis (summary) at the bottom of the All BD/All HD (when on the "Best Sellers" page, select the "All Sales Rank Data" option from the second drop down list) is exactly the same as for the Top 10/25/50/100 lists except that there are no leaders highlighted in blue or red because the lists are not the same length for both sides.

Quote:
Originally Posted by wnorris
And did you read the note at the bottom?

"NOTE: These average sales rank values are for informational purposes only. Since the number of discs on each side can vary, no conclusion can be drawn if the values are compared.

For average sales rank values that can be compared, visit the HD DVD vs. Blu-ray page."

You can't compare this average because it is invalid to do so. This is why this data is not charted. It isn't apples to apples. Thats for the studio summary chart on the bottom.

The method I described will show the Amazon sales ranks for all the titles. that is the same as in the HD DVD versus Blu-ray page.

PS I will check witht the site designer, but I think its how I describe it as each title can open up its own little Amazon sales graph with the same numbers.

Its as I thought. that disclaimer only applied to the studio chart, it was meant to highlight the fact that there was no red or blue highlighting on that chart. to get the red or blur highlighting you have to go to the "HD DVD vs. Blu-ray page" THe averages are not highllighted.

For average sales rank values that can be compared, visit the HD DVD vs. Blu-ray pageThis average sales rankings in the table is meant to have the disclaimer.

But the individual title Amazon Sales Rank is available for all 400 titles and you can graph out everyone of them. that's what you were asking about...

Its as I thought. that disclaimer only applied to the studio chart, it was meant to highlight the fact that there was no red or blue highlighting on that chart. to get the red or blur highlighting you have to go to the "HD DVD vs. Blu-ray page" THe averages are not highllighted.

This average sales rankings in the table is meant to have the disclaimer.

But the individual title Amazon Sales Rank is available for all 400 titles and you can graph out everyone of them. that's what you were asking about...

That isn't what I was asking about. Earlier, DarinP was talking about a Top X analysis of the data, where X included all titles from both formats: "looking at the overall average" [of an entire format]. I said I didn't believe such an analysis existed, because it would be flawed.

For awhile, I thought HDGameDB removed the overall average from the page you linked. It appears to be back now, with a disclaimer. The DVD Wars doesn't include it at all.

That isn't what I was asking about. Earlier, DarinP was talking about a Top X analysis of the data, where X included all titles from both formats: "looking at the overall average" [of an entire format]. I said I didn't believe such an analysis existed, because it would be flawed.

For awhile, I thought HDGameDB removed the overall average from the page you linked. It appears to be back now, with a disclaimer. The DVD Wars doesn't include it at all. Gotcha. I think we both agree now.

I glad I could point out where you could find the individual amazon stats for those long tail titles.

Which format do you guys think has more sales right now outside of its top 100 (meaning they don't even show up in the rankings with looking at top 100 vs top 100)? One thing we can look at is that right now the eproductwars site shows that the 101st title for HD DVD (out of 184) is "Four Brothers" at #12660, while the 101st title for Blu-ray (out of 217) is "The Italian Job" at #8112. For a spot check, titles at spot 150 for each have #23378 on HD DVD and #13213 on Blu-ray. If anything, that would seem to support that Blu-ray has more sales that aren't counting in the top 100 comparison to me, at least right now, and not the other way around.

BTW: I was looking at the in stock numbers for the XBOX360 add-on and the PS3 remote earlier today and it looks like those numbers just changed. For the XBOX360 it changed from 920 to 882, for a drop of 38. The PS3 remote dropped from 1577 in stock to 1314 in stock, for a drop of 263. Yes, I know that there are those who will claim that over half of those buying the PS3 remote won't be using it for Blu-ray movies, but those seem like significant numbers if they hold up to me. Look at the history that looks like a bigger drop than normal for the PS3 remote, but it is hard to know how often they restock these things.

--Darin

The reason I brought it up was 1) Plaz seems to show HD-DVD selling more discs over a given window, but BD still has a higher dailey sales rank over the same window. How? Is it lag? Or something else?I don't remember what time period he looked at, but from looking at the top 10 chart for the last 24 hours, HD DVD did gain some ground on Blu-ray during part of the day. There is a historical component to these sales and we shouldn't expect one to show up as higher just because it outsold the other for a few hours unless than overcomes sales from earlier (which look like they do get discounted with time).

--Darin

All this work in response to Grubert's chart he posted earlier today showing the bottom had fallen off of HD DVD on Amazon :eek:

Not so. I was merely showing that HD DVD supporters had proclaimed the format's 'comeback' a tad too soon.

BTW, I see there's a lot of talk of amazon rankings and how they translate into actual amazon sales, which is not the topic of this thread.

These are the new figures from HMM for the week ending Jan 28:

YTD: BD 100.00 HD 49.21
SI: BD 87.76 HD 100.00

Combining this with the weekly figures from PC Magazine (http://www.pcmag.com/article2/0,1895,2093290,00.asp), BD/HD DVD percentages are:

Day Week YTD SI
01/07 63.3/36.7
01/14 68.2/31.8
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

These are the Top 5 for BD and HD DVD for the week ending Jan 28 according to Nielsen.

Top 5 BD
1. Saw III 100.00
2. The Guardian 56.63
3. Crank 20.50
4. Saw II 17.00
5. Superman Returns 16.63

Top 5 HD
1. Batman Begins 100.00
2. Clerks II 69.30
3. Lucky Number Slevin 66.92
4. Superman Returns 54.11
5. The Mummy Returns 48.26


************

And the Top 10 for the week ending February 4, given by davisdvd, from Rentrak's Home Video Essentials:

BD
1 (-) Open Season
2 (-) Flyboys
3 (4) Crank
4 (3) Saw III
5 (2) The Guardian
6 (-) Terminator 2
7 (-) Superman Returns
8 (-) Employee of the Month
9 (-) Mission Impossible Collection
10 (7 ) Gridiron Gang

HD
1 (6) Superman Returns
2 (2) Batman Begins
3 (4) V For Vendetta
4 (-) Serenity
5 (3) Lucky Number Slevin
6 (-) The Last Samurai
7 (1) Clerks II
8 (-) Fear & Loathing in Las Vegas
9 (-) Mission Impossible III
10 (7) The Mummy Returns

These are the new figures from HMM for the week ending Jan 28:

YTD: BD 100.00 HD 49.21
SI: BD 87.76 HD 100.00

Combining this with the weekly figures from PC Magazine (http://www.pcmag.com/article2/0,1895,2093290,00.asp), BD/HD DVD percentages are:

Day Week YTD SI
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3


Grubert, thank you very much for info !!!

Marek

So Blu-ray gained about 1.6% into HD DVD since inception numbers for the week?

And the rate is about the same as last week?

Even though that 2:1 in favor of Blu-ray, the volumes seem to be relatively small to change the SI numbers that little.

Even with all of the discounts, 2 for 1, rebate incentives and new buyer enthusiasm, on both sides, that seems to be a relatively small change implying low volumes compared to the SI number.

Is that rate faster or slower relative to the three weeks prior?

Or do we not know because we have to through out the first couple weeks?

I'm slightly confused because I don't know where the combo corrected numbers are?

Do we have to through out the first couple weeks of data?

Does the PC mag weekly charts include the effect on combos for all of the historical data?

Now that we have two "good" weeks of data, it still looks like Blu-ray is pulling away with YTD and edging closer to SI. So the data without the combos was correct (for trending), just not as drastic as it truly was.

it still looks like Blu-ray is pulling away with YTD and edging closer to SISo Blu-ray gained about 1.6% into HD DVD since inception numbers for the week? 1.6% gain in SI. 3.2% closing rate.

So in two or three weeks if nothing changes, the gap will have been closed.

In a month and a half the gap will be what it is now in the other direction.

If nothing changes.

Does the PC mag weekly charts include the effect on combos for all of the historical data?

It looks like it.

Just look at their Jan 7 data: 63.3/36.7. This must be also the YTD ratio for the Jan 1-Jan 7 period. Right?

So it follows that during that period, for every 100 BDs sold there were 57.98 HD DVDs sold. Whereas the HMM stat for that period was 100.00 for bd and 47.14 for HD DVD (not including combos). So for that period, of every 100 HD DVDs sold, 81.3 were non-combos and 18.7 were combos.

Using that ratio for the other fields, we can do a fairly accurate correction of the other figures:

Week ended Jan 7
YTD: BD 100.00, HD 57.98 (actual)
SI: BD 69.15, HD 100.00 (extrapolated)

Week ended Jan 14
YTD: BD 100.00, HD 47.18 (extrapolated)
SI: BD 75.12, HD 100.00 (extrapolated)

Now that we have two "good" weeks of data, it still looks like Blu-ray is pulling away with YTD and edging closer to SI. So the data without the combos was correct (for trending), just not as drastic as it truly was.

And the temporary closing up in the Amazon sales rank charts has ended.

It looks like it.

Just look at their Jan 7 data: 63.3/36.7. This must be also the YTD ratio for the Jan 1-Jan 7 period. Right?

So it follows that during that period, for every 100 BDs sold there were 57.98 HD DVDs sold. Whereas the HMM stat for that period was 100.00 for bd and 47.14 for HD DVD (not including combos). So for that period, of every 100 HD DVDs sold, 81.3 were non-combos and 18.7 were combos.

Using that ratio for the other fields, we can do a fairly accurate correction of the other figures:

Week ended Jan 7
YTD: BD 100.00, HD 57.98 (actual)
SI: BD 69.15, HD 100.00 (extrapolated)

Week ended Jan 14
YTD: BD 100.00, HD 47.18 (extrapolated)
SI: BD 75.12, HD 100.00 (extrapolated)

Can you do your weekly sales ratio calculation for those two extrapolated weeks of Jan 7 and 14th and update the complete chart with asterisks?

Thanks for all the work. :)

Even though that 2:1 in favor of Blu-ray, the volumes seem to be relatively small to change the SI numbers that little.
I'm not sure about that - 1.6% change would indicate 4.8% of total sales happened in the last week (at 2:1, if I'm thinking that through right), its not a small proportion of the total HD sales at all IMO.

The HD market is of course small in itself, but thats another issue.

Can you do your weekly sales ratio calculation for those two extrapolated weeks of Jan 7 and 14th and update the complete chart with asterisks?

Thanks for all the work. :)

No problem. Just don't complain if they don't make sense. ;)

Day Week YTD SI
01/07 63.3/36.7 63.3/36.7 40.9/59.1*
01/14 68.2/31.8 67.9/32.1* 42.9/57.1*
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

So a complete chart using the numbers balanced by the combo correction would be:

Date YTD (Change) SI (SI Change)
BD/HD BD/HD

01/07 100/57.98* ( ) 69.15/100* ( )
01/14 100/47.18* (+10.80) 75.12/100* (5.97)
01/21 100/50.51 (-03.33) 82.30/100 (7.18)
01/28 100/49.21 (+01.30) 87.76/100 (5.46)



* extrapolated figures

These are the new figures from HMM for the week ending Jan 28:

YTD: BD 100.00 HD 49.21
SI: BD 87.76 HD 100.00

Combining this with the weekly figures from PC Magazine (http://www.pcmag.com/article2/0,1895,2093290,00.asp), BD/HD DVD percentages are:

Day Week YTD SI
01/07 63.3/36.7
01/14 68.2/31.8
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3


Well, this casts doubt on the theory that the numbers SCEA gave out were based on any Nielsen data. If they were, you would see something close to 100/100 for SI.

So SCEA gave out a number showing BD 1000 discs in the lead and even paidgeek gave out a number showing BD several thousand discs in the lead. If these numbers came from Nielsen, then we would see an SI in BD's favor. So this weeks numbers seem to rule that out, so where is Sony getting its numbers. Are they just making up something that looks good, based on survey results or something?

No problem. Just don't complain if they don't make sense. ;)

Day Week YTD SI
01/07 63.3/36.7 63.3/36.7 40.9/59.1*
01/14 68.2/31.8 67.9/32.1* 42.9/57.1*
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3
So the SI changes for each week would be:
Day SI (SI Change)

01/07 40.9/59.1*
01/14 42.9/57.1* (2.0)
01/21 45.1/54.9 (2.2)
01/28 46.7/53.3 (1.6)Isn't that a slowdown in the rate of the (SI) Since Inception figure closing rate?

It looks like it.

Just look at their Jan 7 data: 63.3/36.7. This must be also the YTD ratio for the Jan 1-Jan 7 period. Right?

So it follows that during that period, for every 100 BDs sold there were 57.98 HD DVDs sold. Whereas the HMM stat for that period was 100.00 for bd and 47.14 for HD DVD (not including combos). So for that period, of every 100 HD DVDs sold, 81.3 were non-combos and 18.7 were combos.

Using that ratio for the other fields, we can do a fairly accurate correction of the other figures:

Week ended Jan 7
YTD: BD 100.00, HD 57.98 (actual)
SI: BD 69.15, HD 100.00 (extrapolated)

Week ended Jan 14
YTD: BD 100.00, HD 47.18 (extrapolated)
SI: BD 75.12, HD 100.00 (extrapolated)

Isn't this extrapolation extremely flawed for SI numbers? Basically, if Nielsen just discovered combos were being omitted, then the data needed to be corrected all the way back to April 2006. I don't think it is reasonable to look at the ratio of combos and non-combos for one week, and say that applies all the way back to April 2006.

I might be willing to buy that you can extrapolate new YTD numbers (obviously you could for week 1, so it isn't a stretch to do it for week 2). However, I think it is seriously flawed to do so for SI.

If Grubert is the keeper of the official chart, then I think in all fairness, no corrected SI numbers should be included, and really no corrected YTD either, but I wouldn't put up much of an arguement if YTD was included for the two weeks.

EDIT: It also just crossed my mind, but isn't each week essentailly weighted? For example, if there are 100,000 total disc sales one week and the ratio is 100/80 in favor of HD-DVD and then the next week there are 50,000 disc sales and the ratio is 80/100 in favor of BD, then the SI wouldn't be 100/100, it would still favor HD-DVD.

When you extrapolate data in the manner you did, isn't that basically assuming all weeks are equal, when in fact, they are weighted by the total number of sales? So you really can't extrapolate the SI based on a single week of data. The more weeks you include in a YTD extrapolation, the more flawed it could become.

I also think there is a possibility that Nielson has also undercounted the total of last years combos and thus has undercounted the SI numbers.

If Grubert is the keeper of the official chart, then I think in all fairness, no corrected SI numbers should be included, and really no corrected YTD either, but I wouldn't put up much of an arguement if YTD was included for the two weeks.

I agree. I'm not putting extrapolated data on the first post.



EDIT: It also just crossed my mind, but isn't each week essentailly weighted? For example, if there are 100,000 total disc sales one week and the ratio is 100/80 in favor of HD-DVD and then the next week there are 50,000 disc sales and the ratio is 80/100 in favor of BD, then the SI wouldn't be 100/100, it would still favor HD-DVD.


Exactly. That's why we need weekly, YTD and since inception figures.

(Of course the absolute best thing would be to have real volume numbers, but beggars can't be choosers.)

I'm not sure about that - 1.6% change would indicate 4.8% of total sales happened in the last week (at 2:1, if I'm thinking that through right), its not a small proportion of the total HD sales at all IMO.

The HD market is of course small in itself, but thats another issue.

Since inception the total number of disks sold according to Nielson is around 1M for both formats combined (slightly less).

5% of that would be around 50,000 titles sold for the week. With a 2:1 ratio that would put it at 16,666 HD DVD titles sold and 33,333 BD titles sold.

So, per day that would be 4761 BD titles sold and 2380 HD DVD titles sold.. For a total of 7141 disks sold a day. This would be all sources covered by Nielson and based on the 1.6% change in SI total and assuming that 1M total titles were sold by the end of Jan (Around 950K is what was told to me by an insider (non HD DVD) was the total disk count according to Nielson/Videoscan).

As a reference point. Within a random 24 hour period that I picked on Amazon - 8:30AM 2/11 to 8:30AM 2/12 BD total disk sales were 617 and HD DVD Total sales were 546., for a percent share of 53% for BD and 47% for HD DVD. So, the total disks sold in that 24 hour period was 1,163. That would make it around 16% of Videoscan avg. on a day. Clearly, I am only looking at one day and I have no idea how that 24 hour period compares to the other days. But it is a data point.

Why am I bringing this up? Amazon is not covered by Videoscan (at least not directly). Their sales ratio may not line up with Videoscan, and it does appear that they are doing a sizeable share of the business. 16% of avg. is pretty significant.

My lab experiment was not to discount Videoscan, but to:

1. See if Amazon is a major source of purchase (compare daily volume on Amazon v. Videoscan)
2. See how their avg. rankings translate to actual sales

I am not implying that one days worth of data from Amazon is solid eveidence. But it certainly is a data point that we should consider when examnining the overall impact of Videoscan and what these numbers mean.

Cheers.

We were also talking about sales of MI3 previously. HMM says 36,000 copies of MI3 (including copies in box sets) sold for BD and HD-DVD combined, through Nov. 30th.

Talking about DVD, HMM also says "It should be noted, however, that Videoscan does not include DVD unit sales from Wal-Mart, which reportedly accounts for accounts for 40% of all domestic DVD retail sales."

So Wal-Mart alone accounts for 40% of all DVD retail sales. This means that Videoscan's claimed 60% of B&M retailers actually represents less than 60% of the market (Wal-Mart is 40%, Amazon is 5-10%, and there are several other major web only retailers not tracked by Videoscan), even for established DVD.

It sounds like the first format to introduce reasonably priced HD-DVD players at Wal-Mart, and get a decent selection of discs at Wal-Mart, may be the one to win the format war. Winning over Wal-Mart (and I mean all Wal-Mart stores, not just one or two test stores) could instantly win you 40% of the market.

Since inception the total number of disks sold according to Nielson is around 1M for both formats combined (slightly less).

5% of that would be around 50,000 titles sold for the week. With a 2:1 ratio that would put it at 16,666 HD DVD titles sold and 33,333 BD titles sold.

So, per day that would be 4761 BD titles sold and 2380 HD DVD titles sold.. For a total of 7141 disks sold a day. This would be all sources covered by Nielson and based on the 1.6% change in SI total and assuming that 1M total titles were sold by the end of Jan (Around 950K is what was told to me by an insider (non HD DVD) was the total disk count according to Nielson/Videoscan).

As a reference point. Within a random 24 hour period that I picked on Amazon - 8:30AM 2/11 to 8:30AM 2/12 BD total disk sales were 617 and HD DVD Total sales were 546., for a percent share of 53% for BD and 47% for HD DVD. So, the total disks sold in that 24 hour period was 1,163. That would make it around 16% of Videoscan avg. on a day. Clearly, I am only looking at one day and I have no idea how that 24 hour period compares to the other days. But it is a data point.

Why am I bringing this up? Amazon is not covered by Videoscan (at least not directly). Their sales ratio may not line up with Videoscan, and it does appear that they are doing a sizeable share of the business. 16% of avg. is pretty significant.

My lab experiment was not to discount Videoscan, but to:

1. See if Amazon is a major source of purchase (compare daily volume on Amazon v. Videoscan)
2. See how their avg. rankings translate to actual sales

I am not implying that one days worth of data from Amazon is solid eveidence. But it certainly is a data point that we should consider when examnining the overall impact of Videoscan and what these numbers mean.

Cheers.

Plaz,

The only problem I have with your analysis is that I don't think we can believe the combined ~1 million numbers that we are seeing from BD insiders. If they were true, or the SCEA numbers, then the latest Nielsen data should have been much closer to 100/100 than it was. This would mean they are getting their numbers from somewhere other than Nielsen. Your analysis uses these BD insider numbers (which I now believe are non-Nielsen) with Nielsen sales ratios. This makes for a completely flawed analysis.

Going back to dvdempire:

last week HD DVD closed back on Blu-ray in weekly sales

Blu-ray still leads, but it closed to a few points

15% swing between last week and this week :eek:


Percentage of total Hi-Def sales:
HD-DVD Blu-ray

Week of Feb. 06th 48.17% 51.83%

Week of Jan. 30th 33.33% 66.67%
Week of Jan. 23rd 41.99% 58.01%
Week of Jan. 16th 45.83% 54.17%
Week of Jan. 09th 43.87% 56.13%
Week of Jan. 02nd 49.32% 50.68%
Week of Dec. 26th 43.11% 56.89%
Week of Dec. 19th 35.38% 64.62%

http://www.dvdempire.com/index.asp?userid=99365291527966&tab_id=61&site_id=68&site_media_id=0


Week of Feb. 06th just got updated to 41.91% / 58.09 %

Amazon has been discounted as a relevant tracking source so many times; must we keep bringing it up?

Going back to dvdempire:




Week of Feb. 06th just got updated to 41.91% / 58.09 %

Yeah, but several of the HD-DVD music discs are still on backorder. As soon as those puppies get in stock, we might see another 7% surge.

Go Toto! :D

Plaz,

The only problem I have with your analysis is that I don't think we can believe the combined ~1 million numbers that we are seeing from BD insiders. If they were true, or the SCEA numbers, then the latest Nielsen data should have been much closer to 100/100 than it was. This would mean they are getting their numbers from somewhere other than Nielsen. Your analysis uses these BD insider numbers (which I now believe are non-Nielsen) with Nielsen sales ratios. This makes for a completely flawed analysis.

I was told it was indeed Neilson/Videoscan numbers. I specifically asked if they were and I got an affirmation to the same....

Why do you believe my analysis is completely flawed? For the 24 hour period on AMZN, I'm not seeing a 2:1 ratio. Perhaps at Best Buy it is. I know Videoscan collects data from BBY, but not AMZN. So, where you get your data may be important....

DVDEmpire also shows a less than 2:1 advantage....

Even with all of the discounts, 2 for 1, rebate incentives and new buyer enthusiasm, on both sides, that seems to be a relatively small change implying low volumes compared to the SI number.

On the contary, 1.6% is a massive change! The volumes will always be low compared to SI and as time goes on the relative volume will drop

Amazon has been discounted as a relevant tracking source so many times; must we keep bringing it up?

My opinion is that Amazon, DVD Empire, and even Nielsen are not relevant tracking sources at this time. In that light, this thread is pointless and should be closed (along with all the other trying to track sales).

No one knows where the bulk of discs are being sold. There have been numerous flaws in the way the data is reported. They all have aspects that just don't make sense (DVD Empires crazy best sellers list for HD-DVD, Nielsen possibly only capturing 30% of the market, and Amazon having the possibility of showing one format with a higher sales rank for a given day, even though it may have sold less discs that day).

Everyone is trying to use this data as ammo for their marketing gun. Most don't realize they are firing blanks. The down side is the mass consumer just duck for cover and hide to it all sorts out, which only makes it take longer to sort out.

On the contary, 1.6% is a massive change! The volumes will always be low compared to SI and as time goes on the relative volume will drop

Only if you are talking about large volumes....

I was told it was indeed Neilson/Videoscan numbers. I specifically asked if they were and I got an affirmation to the same....

Why do you believe my analysis is completely flawed? For the 24 hour period on AMZN, I'm not seeing a 2:1 ratio. Perhaps at Best Buy it is. I know Videoscan collects data from BBY, but not AMZN. So, where you get your data may be important....

Step 1 of your process requires you to take a percentage of an amount of combined units sold that you believe to be a Nielsen number. However, the data released to HMM by Nielsen would indicate that your number of combined units sold may not be correct.

The combined number you are referring to seems oddly the same as the one reported by paidgeek (a BD insider and Sony employee), and closely releated to the one pronounced by SCEA (Sony's gaming division). If SCEA's estimate had been based on Nielsen data, then this weeks publsihed SI ratio would have practically been 100/100. By paidgeeks account, the next published SI number would be several points in favor of BD. However, the SI just published in HMM did not support SCEA's claim. This makes me doubt paidgeek's claim (another Sony source) as well, or at least makes me doubt it is based on Nielsen data (they may have another reporting source we are unaware of). This casts a lot of doubt on any sales info coming from the BD side (and you said your source was non HD-DVD) until we can learn the source.

Since step 1 of your analysis assumes less than 1 million combined sales (the same combined sales as paidgeek and SCEA), you analysis would be flawed if your source for the info was the same as paidgeek or SCEA's source. The fact currently is we have no idea how many combined nits Nielsen tracks for SI. It may be ~1 mil, or 500,000, or 2 mil. I don't think anyone can say for certain.

Of course, another possibility would be that Nielsen reports one set of data to certain paying customers, and a different set of data to be reported to media outlets. I don't believe this to be the case, but if true, would destory Nielsen's credibility.

Step 1 of your process requires you to take a percentage of an amount of combined units sold that you believe to be a Nielsen number. However, the data released to HMM by Nielsen would indicate that your number of combined units sold may not be correct.

The combined number you are referring to seems oddly the same as the one reported by paidgeek (a BD insider and Sony employee), and closely releated to the one pronounced by SCEA (Sony's gaming division). If SCEA's estimate had been based on Nielsen data, then this weeks publsihed SI ratio would have practically been 100/100. By paidgeeks account, the next published SI number would be several points in favor of BD. However, the SI just published in HMM did not support SCEA's claim. This makes me doubt paidgeek's claim (another Sony source) as well, or at least makes me doubt it is based on Nielsen data (they may have another reporting source we are unaware of). This casts a lot of doubt on any sales info coming from the BD side (and you said your source was non HD-DVD) until we can learn the source.

Since step 1 of your analysis assumes less than 1 million combined sales (the same combined sales as paidgeek and SCEA), you analysis would be flawed if your source for the info was the same as paidgeek or SCEA's source. The fact currently is we have no idea how many combined nits Nielsen tracks for SI. It may be ~1 mil, or 500,000, or 2 mil. I don't think anyone can say for certain.

Of course, another possibility would be that Nielsen reports one set of data to certain paying customers, and a different set of data to be reported to media outlets. I don't believe this to be the case, but if true, would destory Nielsen's credibility.

Perhaps the data did not include combos in the total for SPE? But he did say it was Nielson...and the % reported was right (for week).

Perhaps the data did not include combos in the total for SPE? But he did say it was Nielson...and the % reported was right (for week).


I don't know what to tell you Plaz. I reread his post and he did say a 67% market share for BD, which is what it turned out to be. However, your explanation of not including combos doesn't make much sense. If combos weren't included in paidgeek's numbers, then the purblished report that does include combos, would be unlikely to come out to the same 67%, since Superman Returns (a combo) was a Top 5 seller. It would mean an equal error would have to be found and corrected on the BD side. Assuming no BD error, why would he have a YTD market share percentage that did include combos, but total sales that didn't?

Also, if the number of discs sold reported by Paidgeek was correct, the SI would be ~ 100/94 with BD having the lead. However, the Nielsen numbers still show HD-DVD as having the SI lead, which means Paidgeek's number of discs sold could not be correct, or at least aren't based on Nielsen numbers. The other possibility is that the 450,000+ numbers are actually YTD numbers he was reporting. But again, the YTD ratio would not match up to what is reported by Nielsen.

So either paidgeek posted some incorrect information, or he made a lucky guess with the 67% figure. Or if you want to be more sinister, the most convincing lie is the one served with a little bit of truth...

When they say Wal Mart data is included only through 2001, I am assuming it means, it is available until 2001, not after 2001.

When they say some internet sites. I am assuming they are aware of their limited coverage. Or they would have said most, or major or used some other better sounding term.

Obviously I don't know who or what they track. But AFAIK, neither Amazon nor Walmart provide PoS data to Videoscan. Someone who works for Videoscan should probably verify this for us pretty easily....
Well you know what it means to ASSume, don't you Plaz?

Actually, I first read it that way as well.

But look at the context...

Complete sales data in all retail channels are available as far back as January 1999 (however Wal*Mart data is included only through 7/28/01). Prior to January 1999, limited information is available in most channels going back to 1993.

It's not perfectly written, by any means, but standard writing technique would dictate that the text in parentheses is in response to the text prior and it seems very clear that this is indeed the case here. Instead of saying, "through 7/28/01," they should have said something more like, "back to..."

I don't wish to ASSume myself, but it seems much more logical that they mean to say that you can only find data for Wal-Mart from now back to 7/28/01, or from 7/28/01 through to the present.

Well you know what it means to ASSume, don't you Plaz?

Actually, I first read it that way as well.

But look at the context...



It's not perfectly written, by any means, but standard writing technique would dictate that the text in parentheses is in response to the text prior and it seems very clear that this is indeed the case here. Instead of saying, "through 7/28/01," they should have said something more like, "back to..."

I don't wish to ASSume myself, but it seems much more logical that they mean to say that you can only find data for Wal-Mart from now back to 7/28/01, or from 7/28/01 through to the present.

Why is this still being discussed? We have direct quotes from a Wal-Mart rep, published in Video Business magazine that says Wal-Mart does not provide its sales data Nielsen Videoscan (or Soundscan, or any of their divisions). They stopped in 2001. Also late in 2000 ACNielsen merged with Videoscan to form a new business. Wal-Mart was partnered with Videoscan, but didn't do business with ACNielsen. After the merger, Wal-Mart dropped both. We also have Home Media Magazine (the source of all these Nielsen numbers we have) stating in their December 31st, 2006 edition that Wal-Mart does not report to Nielsen videoscan.

End of story, no assuming necessary. It is a FACT that WAL-MART DOES NOT report its' sales data to Nielsen.

While we are speaking of Wal-Mart. Here is Wal-Mart online's list of combined (BD and HD-DVD) best sellers. I don't know how Wal-Mart arrives at their ranks, but here they are:

1. Sopranos: Season Six - Part 1 (HD-DVD)
2. Departed (BD)
3. Departed (HD-DVD)
4. Fear and Loathing In Las Vegas (HD-DVD)
5. MI Ultimate Collection (HD-DVD)
6. MI Ultimate Collection (BD)
7. Saw III (BD)
8. Hollywoodland (HD-DVD)
9. Princess Bride (BD)
10. Poseiden (HD-DVD)

Seems pretty evenly distributed, with the slight edge going to HD-DVD. I can't believe Sopranos is their #1 seller. I guess with their free site-to-store shipping, it makes it the 2nd best price on the net if you are willing to go to the store for pickup.

Why is this still being discussed? We have direct quotes from a Wal-Mart rep, published in Video Business magazine that says Wal-Mart does not provide its sales data Nielsen Videoscan (or Soundscan, or any of their divisions). They stopped in 2001. Also late in 2000 ACNielsen merged with Videoscan to form a new business. Wal-Mart was partnered with Videoscan, but didn't do business with ACNielsen. After the merger, Wal-Mart dropped both. We also have Home Media Magazine (the source of all these Nielsen numbers we have) stating in their December 31st, 2006 edition that Wal-Mart does not report to Nielsen videoscan.

End of story, no assuming necessary. It is a FACT that WAL-MART DOES NOT report its' sales data to Nielsen.
Well the wording makes sense now, thanks for clearing that up.

Now if I had only read your post following the one from Plaz that I quoted, I could've saved us both some trouble. I must admit I stopped reading every post when the thread started filling up with 24hr. sales rankings. I had hoped that our discoveries of flawed data and gathering techniques would mellow us out a bit, but I'm afraid the opposite has occurred.

I just thought I'd paste this from the Videoscan home page for clarity.

"Sell-through POS (consumer purchase) sales data are collected weekly from traditional channels of video distribution including mass merchants, audio/video and video specialty retailers, electronics outlets, grocery stores, drug stores, and some Internet sites. Nielsen VideoScan maintains this weekly POS data in one of the largest databases of VHS and DVD products in the country. Data is collected on more than 40,000 VHS items and 12,000 DVD items. Virtually every UPC code in sell-through release since 1993 can be found in our extensive database. Complete sales data in all retail channels are available as far back as January 1999 (however Wal*Mart data is included only through 7/28/01). Prior to January 1999, limited information is available in most channels going back to 1993."

source: http://www.videoscan.com/about.html

If Videoscan don't extrapolate for the retail market that do not report to them then does anyone know approximately what % of the retail market Videoscan accounts for? Apologies if this has been given out earlier in this thread.

Well you know what it means to ASSume, don't you... ?
.... :D

Nope. Wal-Mart pulled out of the system on that date.

I agree that that is one of the confusing disclaimers I have ever seen.

Wonder if they meant it to be? ;)

Nope. Wal-Mart pulled out of the system on that date.
Yeah, got it. Thanks Kosty. See prior posts.

Now you're the one not reading all the posts. :)

... and Amazon having the possibility of showing one format with a higher sales rank for a given day, even though it may have sold less discs that day).Are you referring to plazman's numbers or the part about some sales being for lower ranking titles outside the range of titles being averaged? Even ignoring the nonlinearity and averaging, it shouldn't surprise us that one can sell more in a day and not be ahead given that sales rankings don't only count sales for the day (e.g., yesterday's are given some weighting). Also, orders of things that are out of stock count, but don't show up in stock changes. And of course preorders count.

--Darin

Yeah, got it. Thanks Kosty. See prior posts.

Now you're the one not reading all the posts. :)yeah, the page refresh thing is a bit tough for these poor old grizzled paws. :)

P.S. it snookered me too, thats the way I first read it. ;)

Amazon has been discounted as a relevant tracking source so many times; must we keep bringing it up?

There is no good reason to discount Amazon results. They are a high volume retailer and their results jive with what is being reported elsewhere. The mistake people make with Amazon is looking at the hourly info and/or the top 10 info and assuming it has relevance to what's happening in total. Instead, they should look at the top 100 discs over a past week, past 2 weeks, or past month time frame.

I just thought I'd paste this from the Videoscan home page for clarity.

"Sell-through POS (consumer purchase) sales data are collected weekly from traditional channels of video distribution including mass merchants, audio/video and video specialty retailers, electronics outlets, grocery stores, drug stores, and some Internet sites. Nielsen VideoScan maintains this weekly POS data in one of the largest databases of VHS and DVD products in the country. Data is collected on more than 40,000 VHS items and 12,000 DVD items. Virtually every UPC code in sell-through release since 1993 can be found in our extensive database. Complete sales data in all retail channels are available as far back as January 1999 (however Wal*Mart data is included only through 7/28/01). Prior to January 1999, limited information is available in most channels going back to 1993."

source: http://www.videoscan.com/about.html

If Videoscan don't extrapolate for the retail market that do not report to them then does anyone know approximately what % of the retail market Videoscan accounts for? Apologies if this has been given out earlier in this thread.
From the Video Business article, May, 2001...

http://www.videobusiness.com/article/CA619951.html

Wal-Mart represents about 20% of VHS sales nationally, a lower percentage of DVD sales and an estimated 20%-25% of videogame sales, according to industry observers.

NPD will continue to collect videogame info from major retailers such as Toys R Us, Best Buy, Babbage's, Electronics Boutique, Circuit City, CompUSA, Kmart, Target and Sears. Its videogame retail sample represents more than 80% of the total U.S. market, but that will change with Wal-Mart's departure.

Plenty of ASSumption is needed to extrapolate from here, but 60% sounds nice. :)

There is no good reason to discount Amazon results. They are a high volume retailer and their results jive with what is being reported elsewhere. The mistake people make with Amazon is looking at the hourly info and/or the top 10 info and assuming it has relevance to what's happening in total. Instead, they should look at the top 100 discs over a past week, past 2 weeks, or past month time frame. I agree.

Thats why I like the new capabiliity at hd game db.

They now allow 30 and 45 day charting as an option besides the daily, 7 day and 14 day options.

Thats are far as they have data.

I'm told they will add 60 and 90 day data charting as their data ages over that mark.

http://www.hdgamedb.com/amazon/history.aspx

These are the new figures from HMM for the week ending Jan 28:

YTD: BD 100.00 HD 49.21
SI: BD 87.76 HD 100.00

Combining this with the weekly figures from PC Magazine (http://www.pcmag.com/article2/0,1895,2093290,00.asp), BD/HD DVD percentages are:

Day Week YTD SI
01/07 63.3/36.7
01/14 68.2/31.8
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3
While 68.8/31.2 looks kind of bad for HD DVD, I'm actually somewhat impressed at how they did in that week given how well it looks like "Saw 3" and "The Guardian" did for Blu-ray, and that "Brokeback Mountain" doesn't look like it was a real popular title for HD DVD. Given that info, I could see HD DVD coming back pretty good in the next week, except that "Flyboys" and "Open Season" probably sold pretty well for Blu-ray and it doesn't look like "Half Baked" did for the HD DVD side. The next week might be good for HD DVD given that "Hollywoodland" was day-and-date and a reasonable title.

--Darin

There is no good reason to discount Amazon results. They are a high volume retailer and their results jive with what is being reported elsewhere. The mistake people make with Amazon is looking at the hourly info and/or the top 10 info and assuming it has relevance to what's happening in total. Instead, they should look at the top 100 discs over a past week, past 2 weeks, or past month time frame.
Yeah this hourly and even 24hourly shite is driving me cookoo. :eek:

While we are speaking of Wal-Mart. Here is Wal-Mart online's list of combined (BD and HD-DVD) best sellers. I don't know how Wal-Mart arrives at their ranks, but here they are:

1. Sopranos: Season Six - Part 1 (HD-DVD)
2. Departed (BD)
3. Departed (HD-DVD)
4. Fear and Loathing In Las Vegas (HD-DVD)
5. MI Ultimate Collection (HD-DVD)
6. MI Ultimate Collection (BD)
7. Saw III (BD)
8. Hollywoodland (HD-DVD)
9. Princess Bride (BD)
10. Poseiden (HD-DVD)Was "The Princess Bride" released on Blu-ray? I thought it was delayed/cancelled and I don't see a new date for it.

--Darin

Well you know what it means to ASSume, don't you Plaz?

Actually, I first read it that way as well.

But look at the context...



It's not perfectly written, by any means, but standard writing technique would dictate that the text in parentheses is in response to the text prior and it seems very clear that this is indeed the case here. Instead of saying, "through 7/28/01," they should have said something more like, "back to..."

I don't wish to ASSume myself, but it seems much more logical that they mean to say that you can only find data for Wal-Mart from now back to 7/28/01, or from 7/28/01 through to the present.

Sketcha, Wal Mart does not report to Videoscan. If their intention is to mislead people in to thinking that they have Wal Mart data from 2001 onwards, instead of until 2001, then shame on Videoscan!

I agree.

Thats why I like the new capabiliity at hd game db.

They now allow 30 and 45 day charting as an option besides the daily, 7 day and 14 day options.

Thats are far as they have data.

I'm told they will add 60 and 90 day data charting as their data ages over that mark.

http://www.hdgamedb.com/amazon/history.aspx
Yeah, if someone was tracking the daily on the 9th, things would look pretty rosey for HD DVD. Otherwise... ;)

There is no good reason to discount Amazon results. They are a high volume retailer and their results jive with what is being reported elsewhere. The mistake people make with Amazon is looking at the hourly info and/or the top 10 info and assuming it has relevance to what's happening in total. Instead, they should look at the top 100 discs over a past week, past 2 weeks, or past month time frame.


I find the "top 100 average rank" to be the most meaningless of all the meaningless Amazon numbers. That list is filled with titles that have such a low volume, a very few sales can swing the averge wildly. "Outragous outliers" are the norm on that list.

I think the Amazon data can only be used to say very generally that it looks like more people are buying BD than HD DVD right now. I don't think one can convert specific trends or amounts out of it.

I find the "top 100 average rank" to be the most meaningless of all the meaningless Amazon numbers. That list is filled with titles that have such a low volume, a very few sales can swing the averge wildly. "Outragous outliers" are the norm on that list.

I think the Amazon data can only be used to say very generally that it looks like more people are buying BD than HD DVD right now. I don't think one can convert specific trends or amounts out of it.

You are interpreting the information incorrectly. The info is summed, making any low selling individual title irrelevant. The top 100, by definition, is more closely correlated with what is happening in total in the market.

Sketcha, Wal Mart does not report to Videoscan. If their intention is to mislead people in to thinking that they have Wal Mart data from 2001 onwards, instead of until 2001, then shame on Videoscan!
Looks like you're just catching up too, ala Kosty.

No, their wording works... just not extremely well. I doubt it was intentional, it just wasn't written very clearly.

Originally Posted by Kosty
Nope. Wal-Mart pulled out of the system on that date. Yeah, got it. Thanks Kosty. See prior posts.
Now you're the one not reading all the posts Sketcha, Wal Mart does not report to Videoscan. If their intention is to mislead people in to thinking that they have Wal Mart data from 2001 onwards, instead of until 2001, then shame on Videoscan!Tag, plazman, Your it. :) Now plaz, you're the one not reading all the posts. :D

He already got it. No need for us to beat him up after he already said he understands! :eek:

You are interpreting the information incorrectly. The info is summed, making any low selling individual title irrelevant. The top 100, by definition, is more closely correlated with what is happening in total in the market. I think the top 100 ranking will be the most relevant in the future. Along with the titles in the top 1000. Both of those will start to show titles that have some sort of sales volume.

Tag, plazman, Your it. :) Now plaz, you're the one not reading all the posts. :D

He already got it. No need for us to beat him up after he already said he understands! :eek:
http://www.cigarbid.com/auction/picpost/27805.gif

You are interpreting the information incorrectly. The info is summed, making any low selling individual title irrelevant. The top 100, by definition, is more closely correlated with what is happening in total in the market.If you are talking about the average for the top 100 titles (and not the number in the top 100 rankings for DVDs), then the lower sellers in that list actually have more influence than the higher sellers on a per sale basis. A title that is at a ranking of 8000 could sell 2 more copies and move to 6000, changing the average for the top 100 by 20 points. But a title that is at a ranking of 501 could sell 100,000 copies and only move the average for the top 100 by 5 points (moving to #1 would bring the total down by 500 places). I think that is basically skogan's point.

If we could figure out that approximate number of sales required for each place at an average time, we could probably come up with a good estimate of an equation for weighting each spot to give a better idea of overall sales between the two.

--Darin

I just thought I'd paste this from the Videoscan home page for clarity.

"Sell-through POS (consumer purchase) sales data are collected weekly from traditional channels of video distribution including mass merchants, audio/video and video specialty retailers, electronics outlets, grocery stores, drug stores, and some Internet sites. Nielsen VideoScan maintains this weekly POS data in one of the largest databases of VHS and DVD products in the country. Data is collected on more than 40,000 VHS items and 12,000 DVD items. Virtually every UPC code in sell-through release since 1993 can be found in our extensive database. Complete sales data in all retail channels are available as far back as January 1999 (however Wal*Mart data is included only through 7/28/01). Prior to January 1999, limited information is available in most channels going back to 1993."

source: http://www.videoscan.com/about.html

If Videoscan don't extrapolate for the retail market that do not report to them then does anyone know approximately what % of the retail market Videoscan accounts for? Apologies if this has been given out earlier in this thread.

Look up a few posts for more details. Videoscan claims to catch 60% of retailers (since this figure has been used for years, I'm thinking they mostly refer to brick and mortar retailers, and their web presence). We don't know what percentage of the market that is, but Wal-Mart is about 40% for DVD's, Amazon is probably at least 5-10%. So at best Videoscan is ~50% of the market in unit sales, but it is likely less than that (my guess is 30-40%).

I think the top 100 ranking will be the most relevant in the future. Along with the titles in the top 1000. Both of those will start to show titles that have some sort of sales volume.

Agreed. I don't think Skogan understands that the top 100 measures about 2/3 (assuming Amazon carries all titles) of the titles on both formats, whereas the top 10 measures only 1/16 of the total titles. Top 10 measures the hottest titles at any one moment in time, but still only 1/16 of the total...soon to be 1/20th of the total.

I must be going nuts, because I can't make sense of this.

Date, Weekly, YTD, SI
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

A 1% change in weekly sales created a 1.6% change SI?
That isn't possilbe, is it? Because the weekly number is necessarily smaller and included within the SI number, so no matter what the change is in weekly, it must be less than the change Since Inception.

What am I missing here?

It seems to me that whatever change occurs in the weekly percentage, it must necessarily result in a smaller change in the YTD, which must also be smaller than SI - because SI includes YTD, which includes weekly. So, regardless of the volume or percentage sold, the change in weekly sales can not be less than the change in YTD, which can not be less than SI.

I know I can't be right about this, but I don't know why. It's been a hard morning, someone explain please.

If you are talking about the average for the top 100 titles (and not the number in the top 100 rankings for DVDs), then the lower sellers in that list actually have more influence than the higher sellers on a per sale basis. A title that is at a ranking of 8000 could sell 2 more copies and move to 6000, changing the average for the top 100 by 20 points. But a title that is at a ranking of 501 could sell 100,000 copies and only move the average for the top 100 by 5 points (moving to #1 would bring the total down by 500 places). I think that is basically skogan's point.

If we could figure out that approximate number of sales required for each place at an average time, we could probably come up with a good estimate of an equation for weighting each spot to give a better idea of overall sales between the two.

--Darin

There is no reason why the data would be computed that way. All they need to do is sum the total sales and rank them. If what you're saying was true, the 100 title data would show more short term variation over time than the top 10 data, but that isn't the case. The 100 title data is smoother, showing less short term variation, as it should be.

Was "The Princess Bride" released on Blu-ray? I thought it was delayed/cancelled and I don't see a new date for it.

--Darin

Wal-Mart includes pre-orders, which the Departed in both formats is still a preorder at Wal-Mart. Wal-Mart has it as a pre-order set to release tomorrow (the original date). I have not heard a new date. I guess Wal-Mart will be getting a bunch of complaints this week (since it is Top 10).

I guess that is good for HD-DVD, as the more missed or delayed preorders a retailer has, the more flak that will get logged with customer service (BD Complaint, BD Complaint, ...)

I must be going nuts, because I can't make sense of this.

Date, Weekly, YTD, SI
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

A 1% change in weekly sales created a 1.6% change SI?
...
I know I can't be right about this, but I don't know why. It's been a hard morning, someone explain please.If the weekly numbers had stayed at 67.8/32.2 then the SI numbers should have still changed in BD's direction. That is, a 0% change should have resulted in a higher than 0% change to the overall sales because the runrate is higher. Much like one car closing on another car in a drag race at a constant rate.

--Darin

Agreed. I don't think Skogan understands that the top 100 measures about 2/3 (assuming Amazon carries all titles) of the titles on both formats, whereas the top 10 measures only 1/16 of the total titles. Top 10 measures the hottest titles at any one moment in time, but still only 1/16 of the total...soon to be 1/20th of the total. For now. But I am more interested in when there are more titles in each format and there sales volumes are higher.

A top 100 spot is selling (edit) 10 + plus copies a day . A top 1000 spot is a least selling a copy day (edit ), so it is in active buying.

When we have a lot of titles moving in those categories, more than the dribble we now have, those should even out to give us a large enough basic to give us real trends that will be showing format strength not just individual title releases.

The top 10 graph will show us those spikes.

I like the fact that we can basket at all of those levels on the eproduct wars site.

I like the fact we can basket the average rank of the top 10 25 50 100 titles on the hd game db site.

If all are in sync, the trend is clear. If some are different than the others, we can look at why the outlier graphs are not consistent with the others.

(edit; no way my original number sweer correct)

There is no reason why the data would be computed that way. All they need to do is sum the total sales and rank them. If what you're saying was true, the 100 title data would show more short term variation over time than the top 10 data, but that isn't the case. The 100 title data is smoother, showing less short term variation, as it should be.Which site are you talking about? The hdgamedb and eproductwars sites aren't tracking total sales, they are tracking rankings. Their averages for the top 100 are averaging all of the rankings for 100 titles. How do you think it works since Amazon doesn't release actual sales numbers?

--Darin

If the weekly numbers had stayed at 67.8/32.2 then the SI numbers should have still changed in BD's direction. That is, a 0% change should have resulted in a higher than 0% change to the overall sales because the runrate is higher. Much like one car closing on another car in a drag race at a constant rate.

--Darin
Yup. If the weekly BD number remains at least a hair higher than that of HD DVD, the YTD and SI for BD will continue to climb.

I must be going nuts, because I can't make sense of this.

Date, Weekly, YTD, SI
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

A 1% change in weekly sales created a 1.6% change SI?
That isn't possilbe, is it? Because the weekly number is necessarily smaller and included within the SI number, so no matter what the change is in weekly, it must be less than the change Since Inception.

What am I missing here?

It seems to me that whatever change occurs in the weekly percentage, it must necessarily result in a smaller change in the YTD, which must also be smaller than SI - because SI includes YTD, which includes weekly. So, regardless of the volume or percentage sold, the change in weekly sales can not be less than the change in YTD, which can not be less than SI.

I know I can't be right about this, but I don't know why. It's been a hard morning, someone explain please.

The weekly sales are growing each week (barring holiday or promotional effects). A 1% difference in week 36 is a lot more units than a 1% difference in week 10 (because there are a lot more players in homes in week 36). That is what you are missing.

Yup. If the weekly BD number remains at least a hair higher than that of HD DVD, the YTD and SI for BD will continue to climb.Okay, being technical, if Blu-ray led only 50.1 to 49.9 for a week, then the YTD numbers would most likely move HD DVD's direction (at least with the way things are now).

--Darin

Which site are you talking about? The hdgamedb and eproductwars sites aren't tracking total sales, they are tracking rankings. Their averages for the top 100 are averaging all of the rankings for 100 titles. How do you think it works since Amazon doesn't release actual sales numbers?

--Darin

They don't release unit sales numbers, but they obviously use them for computation. Look at the graphs. They don't lie.

I must be going nuts, because I can't make sense of this.

Date, Weekly, YTD, SI
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

A 1% change in weekly sales created a 1.6% change SI?
That isn't possilbe, is it? Because the weekly number is necessarily smaller and included within the SI number, so no matter what the change is in weekly, it must be less than the change Since Inception.



I know I can't be right about this, but I don't know why. It's been a hard morning, someone explain please.

It's not possible. Good catch!

The weekly and YTD are fine. The SI is off. SI for sure must increase by a smaller % than YTD.

Agreed. I don't think Skogan understands that the top 100 measures about 2/3 (assuming Amazon carries all titles) of the titles on both formats, whereas the top 10 measures only 1/16 of the total titles. Top 10 measures the hottest titles at any one moment in time, but still only 1/16 of the total...soon to be 1/20th of the total.

Read Darin's post. Maybe he explained it better than I did.

It's not possible. Good catch!

The weekly and YTD are fine. The SI is off. SI for sure must increase by a smaller % than YTD.
Good call.

They don't release unit sales numbers, but they obviously use them for computation. Look at the graphs. They don't lie.

The graphs show sales ranks. it includes pre-orders that are not shipping and hence = 0 sales, and it also includes items out of stock, which is also =0 sales.

Those graphs are not sales graphs and more like orders placed graphs. However, since 7/10 top titles are not available to ship, it is reasonable that the actual sales and orders graph are out of synch!

Also, as Darin says, the sales rank includes more than just the current sales of a title and past sales are factored in as well. So, if Batman sold 100 copies yesterday, but today has sold 0, and Clerk II sold 0 copies yesterday but sold 5 copies today. On the sales rank, Batman will still be above Clerks II at Amazon. Over time, if Clerks II continues to sell more than Batman it will gradually climb in sales ranking...

The weekly sales are growing each week (barring holiday or promotional effects). A 1% difference in week 36 is a lot more units than a 1% difference in week 10 (because there are a lot more players in homes in week 36). That is what you are missing.

Absolutely not! It does not matter even if week 36 sold 100x over week 10. But your SI percent cannot increase by more than your weekly percent. At best they can be equal....and that would only happen in week 1.

I must be going nuts, because I can't make sense of this.

Date, Weekly, YTD, SI
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

A 1% change in weekly sales created a 1.6% change SI?
That isn't possilbe, is it? Because the weekly number is necessarily smaller and included within the SI number, so no matter what the change is in weekly, it must be less than the change Since Inception.

What am I missing here?

It seems to me that whatever change occurs in the weekly percentage, it must necessarily result in a smaller change in the YTD, which must also be smaller than SI - because SI includes YTD, which includes weekly. So, regardless of the volume or percentage sold, the change in weekly sales can not be less than the change in YTD, which can not be less than SI.

I know I can't be right about this, but I don't know why. It's been a hard morning, someone explain please.

Well it's a 1% weekly change and a .6% change in YTD, so add them together and you get 1.6%.... :D ..... now do you trust Nielsen....just kidding!

I see your point. Let's look at a simple example. Let's say that on 1/21, the SI number was based off 10,000 combined discs being sold SI. So a total sales volume of 10000 hi-def discs.

Now on 1/28, we sold 10,000 discs that week, the new SI would be 57% BD and 43% HD-DVD, a total reversal of the numbers. If it was 1000 units (10% of the SI amount) on 1/28, the new SI would be 52.7% HD-DVD and 47.3% BD. So here you would have a weekly change of 1%, but an SI change of 2.2%.

Basically, the data each week is weighted by the number of units sold each week (a piece of data we don't have). Weeks with small weekly percentage changes can cause bigger SI changes, with a higher volume. Likewise, you could have a huge weekly percentage change, but if the volumes were low, you might see little or no change in SI.

Anybody elses brain about to explode trying to follow all this? :rolleyes:

It will all end up being much simplier when the volumes increase.

All these details will then even out in effect and the variations will be less volatile.

For now. But I am more interested in when there are more titles in each format and there sales volumes are higher.

A top 100 spot is selling 100-1000 plus copies a day. A top 1000 spot is a least selling a copy day to 10 a day, so it is in active buying.

When we have a lot of titles moving in those categories, more than the dribble we now have, those should even out to give us a large enough basic to give us real trends that will be showing format strength not just individual title releases.

The top 10 graph will show us those spikes.

I like the fact that we can basket at all of those levels on the eproduct wars site.

I like the fact we can basket the average rank of the top 10 25 50 100 titles on the hd game db site.

If all are in sync, the trend is clear. If some are different than the others, we can look at why the outlier graphs are not consistent with the others.

I hate to burts your bubble, but I'm 95% positive a disc ranked 100 or higher is not selling 100-1000 copies per day. Maybe per week, but not per day. I might buy your claim if you said a disc ranked 25 or higher is selling in that range. There is a huge drop in volume on Amazon once you start to leave the Top 25 ranks. By the time you get to the 50th rank, sales are starting to taper off to very low amounts.

It's not possible. Good catch!

The weekly and YTD are fine. The SI is off. SI for sure must increase by a smaller % than YTD.

Actually it is quite possible if the number of discs sold prior to January is relatively small enough.

But your SI percent cannot increase by more than your weekly percent. At best they can be equal....and that would only happen in week 1.Yes it can. Do you disagree with:
If the weekly numbers had stayed at 67.8/32.2 then the SI numbers should have still changed in BD's direction. That is, a 0% change should have resulted in a higher than 0% change to the overall sales because the runrate is higher. Much like one car closing on another car in a drag race at a constant rate.If the weekly numbers stayed at 68.8/31.2 every week from now on then the SI numbers would keep increasing until they approached 68.8/31.2 themselves. Same kind of things with the YTD numbers. If they had stayed at 66.4/33.6 then the SI numbers should have still changed in the direction toward them.

fa8362,

If you still think that actual sales numbers and not sales rankings are used for the top 100 average, first start by telling us what site you are looking at and then one of us can explain it further.

--Darin

If the weekly numbers had stayed at 67.8/32.2 then the SI numbers should have still changed in BD's direction. That is, a 0% change should have resulted in a higher than 0% change to the overall sales because the runrate is higher. Much like one car closing on another car in a drag race at a constant rate.

--Darin

I see, thanks.

Yes it can. Do you disagree with:
If the weekly numbers stayed at 68.8/31.2 every week from now on then the SI numbers would keep increasing until they approached 68.8/31.2 themselves. Same kind of things with the YTD numbers. If they had stayed at 66.4/33.6 then the SI numbers should have still changed in the direction toward them.

fa8362,

If you still think that actual sales numbers and not sales rankings are used for the top 100 average, first start by telling us what site you are looking at and then one of us can explain it further.

--Darin

Darin is right guys, just look at my simple example above. If the weekly rate stays the same, each week BD is selling more discs than HD-DVD. After a long enough period of time, BD will actually have sold more discs than HD-DVD, and the SI flips. After a long enough period of time, it would match the constant weekly rate. The volume of sales each week would determine how quickly this would occur.

Well it's a 1% weekly change and a .6% change in YTD, so add them together and you get 1.6%.... :D ..... now do you trust Nielsen....just kidding!

I see your point. Let's look at a simple example. Let's say that on 1/21, the SI number was based off 10,000 combined discs being sold SI. So a total sales volume of 10000 hi-def discs.

Now on 1/28, we sold 10,000 discs that week, the new SI would be 57% BD and 43% HD-DVD, a total reversal of the numbers. If it was 1000 units (10% of the SI amount) on 1/28, the new SI would be 52.7% HD-DVD and 47.3% BD. So here you would have a weekly change of 1%, but an SI change of 2.2%.

Basically, the data each week is weighted by the number of units sold each week (a piece of data we don't have). Weeks with small weekly percentage changes can cause bigger SI changes, with a higher volume. Likewise, you could have a huge weekly percentage change, but if the volumes were low, you might see little or no change in SI.

Not sure I follow this.

Your example is:

1. 1/21 SI total = 10,000; so based on the ratio above: HD DVD: 5490, BD 4510

2. 1/28 10,000 disks sold:

3. That means the SI total is now 20,000

4. The weekly sales is: HD DVD: 3120, BD: 6,820 (given the 68.2% v. 31.2%)

5. So the SI share for HD DVD is: 3120+5490 = 8610/20000 = 43.05

6. Similarly the SI for BD is= 56.95

So,

Unless your example was different that what I wrote out, the numbers don't add up!

I hate to burts your bubble, but I'm 95% positive a disc ranked 100 or higher is not selling 100-1000 copies per day. Maybe per week, but not per day. I might buy your claim if you said a disc ranked 25 or higher is selling in that range. There is a huge drop in volume on Amazon once you start to leave the Top 25 ranks. By the time you get to the 50th rank, sales are starting to taper off to very low amounts. I agree with that. I have to be wrong there.

Its probably closer to 100-200 copies a week for the 50-100 lower ranked titles and 200-400 copies per week for the 25-50 ranks.

The top 25 or 30 titles could be selling 50 -100 copies a day though.

Darin is right guys, just look at my simple example above. If the weekly rate stays the same, each week BD is selling more discs than HD-DVD. After a long enough period of time, BD will actually have sold more discs than HD-DVD, and the SI flips. After a long enough period of time, it would match the constant weekly rate. The volume of sales each week would determine how quickly this would occur.

The question is. Can a 1% increase in market share in a week, translate to a more than 1% market share for that month (since the weekly share is part of the monthly share)?

This is like saying can your GPA increase faster than your course grade?

If I get a B is a class: 3.0

Can my GPA go up by 3.2?

I would say it cannot, because the grade in that class is a part of the GPA.

Maybe my math is just too rusty ;)

The graphs show sales ranks. it includes pre-orders that are not shipping and hence = 0 sales, and it also includes items out of stock, which is also =0 sales.

Those graphs are not sales graphs and more like orders placed graphs. However, since 7/10 top titles are not available to ship, it is reasonable that the actual sales and orders graph are out of synch!

Also, as Darin says, the sales rank includes more than just the current sales of a title and past sales are factored in as well. So, if Batman sold 100 copies yesterday, but today has sold 0, and Clerk II sold 0 copies yesterday but sold 5 copies today. On the sales rank, Batman will still be above Clerks II at Amazon. Over time, if Clerks II continues to sell more than Batman it will gradually climb in sales ranking...

It doesn't matter whether orders are included or not. That doesn't change anything. The sales rank of the top 100 discs is still more reflective of the market total than the sales rank of the top 10 discs.

I must be going nuts, because I can't make sense of this.

Date, Weekly, YTD, SI
01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

A 1% change in weekly sales created a 1.6% change SI?
That isn't possilbe, is it? Because the weekly number is necessarily smaller and included within the SI number, so no matter what the change is in weekly, it must be less than the change Since Inception.

What am I missing here?

It seems to me that whatever change occurs in the weekly percentage, it must necessarily result in a smaller change in the YTD, which must also be smaller than SI - because SI includes YTD, which includes weekly. So, regardless of the volume or percentage sold, the change in weekly sales can not be less than the change in YTD, which can not be less than SI.

I know I can't be right about this, but I don't know why. It's been a hard morning, someone explain please.

Weekly percentage doesn't have to change at all in order for since inception figures to change.

Just imagine the following scenario:
HD DVD sales were 300,000 in 2006. BD sales were 200,000 in 2006

Week one: HD DVD sells 10,000. BD sells 20,000.
YTD is 66:33 for BD. Since inception is 58.5/41.5 for HD

Week two: HD DVD sells 10,000. BD sells 20,000 (again).
YTD is 66:33 for BD (again). Since inception is 57.1/42.9 for HD.

And so it happens that without any improvement of YTD or weekly sales, SI increased 1.4 points.

It doesn't matter whether orders are included or not. That doesn't change anything. The sales rank of the top 100 discs is still more reflective of the market total than the sales rank of the top 10 discs.

But neither is more accurate than looking at actual sales! That is what I did. I tracked the change in total stock on hand to calculate how many titles sold.

However, I do agree that the top 100 is a better reflection than the top 10, since it covers more :)

Read Darin's post. Maybe he explained it better than I did.

No, he didn't. He seems to think the folks at Amazon are incompetent enough to average the rankings of each of the top 100 discs, instead of summing the numbers that the rankings are based on and then computing a ranking for that summed total.

The question is. Can a 1% increase in market share in a week, translate to a more than 1% market share for that month (since the weekly share is part of the monthly share)?It can with the kinds of numbers we are talking about.

Just consider 3 weeks. In week one format A sells 1 copy and format B sells 3 copies. The week and SI are 25:75. The next week it flops, so the week is 75:25 and the SI is 50:50. The next week it stays the same at 75:25. That would be 0% by your calculation. Yet the SI would change to 58.3:41.7 (since the totals would be 5 to 7).

--Darin

No, he didn't. He seems to think the folks at Amazon are incompetent enough to average the rankings of each of the top 100 discs, instead of summing the numbers that the rankings are based on and then computing a ranking for that summed total.Again, please tell us what site you are talking about for this average of the top 100. Are you talking about hdgamedb.com, eproductwars.com, or Amazon itself? A link would be nice since your info is different than what hdgamedb.com and eproductwars.com do.

--Darin

Weekly percentage doesn't have to change at all in order for since inception figures to change.

Just imagine the following scenario:
HD DVD sales were 300,000 in 2006. BD sales were 200,000 in 2006

Week one: HD DVD sells 10,000. BD sells 20,000.
YTD is 66:33 for BD. Since inception is 58.5/41.5 for HD

Week two: HD DVD sells 10,000. BD sells 20,000 (again).
YTD is 66:33 for BD (again). Since inception is 57.1/42.9 for HD.

And so it happens that without any improvement of YTD or weekly sales, SI increased 1.4 points.

Perfect!

Weekly percentage doesn't have to change at all in order for since inception figures to change.

Just imagine the following scenario:
HD DVD sales were 300,000 in 2006. BD sales were 200,000 in 2006

Week one: HD DVD sells 10,000. BD sells 20,000.
YTD is 66:33 for BD. Since inception is 58.5/41.5 for HD

Week two: HD DVD sells 10,000. BD sells 20,000 (again).
YTD is 66:33 for BD (again). Since inception is 57.1/42.9 for HD.

And so it happens that without any improvement of YTD or weekly sales, SI increased 1.4 pointsBetween that scenario Grubert stated above and two or three times its magnitude, those are the volumes we are playing at currently between HD DVD and Blu-ray sales. Right?

500,000 or 1,000,000 or 1,500,000 total movie units? Depending on how many were sold and how many given away?

Were talking a possible delta between HD DVD and Blu-ray even at a 67% to 33% ratio at only being 5,000 to 10,000 to maybe 20,000 units per week?

No, he didn't. He seems to think the folks at Amazon are incompetent enough to average the rankings of each of the top 100 discs, instead of summing the numbers that the rankings are based on and then computing a ranking for that summed total.BTW: If you go to http://www.hdgamedb.com/amazon/versus.aspx and pull down the top 100 box you should be able to see that what they give is an average for the rankings (not sales volumes), just like eproductswars.com. What do you think a number of 4100 means for the top 100?

--Darin

Originally Posted by plazman
It's not possible. Good catch!

The weekly and YTD are fine. The SI is off. SI for sure must increase by a smaller % than YTD.
Actually it is quite possible if the number of discs sold prior to January is relatively small enough.
What? How?

Again, please tell us what site you are talking about for this average of the top 100. Are you talking about hdgamedb.com, eproductwars.com, or Amazon itself? A link would be nice since your info is different than what hdgamedb.com and eproductwars.com do.

--Darin

It's not an "average", it's the sales ranking of the top 100 discs. It would be computed the same way they compute the ranking for a single disc or the top 10 discs. "Average" is your terminology, since you seem to think they just averaged the rankings of the top 100 discs.

http://www.eproductwars.com/dvd/

The question is. Can a 1% increase in market share in a week, translate to a more than 1% market share for that month (since the weekly share is part of the monthly share)?


Not to beat a dead horse, but here is where we(I) messed up:

It's not a 1% increase in market share. It was a 37% increase in market share for the week. (68.8/31). The fact that it was greater or lessor than the previous week's market share is irrelevent for this purpose. 37% greater sales in that week resulted in a 1.6% change in the sales ratio since inception.

What? How?

Actually, it's possible as Grubert shows :)

I made an error....

Not to beat a dead horse, but here is where we(I) messed up:

It's not a 1% increase in market share. It was a 37% increase in market share for the week. (68.8/31). The fact that it was greater or lessor than the previous week's market share is irrelevent for this purpose. 37% greater sales in that week resulted in a 1.6% change in the sales ratio since inception.

Exactly! I realized this as I was looking at Gruberts example!!!!!!

Sorry for the back and forth....

It's not an "average", it's the sales ranking of the top 100 discs. It would be computed the same way they compute the ranking for a single disc or the top 10 discs. "Average" is your terminology, since you seem to think they just averaged the rankings of the top 100 discs.

http://www.eproductwars.com/dvd/You still don't seem to get how it works. It is an average sales ranking for those top 100. What do you think a number of 4100 means? Do you think that is a rank of 4100 compared to groupings of 100 other titles? Do you think the number for the top 100 is computed with a different methodology than the number for top 10? Add up the individual rankings for the top 10 and then divide by 10 and you get the number they give there for the graph of the top 10. Do you see that?

--Darin

Not to beat a dead horse, but here is where we(I) messed up:

It's not a 1% increase in market share. It was a 37% increase in market share for the week. (68.8/31). The fact that it was greater or lessor than the previous week's market share is irrelevent for this purpose. 37% greater sales in that week resulted in a 1.6% change in the sales ratio since inception.
I'm fully with you on that, but could someone please explain how the SI could go up by 1.6% while the YTD only goes up .6%?

Before I forget. Videoscan are not extrapolating their HD DVD and BD sales data. The numbers being reported by Videoscan are the numbers as reported by their sources with no extrapolation. I was under the assumption they extrapolated. But they do not. So that's one thing.

Amazon, Wal-Mart and most internet sites are NOT part of Videoscan data. So, when you say that the internet sales is factored into Videoscan. It is not. AFAIK, the only internet sites that provide data to Videoscan are the B&M sites that provide them data. So Bestbuy.com and circuitcity.com would be sources.

Given their low peneration in the internet sales channel, we need to factor that into their published numbers....I don't know if anyone posted this or not, but Amazon.com is very much a part of VideoScan data. Wal-Mart is not.

I think we can put this "Amazon not included" misinformation to rest.

I received a couple of documents from VideoScan with the full list. However, each document is too big to attach here. Here is the partial list of retailers (less than 3% of the full list). These are all covered under their "First Alert" service:
Discount Mass Merchants:
BI-MART
KMART (INCL. SUPERCENTERS)
ROSE'S
SHOPKO
PAMIDA
TARGET

Other Mass Merchants:
AAFES
AMAZON.COM
AOL.COM
BEST BUY
BORDERS.COM
BUY.COM
CD NOW/CDNOW.COM
CIRCUIT CITY/CIRCUITCITY.COM
COSTCO
EURPAC
GAIAM
SPRING HOUSE/GAITHERS
TOWERRECORDS.COM
TOYSRUS.COM
VIDEOUNIVERSE.COM
VIRGINMEGA.COM
WESTWOOD PROMOTION
WHEREHOUSE.COM

It's not an "average", it's the sales ranking of the top 100 discs. It would be computed the same way they compute the ranking for a single disc or the top 10 discs. "Average" is your terminology, since you seem to think they just averaged the rankings of the top 100 discs.

http://www.eproductwars.com/dvd/

No it's not.

Actually, it's possible as Grubert shows :)

I made an error....
Alright, I'm looking back. Be with you in a minute.

Actually, it's possible as Grubert shows :)

I made an error....
The weekly vs. YTD/SI issue was never a problem for me. I didn't see Grubert explain how SI could go up more than YTD. Can you show me what post this was answered in?

I'm fully with you on that, but could someone please explain how the SI could go up by 1.6% while the YTD only goes up .6%?Not sure if this helps, but if you look at:

01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

the YTD numbers were already pretty close to the weekly numbers, meaning their percentage change is likely to be small. In other words, the YTD was 66.4 for Blu-ray and the next week ended up being 68.8. The new YTD closed .6 out of a gap of 2.4, or about 1/4th of that distance. The SI was 23.7 away (68.8-45.1), and there a change of 1.6 is only 1/15th of that distance. Doesn't seem unreasonable.

One thing to consider in the above is that if the weekly numbers stay at some value (like 68.8/31.2) then both the YTD and SI will approach them over time, but the closer they are, the less they will close in general.

--Darin

For the Record,


Here are the individual ranking's of the top 10 BD's

1. 24
2. 45
3. 132
4. 231
5. 246
6. 349
7. 547
8. 644
9. 693
10. 745

Those individual rankings total to 3656.


Anyone care to take a guess on what the "Top 10" rank is at eproducts?

The weekly vs. YTD/SI issue was never a problem for me. I didn't see Grubert explain how SI could go up more than YTD. Can you show me what post this was answered in?
Nevermind. Got it. It's kind of a trip. Should've been looking at YTD as more like the weekly in its effect.

Not sure if this helps, but if you look at:

01/21 67.8/32.2 66.4/33.6 45.1/54.9
01/28 68.8/31.2 67.0/33.0 46.7/53.3

the YTD numbers were already pretty close to the weekly numbers, meaning their percentage change is likely to be small. In other words, the YTD was 66.4 for Blu-ray and the next week ended up being 68.8. The new YTD closed .6 out of a gap of 2.4, or about 1/4th of that distance. The SI was 23.7 away (68.8-45.1), and there a change of 1.6 is only 1/15th of that distance. Doesn't seem unreasonable.
It does, thanks.

One thing to consider in the above is that if the weekly numbers stay at some value (like 68.8/31.2) then both the YTD and SI will approach them over time, but the closer they are, the less they will close in general.

--Darin
This part I understood before. Must've had a brain fart on the former.

I found this post in another thread to be very interesting:

http://www.pcmag.com/article2/0,1895,2093290,00.asp

MAKING SENSE OF THE Hi Def DVD NUMBERS

John Dvorak in PC Mag

I found this post in another thread to be very interesting:

http://www.pcmag.com/article2/0,1895,2093290,00.asp

MAKING SENSE OF THE Hi Def DVD NUMBERS

John Dvorak in PC Mag

This was already posted many days ago. Go back to page 45 or so.

Originally Posted by fa8362
It's not an "average", it's the sales ranking of the top 100 discs. It would be computed the same way they compute the ranking for a single disc or the top 10 discs. "Average" is your terminology, since you seem to think they just averaged the rankings of the top 100 discs.

http://www.eproductwars.com/dvd/That is incorrect, Skogan is right, it is the average on the "Salesrank of the top xx product charts"

For the Record,
Here are the individual ranking's of the top 10 BD's

1. 24
2. 45
3. 132
4. 231
5. 246
6. 349
7. 547
8. 644
9. 693
10. 745

Those individual rankings total to 3656.


Anyone care to take a guess on what the "Top 10" rank is at eproducts?err ....365.6 :)

Which by coincidence is the average of all 10 rankings? or 3656/10. ;)

Thats the way it works for the "Salesrank of the top 25 (and 100) products" charts as well.

"Average" is your terminology, since you seem to think they just averaged the rankings of the top 100 discs.
That's exactly what they do. :)

http://www.pcmag.com/article2/0,1895,2093290,00.asp

MAKING SENSE OF THE Hi Def DVD NUMBERS By Mark Hachman not John Dvorak

My first thought was that we should forgive fa8362 because he is a new member and didn't know what many of us have learned long ago: If you disagree with Darin, you are probably wrong. Being a December 06 member, he probably wasn't aware of that axiom yet.

But I've came to the conclusion that he was manufacturing facts to support his arguments, and then acting a indignant with anyone who was skeptical. Therefore, he probably should be marked as someone who doesn't argue honestly. But out of prudence, maybe I'll give him another chance :)

The new issue of Home Media Mag is up and I haven't found a new graphic of the BD vs. HD-DVD Videoscan numbers. Anyone else given it a look yet ??

http://www.nxtbook.com/nxtbooks/questex/hom021107/

Edit: Found it.. see attached

b2b

The new issue of Home Media Mag is up and I haven't found a new graphic of the BD vs. HD-DVD Videoscan numbers. Anyone else given it a look yet ??

http://www.nxtbook.com/nxtbooks/questex/hom021107/

b2b

Page 3.

The new issue of Home Media Mag is up and I haven't found a new graphic of the BD vs. HD-DVD Videoscan numbers. Anyone else given it a look yet ??

http://www.nxtbook.com/nxtbooks/questex/hom021107/

b2b
Page 3 has the 1/28 figures.

While I expected a bigger jump due to sales of "Saw III" on Blu-Ray in the YTD section, the jump from 82.3 to 87.76 is quite a big jump!!

~Alan

While I expected a bigger jump due to sales of "Saw III" on Blu-Ray in the YTD section, the jump from 82.3 to 87.76 is quite a big jump!!

~Alan

What was it the Universal bigwig said in his "never say never" about supporting Blu-ray ??

60% sales advantage for six months or 80% for three months. I guess that means if BD can keep up the 2 to 1 sales ratio lead for another 5 months Uni might be changing their plans.. :)

b2b

.... if BD can keep up the 2 to 1 sales ratio lead for another 5 months... Kinda a big if...

hatching .....before..... chickens .....counting..... :)

Something that I haven't seen discussed is the rental equation. I would certainly assume that Netflix and BB's purchases are not included in any of the sales rankings that we are discussing, yet I think the rental market could play an important role and may not jive with the numbers discussed here.

Here's why...

The PS3.

1. Obviously the biggest selling HD optical player by far.

2. Adoption rates are, again, obviously not what they would be for any standalone.

3. I think it stands to reason that PS3 owners are of a more "fair weathered" sort, not die-hard movie fans like HD DVD owners. But what they may do a fair amount of, since it costs them no more than SD, is rent. And if this overwhelming number of PS3 (vs. standalone) owners do anything BD a fair amount, it is significant.

Isn't the rental market important to DVD? Do the studios not make loads of money in the rental market? Why would the rental market not be important to HD optical?

Now the online rental availabilities have been atrocious, but the demand is obviously there for both sides. Just wish I knew which had more demand. I haven't checked recently, can anyone say whether or not Netflix and BB have gotten their act together yet?

I'm guessing somebody knows the answer to this and I'm guessing they think it's important too.


P.S. Could the discrepancy we've seen in shipped vs. sold be due to the rentals being counted as shipped? Add to that the numbers in the hundreds that Amazon apparently keeps on hand. And how many are stored in the back of Wal-Marts, CCs and BBs?

Kinda a big if...

hatching .....before..... chickens .....counting..... :)
Yeah and don't forget all those exclusive "titles" HD DVD has coming vs. the paltry few of BD. ;)

Initial post updated.

Something that I haven't seen discussed is the rental equation. I would certainly assume that Netflix and BB's purchases are not included in any of the sales rankings that we are discussing, yet I think the rental market could play an important role and may not jive with the numbers discussed here.

Here's why...

The PS3.

1. Obviously the biggest selling HD optical player by far.

2. Adoption rates are, again, obviously not what they would be for any standalone.

3. I think it stands to reason that PS3 owners are of a more "fair weathered" sort, not die-hard movie fans like HD DVD owners. But what they may do a fair amount of, since it costs them no more than SD, is rent. And if the overwhelming number of PS3 (vs. standalone) owners do anything BD a fair amount, it is significant.

Isn't the rental market important to DVD? Do the studios not make loads of money in the rental market? Why would the rental market not be important to HD optical?

Now the online rental availabilities have been atrocious, but the demand is obviously there for both sides. Just wish I knew which had more demand. I haven't checked recently, can anyone say whether or not Netflix and BB have gotten their act together yet?

I'm guessing somebody knows the answer to this and I'm guessing they think it's important too.


P.S. Could the discrepancy we've seen in shipped vs. sold be due to the rentals being counted as shipped? Add to that the numbers in the hundreds that Amazon apparently keeps on hand. And how many are stored in the back of Wal-Marts, CCs and BBs? Fantastic point.

I've thinking of starting a Netflix/blockbuster thread.

We know that Netflix CEO stated that they went neutral because of HD DVD launch despite earlier plans for staying Blu-ray only.

In the recent 60 minutes segment on Netflix in one center they had boxes and boxes of Spiderman II DVDs , at least 10,000 copies maybe even more on the shelf in one of a IIRC dozen centers. They are among the largest purchasers of standard DVDs. They normally buy at least 100-500 copies of every release they stock.

They have a long tail strategy for titles and they will buy a lot of both HD DVD and Blu-ray. Not sure how it compares to normal sales, but with low volumes sales of titles now, they may even be a factor in their purchases for the rental market.

I mean if they buy 50,000 copies of some standard DVDs, its not out of the question that they may buy hundreds in not 1000's of copies of the HD DVD and Blu-ray releases since they know their might eventually be a demand.

http://www.cbsnews.com/stories/2006/12/01/60minutes/main2222059_page3.shtml

I mean if they buy 50,000 copies of some standard DVDs, its not out of the question that they may buy hundreds in not 1000's of copies of the HD DVD and Blu-ray releases since they know their might eventually be a demand.
Well if you talk to any members of either Netflix of BB, their Queues will indicate there is plenty of demand. It's the supply they're more concerned about.

But as I said before, I haven't checked those threads recently. Maybe they're caught up by now.

In addition to above, what I am curious about are the numbers from Netflix regarding HD-DVD and Blu-ray rental activity. That may tell us how much interest people have in HiDef in general, apart from disc sales.

New Netflix thread starte here. I am transfering some posts over to jump start it.

http://www.avsforum.com/avs-vb/showthread.php?p=9757553#post9757553

Fantastic point.
In the words of Jon Stewart...

"Seriously?"

I've thinking of starting a Netflix/blockbuster thread.
I want a mention. :D

Done. come on over. Great idea.

The new Netflix /Blockbuster Rental thread is started.
I am transferring some posts over to jump start it.


How does Netflix and Blockbuster and other Rental Outlets affect the HD format war

http://www.avsforum.com/avs-vb/showthread.php?p=9757553#post9757553

Please discuss any rental issues there.

done. come on over

New Netflix thread starte here. I am transfering some posts over to jump start it.
Saw it, thanks.

That's all I have to contribute for now. I'll certainly be keeping an eye on it. I may post nothing just to subscribe. Is there another way to do that?

Sketcha, Seriously. That's an excellent point!

I believe most people simply rent DVD, and probably, will do for the hi-def formats as well. So, what about the rental attachment rate?

:)

Saw it, thanks.

That's all I have to contribute for now. I'll certainly be keeping an eye on it. I may post nothing just to subscribe. Is there another way to do that? yep, but thats OK.

Its good to have a reserved post near the top for updates.

Please direct all Netflix Blockbuster Rental issue comments there.

How does Netflix and Blockbuster and other Rental Outlets affect the HD format war

http://www.avsforum.com/avs-vb/showthread.php?p=9757553#post9757553

How does Netflix and Blockbuster and other Rental Outlets affect the HD format war

http://www.avsforum.com/avs-vb/show...553#post9757553Just an FYI. Your link a few posts back worked, but I think this one has the ...s in the underlying link, instead of the whole:

http://www.avsforum.com/avs-vb/showthread.php?p=9757553#post9757553

Note: Even though they look the same, they are different.

--Darin

http://www.avsforum.com/avs-vb/showthread.php?p=9757553#post9757553

Thanks I corrected it

Sketcha, Seriously. That's an excellent point!

I believe most people simply rent DVD, and probably, will do for the hi-def formats as well. So, what about the rental attachment rate?

:)Especially, I think, in the beginning during the war.

Good point, Sketcha.

A top 100 spot is selling (edit) 10 + plus copies a day . A top 1000 spot is a least selling a copy day (edit ), so it is in active buying.I think I just found a useful piece of information (hoping that the in stock numbers are accurate though). I didn't realize that the graphs on the eproductwars.com site would allow you to put the cursor over a spot and it would tell the relevant info for that time, like the in stock number and the time. Looks like "Black Hawk Down" could tell us something by going here:

http://www.eproductwars.com/dvd/asinsaleshistory.cfm?asin=B000G0O5N2&db=dvd

Putting the cursor over the green part on top, at 13:30 on 2/6/07 that shows 485 in stock (although it seems to be marked as sales rank). At 13:45 on 2/12/07 it shows 373 in stock. That is about 6 days worth, with a drop of 112. It looks like they didn't get any extra stock in there based on the way the graph looks, so that would be an average of just under 19 per day. And the sales rank was mostly between 600 and 1200 during that time.

"Batman Begins" looks like it dropped about 96 in stock between 1 am on the 8th and 1 am on the 12th, for an average of about 24. It was also ranked a little higher at mostly between 350 and 850 during that time.

--Darin

I think I just found a useful piece of information (hoping that the in stock numbers are accurate though). I didn't realize that the graphs on the eproductwars.com site would allow you to put the cursor over a spot and it would tell the relevant info for that time, like the in stock number and the time. Looks like "Black Hawk Down" could tell us something by going here:

http://www.eproductwars.com/dvd/asinsaleshistory.cfm?asin=B000G0O5N2&db=dvd

Putting the cursor over the green part on top, at 13:30 on 2/6/07 that shows 485 in stock (although it seems to be marked as sales rank). At 13:45 on 2/12/07 it shows 373 in stock. That is about 6 days worth, with a drop of 112. It looks like they didn't get any extra stock in there based on the way the graph looks, so that would be an average of just under 19 per day. And the sales rank was mostly between 600 and 1200 during that time.

"Batman Begins" looks like it dropped about 96 in stock between 1 am on the 8th and 1 am on the 12th, for an average of about 24. It was also ranked a little higher at mostly between 350 and 850 during that time.

--Darin I see you have to click on the inventory line to activate it the tool. You can click on the lower sales rank line to get the sales rank for that time.

That's freaky geeky cool. :cool:

Anybody tell Plazman Skogan Stetcha Kdragon and Grubert about it ?? :D

Wonder what happened to the inventory stock during the last $5 off Amazon Prime surgette for HD DVD last weekend.

I think I just found a useful piece of information (hoping that the in stock numbers are accurate though). I didn't realize that the graphs on the eproductwars.com site would allow you to put the cursor over a spot and it would tell the relevant info for that time, like the in stock number and the time. Looks like "Black Hawk Down" could tell us something by going here:

http://www.eproductwars.com/dvd/asinsaleshistory.cfm?asin=B000G0O5N2&db=dvd

Putting the cursor over the green part on top, at 13:30 on 2/6/07 that shows 485 in stock (although it seems to be marked as sales rank). At 13:45 on 2/12/07 it shows 373 in stock. That is about 6 days worth, with a drop of 112. It looks like they didn't get any extra stock in there based on the way the graph looks, so that would be an average of just under 19 per day. And the sales rank was mostly between 600 and 1200 during that time.

"Batman Begins" looks like it dropped about 96 in stock between 1 am on the 8th and 1 am on the 12th, for an average of about 24. It was also ranked a little higher at mostly between 350 and 850 during that time.

--Darin

You're going to make plazman's head explode after all the work he did to try to show HD DVD was inexplicably selling better than BD on Amazon.

You two have too much time on your hands!

You two have too much time on your hands! We just don't sleep much when we're on a roll :eek: and we still do our day jobs and watch HD movies and have a life too, this is just a obsessive hobby..... ;)

Its helps that were both pretty good at multiple tabbed browser window switching and we type pretty fast too. :p

....plus, I just love saying....

That's freaky geeky cool :cool: :

Not sure I follow this.

Your example is:

1. 1/21 SI total = 10,000; so based on the ratio above: HD DVD: 5490, BD 4510

2. 1/28 10,000 disks sold:

3. That means the SI total is now 20,000

4. The weekly sales is: HD DVD: 3120, BD: 6,820 (given the 68.2% v. 31.2%)

5. So the SI share for HD DVD is: 3120+5490 = 8610/20000 = 43.05

6. Similarly the SI for BD is= 56.95

So,

Unless your example was different that what I wrote out, the numbers don't add up!

You got the same thing I did and the numbers add up perfectly. What don't you think is right? In this example, you have a 1% weekly change, but because the sales volume for that week is so large compared to SI, you get a 11-12% increase SI for BD. So 1% weekly and 12% SI. The SI gain does not have to be less than the weekly gain, if the volume is large enough for the week. It's just an example to point out that the SI change DOES NOT need to be smaller than the weekly or YTD change.

Wonder what happened to the inventory stock during the last $5 off Amazon Prime surgette for HD DVD last weekend.I'm not sure. Looks like that was a little after 11 pm Central time on the 8th. Looks like "The Mummy Returns" might have changed stock by maybe 10 or so in there, but not sure if that was due to the deal.

BTW: You might want to sit down while I break the news on this one. Doing the same thing with the HD-A2 makes it look like they've been selling about 12-15 a day. Not as impressive as many of us would have thought with it being ranked so high among DVD players. Seems like it is going to be tough to get to 100,000 players a month if they can't do more than that on Amazon. However, I thought that Amazon was mostly selling the HD-A2 through others. From those in stock charts it looks like Amazon started carrying the HD-A2 itself starting on January 26th, but still sells the HD-XA2 through another place (thus showing zero stock itself).

The XBOX360 add-on looks like it has been selling a little over 20 per day on Amazon, on average. Looks like the PS3 remote was selling a little over 60 per day on there, but yesterday stock dropped over 260.

--Darin

Fantastic point.

I've thinking of starting a Netflix/blockbuster thread.

We know that Netflix CEO stated that they went neutral because of HD DVD launch despite earlier plans for staying Blu-ray only.


Really. I didn't know that. In fact, I knew the opposite. As a matter of fact, here is a quote from the CEO of Netflix.

"Protracted competition will hurt the adoption of high-definition DVD," Hastings reportedly said. "If all studios were to embrace both formats agnostically, consumers would be more comfortable making a format decision based on hardware pricing and features."

My impression is that he has always wanted a single format, or equaly support for multiple formats. I've never read where he stated one format over the other.

I would say Darin's method is a little more labor intensive. Just tracking the overall stock trend will do it across all title :)

But you gotta do it on a Sunday, when no inventory is added for any title....

I don't know if anyone posted this or not, but Amazon.com is very much a part of VideoScan data. Wal-Mart is not.

I think we can put this "Amazon not included" misinformation to rest.

I received a couple of documents from VideoScan with the full list. However, each document is too big to attach here. Here is the partial list of retailers (less than 3% of the full list). These are all covered under their "First Alert" service:
Discount Mass Merchants:
BI-MART
KMART (INCL. SUPERCENTERS)
ROSE'S
SHOPKO
PAMIDA
TARGET

Other Mass Merchants:
AAFES
AMAZON.COM
AOL.COM
BEST BUY
BORDERS.COM
BUY.COM
CD NOW/CDNOW.COM
CIRCUIT CITY/CIRCUITCITY.COM
COSTCO
EURPAC
GAIAM
SPRING HOUSE/GAITHERS
TOWERRECORDS.COM
TOYSRUS.COM
VIDEOUNIVERSE.COM
VIRGINMEGA.COM
WESTWOOD PROMOTION
WHEREHOUSE.COM

I'm afraid I can't believe what you are saying. Nielsen has a confidentiality agreement with their partners that states they will not publish a list of participating members. If you actually have a list it means one of two things. You either work for Nielsen and just published confidential information on the web, or, Nielsen has actually published a list and violated the confidentiality clause of every partner contract.

....plus, I just love saying....


That's freaky geeky cool
I doubt there are many of us that can claim non-geekitude, but, my dear friend, you do understand that the above statement firmly plants you deep within the the geek borders. Possibly near the capital. :D

Now truth be told, I thought that was some really nice detective work on the part of darinp2. Definitely some strong geek auras emanating from his direction.

It does seem like there is some potential here for anyone with plenty of time on their hands. Is it possible that these figures can be tracked, combined with our market share data and later extrapolated to give us some fairly reliable sales figures?

A couple of weeks ago, I tracked the A-2 and it sold around 250 units. At that time it was ranked in the 100s. The Sammy sold 20 and was in the 2000s....

Also, if you look at the entire 24 hour period, BD outsold HD DVD by a small margin. BD overtook HD DVD between 9:30 PM and 8:30AM!

Guess, HD DVD sold better during the day, and BD at night. It was around 1,200 disks sold in total.

I'm afraid I can't believe what you are saying. Nielsen has a confidentiality agreement with their partners that states they will not publish a list of participating members. If you actually have a list it means one of two things. You either work for Nielsen and just published confidential information on the web, or, Nielsen has actually published a list and violated the confidentiality clause of every partner contract.

I can't really see where a list of which companies provide their sales data would need to be confidential. I could see where the specific data from each member should be kept confidential (wouldn't want a competitor knowing your sales data).

b2b

I hope no one ordered a PS3 from them :)

http://www.eproductwars.com/dvd/asinsaleshistory.cfm?asin=B0009VXAM0&db=dvd

But as interesting, look at that one shipment, and it's effect on the product ranking.

I hope no one ordered a PS3 from them :)

http://www.eproductwars.com/dvd/asinsaleshistory.cfm?asin=B0009VXAM0&db=dvd

But as interesting, look at that one shipment, and it's effect on the product ranking. So the sales rank for the PS3 went to no1 when they got them in stock, and then that one sales day lasted for weeks in effect for its historical component.

42 days so far with a 142 drop (rise) in its rating, with no additional sales.

But what category was it competing in, the DVD category may be different?

The scale o n the bottom is reversed right?

Hmm, I'm not so sure that no inventory is added at Amazon on Sundays.

If you look at Blu-Ray 'Superman Returns' stock levels on eproductwars, then you'll see that on Sunday at about 16:30 there was a jump of 30 units up, which presumably was a stock replenishment.

(If they ship out, why wouldn't they also have some of their trucks delivering?)

I can't really see where a list of which companies provide their sales data would need to be confidential. I could see where the specific data from each member should be kept confidential (wouldn't want a competitor knowing your sales data).

b2b

When I contacted Nielsen to request a list, I was told that by contract, a list of thier partners was confidential and could not be released.

"Due to confidentiality agreements with our information providers, we cannot supply a list of reporting stores or web sites."

I don't believe they ship out or ship in on Sunday - simply when a sale is made, it takes it off the available stock.

I didn't notice the addition to stock.

I think the Amazon data can only be used to say very generally that it looks like more people are buying BD than HD DVD right now. I don't think one can convert specific trends or amounts out of it.

Skogan Agree, I would also add, that it never could and that all it can be done is does it look like it matches or contradicts the others.

It's not an "average", it's the sales ranking of the top 100 discs. It would be computed the same way they compute the ranking for a single disc or the top 10 discs. "Average" is your terminology, since you seem to think they just averaged the rankings of the top 100 discs.

AMazone does the ranking, the other sites dothe top 100. They don't have access to the real numbers only the ranks to calculate them

A couple of weeks ago, I tracked the A-2 and it sold around 250 units. At that time it was ranked in the 100s. The Sammy sold 20 and was in the 2000s....How long of a period was the 250 units over?
I don't believe they ship out or ship in on Sunday - simply when a sale is made, it takes it off the available stock.

I didn't notice the addition to stock.You can see the bump from 54 to 83 in stock for SR this Sunday that Nescio mentioned here:

http://www.eproductwars.com/dvd/asinsaleshistory.cfm?asin=B000JVT09C&db=dvd

Even if they didn't restock, only tracking the overall stock skews things away from the side which has items that are out of stock, but which are still being ordered.

--Darin

But neither is more accurate than looking at actual sales! That is what I did. I tracked the change in total stock on hand to calculate how many titles sold.Plazman, you've done some interesting analysis but your premise is flawed. Sales of pre-release discs won't show as a drop in remaining stock, so they'll be invisible based on your method. Given that Blu-ray has many more titles available for pre-order than HD DVD you would expect there are many more pre-orders being booked for Blu-ray, yet these won't show up in your calculations.

- Talk

You can see the bump from 54 to 83 in stock for SR this Sunday that Nescio mentioned here:

Not sure what happened here .. but one way the stock can go up without restocking is if there are cancellations.

Plazman, you've done some interesting analysis but your premise is flawed. Sales of pre-release discs won't show as a drop in remaining stock, so they'll be invisible based on your method. Given that Blu-ray has many more titles available for pre-order than HD DVD you would expect there are many more pre-orders being booked for Blu-ray, yet these won't show up in your calculations.

- Talk

So what? They are available for pre-order so they can be cancelled a few months later. Should that sale really count? Or as Plaz has stated before. You pre-order and then cancel later because you find a better priced pre-order. Again, should that count?

I tracked the A2 for 14 days, but for that long replenishment could have taken place.

As for Superman, the stock increase may have to do with cancelled orders. I don't know. But when, Amazon has an order for an item in stock, the stock goes down, when they don't have stock it does not change. So, if you are ordering out of stock items, no changes to stock takes place. It's like placing a pre-order.

However, you can order an instock item and cancel before it ships. That case stock goes down and then goes back up. But I have no way of knowing if this indeed happened on Sunday with Superman. Also, I am basing my information on no physical shipments in or out from an employee at Amazon who works in their IT group. He could be wrong, but typically if someone who workd for a company tells me something about their company I believe it. However, I don't work at Amazon. so I don't know for sure.

I have no problems being set right. For the education part is more important than the being right part :)

How many cancellations could there be? How often do you order something and change your mind and cancel?

My guess is the cancellations are, pretty much an insignificant number.

How many cancellations could there be? How often do you order something and change your mind and cancel?

My guess is the cancellations are, pretty much an insignificant number.


It depends, with Amazon it happens a lot because people order two items together to save on postage, and from the info provided they calculate they will ship together, but then one gets delayed and the whole order held up. I speak from experience on this.

Also some items go on sale (not common in HD) and you cancel to get the better price there or somewhere else. It's the luxury super-slow super-saver shipping allows.

It depends, with Amazon it happens a lot because people order two items together to save on postage, and from the info provided they calculate they will ship together, but then one gets delayed and the whole order held up. I speak from experience on this.

Also some items go on sale (not common in HD) and you cancel to get the better price there or somewhere else. It's the luxury super-slow super-saver shipping allows.
Hmmm. Interesting. Roughly what percentage of the time have either of these scenarios occurred with you?

Even if you choose the free shipping, Amazon will just ship the items separately is there is a delay.

I'm starting to see a trend... :D

Although I almost bought into HD-DVD early last month (but waited to see what CES brought), I think the data and 2007+ studio support pretty clearly shows BR is going to be around for the long haul. I'll be looking for a BR player this weekend. :)

Hmmm. Interesting. Roughly what percentage of the time have either of these scenarios occurred with you?

And another situation. Just this week I had two separate orders with two movies each. They said they were going to ship middle to end of March. I looked at the listing for the movies and they said all four movies were in stock (when I ordered, end of January, at least one was pre-order or out of stock in each order). I cancelled both of the orders and re-placed them last Friday. They bothed shipped that afternoon. I have done this several times now, and always keep an eye on the expected shipping dates.

As pointed out, the graphs in thedvdwars.com show a history of stock levels at Amazon. While tracking titles in the aggregate may lead to invalid results, tracking individual titles over a period when it looks like there was no stock replenishment might give some idea of the relationship between sales rank and sales numbers.

With this in mind I looked at the stock level history for the top 15 blu-ray and top 15 HD titles that are now in stock. The attached pdf has the data for the 30 titles. It shows the period I tracked (most recent period without a stock replenishment, i.e. no stock increase showing during the period), the average daily sell rate, and the current (as of a few hours ago) sales rank. On average there are 5-6 days of history for each title. I did not include "The Departed" even though it is now in stock on the HD side, this is both because less than a full day has elapsed since it went in stock (presumably fewer people buy a title at 2 AM that at noon) and because the inclusion of a single title probably selling more than 250 copies over the next 24 hours would significantly distort the results.

Conclusions drawn from data in the pdf

1 For in stock titles blue-ray outsold HD by 56% to 44%, based on total sales in tyhe data I looked at of about 2000.

2. Average blu-ray rank was 1323 versus 2217 for HD, a much bigger difference than the difference in sales.

3. Sales are about 25 a day for a rank of 600, and fall off to about five at rank 3000.

4. Some titles seem to have very high sales relative to their current rank, for example Rambo, and Superman returns (HD version).

5. For stocked titles Amazon keeps about 18 days of inventory on average. This number is distorted by Saw 3 which has a very high inventory level relative to sales.

When looking at the data I did notice two anomalies, both on the HD side. For Batman Begins, and for the Mummy, there were large increases in stock levels (200 copies) followed by a quick drop to almost the same starting level within a few hours or days. Unless someone is buying 200 copies of something the most likely cause would be input error followed by correction. One needs to be careful when looking at the raw data.

I will be looking for relationships between sales rank, sales, and stock levels. The amazon sales rank clearly depends on more than just a six day average, it is highly weighted to recent (last day or so) sales. Over the next week I hope to add a few more titles as they come in stock, especially those in the under 100 region. I might then be able to fit a curve relating sales to rank to get a quick idea of total sales without having to tediously check the in stock levels.

So, again, what percentage of the time does this kind of stuff occur?

So what? They are available for pre-order so they can be cancelled a few months later. Should that sale really count? Or as Plaz has stated before. You pre-order and then cancel later because you find a better priced pre-order. Again, should that count?

It wouldn't count.

Measuring changes in stock numbers doesn't tell you a thing about how many went out the door from the pre-orders.

The Departed for Blu-ray is sold out on Amazon.com, despite spending weeks in the top 100. So, what can we conclude? Zero stock -> No change in stock -> No sales -> It isn't popular for BD?

Gary

I will be looking for relationships between sales rank, sales, and stock levels. The amazon sales rank clearly depends on more than just a six day average, it is highly weighted to recent (last day or so) sales. Over the next week I hope to add a few more titles as they come in stock, especially those in the under 100 region. I might then be able to fit a curve relating sales to rank to get a quick idea of total sales without having to tediously check the in stock levels. Great work.

Both eproduct wars/dvdwars and http://www.hdgamedb.com/amazon/history.aspx
now have some individal title statistics and inventory information for each title. Both sides complement each other in their data sets.

The stock level analysis is definitely interesting, both on the aggregate level and on the movie level. It gives a much better idea of what is going on than rankings. But, as any source, it has weaknesses. That's all I wanted to suggest with the Superman observation.

(Less than half the titles in the top 10s having any stock levels is obviously another issue. And hardware sales are IMHO more telling since they're less dependent on the quality of a particular title and thus reflect better how well a format is doing overall in the public's eye. So the drop of the XBox add-on and the A2 in Amazon's rankings (despite a short Superbowl bump) may be a bad omen for HD DVD. However, Amazon is a less important source of hardware than of software, so these data may not mean much.)

I would also woory about stock level analysis in this way.

The on hand stock listing is meant first as a consumer sales tool, to give the impression of availablilty time to reduce a consumers anxiety. But its not meant or toughted as an sales repoprt. Amazon specifically does not want to give that data out. it only wants to give ratios or rankings, not sale quantity data.

So there is no incentive for Amazon to keep the arrival inventory data posted or correctly showw all inventory reductions. They may even summarize stuff in their input as long as teh end of the cycle number on han dis accurate for a consumer buying decision purpose.

What I'm saying is there may be hidden sales volumes that aer not tracked on the inventory reports even if the on hand data is accurate.

Also I do mot know if any its are drop shipped form seperate inventory systems. Amazon does this with books . If a book is out of stock and is at a publishers wherehouse, it can be shiped direct to the consumer without going through the Amazon inventory system as ever being in stock. Not sure if this is happening with HD DVD of Blu-ray disks yet, but I know it happens with some DVDs and books.

So, again, what percentage of the time does this kind of stuff occur?

I've ordered from amazon only to cancel later after I buy something else at Best Buy. Waiting for a movie to come in can be kind of hard :)

Is there a reason no one seems to be able to offer a rough percentage of personal cancellations?

I
What I'm saying is there may be hidden sales volumes that aer not tracked on the inventory reports even if the on hand data is accurate.


That makes a lot of sense. If this data was totally accurate, you could collect actual Amazon absolute sales numbers.

Why would they want their competitors to be able to have access to that?

Gary

I've ordered from amazon only to cancel later after I buy something else at Best Buy. Waiting for a movie to come in can be kind of hard :) A fair amount of peole might do that for out of stock hit movies.

Another thing, didn't someone talk about the good old " Best Buy Amazon Switcheroo." Thats when you ordered the cheaper movie from Amazon, bought the full price movie at Best Buy and kept the receipt, and then returned the Amazon copy when it arrived to Best Buy (with the Best Buy receipt) and got the full price of the Best Buy copy back.

So you ended up watching the Best Buy copy immediately but paying the Amazon price, because you got the Best Buy price refunded as you returned a new item back to them.

Not that I condone that , but thats a factor they deal with their liberal return policies.

I wonder how often normal people do that , I don't think its enough to affect sales much.

Is there a reason no one seems to be able to offer a rough percentage of personal cancellations?

Yes. No one tracks how often they order, compared to how often they cancel, so they can't give a percentage.

I've made two Amazon cancellations since last November. I only place three orders. So 66% of the time I cancel my order from Amazon.

See how useless your question becomes?

I can't give you an answer overall, because I have no idea how many orders vs. how many cancellations I've made with Amazon in my lifetime. That would be a better number, but I can only give you what I remember.

Yes. No one tracks how often they order, compared to how often they cancel, so they can't give a percentage.

I've made two Amazon cancellations since last November. I only place three orders. So 66% of the time I cancel my order from Amazon.

See how useless your question becomes?

I can't give you an answer overall, because I have no idea how many orders vs. how many cancellations I've made with Amazon in my lifetime. That would be a better number, but I can only give you what I remember.
Well thanks for the condescension. You're a class act.

What are you, [a certain 80s president that I happen to admire, but had memory issues] reincarnated? All I ask for is an estimate. You gave me a good estimate since November, if you really have no idea how many orders you have made vs. how many you've cancelled, then that's all you had to say.


EDITED: Forum rules; Politics. Kinda' takes some of the spark out, but such is life.

As for Superman, the stock increase may have to do with cancelled orders.I doubt that that number of cancellations is likely to happen in such a short time. I also believe that cancellations count against the sales ranking just like orders count for the sales ranking. The sales ranking for SR did not change much over the approximately 2 hour period where we have the snapshots on both ends that show the stock going up by 29 (more than the average it was selling per day).

--Darin

I see you have to click on the inventory line to activate it the tool. You can click on the lower sales rank line to get the sales rank for that time.

That's freaky geeky cool. :cool:

Anybody tell Plazman Skogan Stetcha Kdragon and Grubert about it ?? :DGeeky indeed! Thanks Darin!

Well thanks for the condescension. You're a class act.

What are you, Ronald Reagan reincarnated? All I ask for is an estimate. You gave me a good estimate since November, if you really have no idea how many orders you have made vs. how many you've cancelled, then that's all you had to say.

Your the one who asked the same question four times in two hours (three times in one hour). If you want to aggrevate with the same question asked repeatedly over and over, expect condescension.

Take it to PM guys. Don't let the thread get ugly. I learning from both of you. :)

I'm afraid I can't believe what you are saying.I have no problem with that! :)

Nielsen has a confidentiality agreement with their partners that states they will not publish a list of participating members. If you actually have a list it means one of two things. You either work for Nielsen and just published confidential information on the web, or, Nielsen has actually published a list and violated the confidentiality clause of every partner contract.

1. I don't work for ACNeilsen / VideoScan. (contrary to my interest in this thread, I am not good with numbers. :o )
2. I don't know what contracts they have with their partners. Please provide a link, or quote such contract so that I can believe you! :)

To help you believe me, here is my story: I sent an email to David.S.Anderson@nielsen.com (that I found on their website). David forwarded it to someone else within the organization (I won't publish that email id, of course). Finally, I received an email back with two documents. One document had full list of their retail partners (over 900 in total). The other document was an overview of their data collection methods and services. In both these documents, Amazon.com was mentioned as an example of a channel (they have combined reports as well as reports of individual channels).

The only reason I could not post those documents was because the server refused to upload them due to size (in hindsight, maybe it was for better!). If anyone is interested in these documents, please let me know how I can make them available.

Both these documents are (apparently) designed to give prospective customers a glimpse into the methods and resources used by Neilsen/VideoScan. It is quite impressive, and I can understand why people rely on their data.

I don't think this violates any confidentiality agreements. I remind you again to link/quote such a contract that you mentioned.

The purpose of my post was to put the misinformation regarding Amazon.com to rest after what I found out from Neilsen. If you have information to the contrary, please do post it here.

You can email it to me at KostyAVS@gmail.com if you want.

We just don't sleep much when we're on a roll :eek: and we still do our day jobs and watch HD movies and have a life too, this is just a obsessive hobby..... ;)

Its helps that were both pretty good at multiple tabbed browser window switching and we type pretty fast too. :p

Kosty & Darin,

Just joshing ya and trying to be obnoxious. Keep up the good work gentlemen.

To help you believe me, here is my story: I sent an email to David.S.Anderson@nielsen.com (that I found on their website). David forwarded it to someone else within the organization (I won't publish that email id, of course). Finally, I received an email back with two documents. One document had full list of their retail partners (over 900 in total). The other document was an overview of their data collection methods and services. In both these documents, Amazon.com was mentioned as an example of a channel (they have combined reports as well as reports of individual channels).
.
Well done sir.

Kosty,

While trying to mail those documents to you, I ran into difficulties with the mail server!!! They are only about 5MB each, word documents. No wonder MS went XML in Office 2007.

Anyway, I decided to convert them into PDF (I wanted to keep them in the form I received, but I have no choice right now). Instead of sending them to you, I am posting them here. I will try again later to send the original word documents to you.

Here they are:

Thanks, Skogan.

Amazon is sold out on the XA2 again, I wonder if that's good or bad for HD-DVD?

Kosty,

While trying to mail those documents to you, I ran into difficulties with the mail server!!! They are only about 5MB each, word documents. No wonder MS went XML in Office 2007.

Anyway, I decided to convert them into PDF (I wanted to keep them in the form I received, but I have no choice right now). Instead of sending them to you, I am posting them here. I will try again later to send the original word documents to you.

Here they are:

What is the difference between census and sample?

Amazon is sold out on the XA2 again, I wonder if that's good or bad for HD-DVD?It shows in stock for me right now, but they sell that one through a different place. The HD-A2 shows 374 in stock as of last night. From looking at the stock numbers it looks to me like they were selling about 12-15 per day on average, but the last couple of days have been selling about 4 per day (with the ranking dropping). The XA2 is ranked a little lower, so probably less than that.

BTW: It is possible that Amazon could get new stock just about every single day on the HD-A2 that just happens to make it look like they only sold 4-20 each day, but I doubt that. The line for in stock shows quite a bit of consistency and I doubt they get new stock just about every day that just happens to be slightly less than what they sold that day. With the ranking moving up today, I bet that the in stock numbers tonight point to them selling between 6 and 10 of the HD-A2s today.

--Darin

I have no problem with that! :)



1. I don't work for ACNeilsen / VideoScan. (contrary to my interest in this thread, I am not good with numbers. :o )
2. I don't know what contracts they have with their partners. Please provide a link, or quote such contract so that I can believe you! :)

To help you believe me, here is my story: I sent an email to David.S.Anderson@nielsen.com (that I found on their website). David forwarded it to someone else within the organization (I won't publish that email id, of course). Finally, I received an email back with two documents. One document had full list of their retail partners (over 900 in total). The other document was an overview of their data collection methods and services. In both these documents, Amazon.com was mentioned as an example of a channel (they have combined reports as well as reports of individual channels).

The only reason I could not post those documents was because the server refused to upload them due to size (in hindsight, maybe it was for better!). If anyone is interested in these documents, please let me know how I can make them available.

Both these documents are (apparently) designed to give prospective customers a glimpse into the methods and resources used by Neilsen/VideoScan. It is quite impressive, and I can understand why people rely on their data.

I don't think this violates any confidentiality agreements. I remind you again to link/quote such a contract that you mentioned.

The purpose of my post was to put the misinformation regarding Amazon.com to rest after what I found out from Neilsen. If you have information to the contrary, please do post it here.

Thanks for the correction! The information I had about Videoscan not getting data from Amazon appears to be incorrect. I don't believe, Nielson can state that they get data from a source that they do not....

Did they describe the difference between how census v. sample works?

Also, given their extensive coverage, I am wondering why 2 independent sources have now told me that they've seen Nielson data and it shows around 1M combined disk sales for both formats. Of course one was Piadgeek. :confused:

It shows in stock for me right now, but they sell that one through a different place. The HD-A2 shows 374 in stock as of last night. From looking at the stock numbers it looks to me like they were selling about 12-15 per day on average, but the last couple of days have been selling about 4 per day. The XA2 is ranked a little lower, so probably less than that.

BTW: It is possible that Amazon could get new stock just about every single day on the HD-A2 that just happens to make it look like they only sold 4-20 each day, but I doubt that. The line for in stock shows quite a bit of consistency and I doubt they get new stock just about every day that just happens to be slightly less than what they sold that day.

--Darin

Darin, I have to agree with you. However, if the A-2 is selling 4 a day and is ranked in the 300s. I wonder what the BD players are doing on this site!

Darin, I have to agree with you. However, if the A-2 is selling 4 a day and is ranked in the 300s. I wonder what the BD players are doing on this site!You'll laugh at this one. The BDP-S1 looks like they might have sold one or maybe zero in the last 2 weeks or so. With stock going up by 1 I'm guessing a return. They may end up getting stuck with more Samsungs than they want. They have 191 in stock and looks like they might be selling 2-3 per day. But the Sony is worse. They say they have 337 in stock and it doesn't look like they are selling at all. I think the PS3 is probably the dominant factor for Blu-ray software sales.

--Darin

You'll laugh at this one. The BDP-S1 looks like they might have sold one or maybe zero in the last 2 weeks or so. With stock going up by 1 I'm guessing a return. They may end up getting stuck with more Samsungs than they want. They have 191 in stock and looks like they might be selling 2-3 per day. But the Sony is worse. They say they have 337 in stock and it doesn't look like they are selling at all. I think the PS3 is probably the dominant factor for Blu-ray software sales.

--Darin

The standlones sell much better at B&M stores. In fact my friend's best buy (he's a manager) received 11 Sony Blu-ray players in Jan. 4 Sold in Jan, 6 sold in Feb. 1 left. Here's what happens:

"Ok, I want this Sony Tv, what's the best deal you can give me?"

"How about this Sir, see this $900 High definition player from Sony? - we'll give it to you for half <proceeds to give speech on 1080p and dvds being 480p>!"

"Sounds good, we have a deal"

<salesman takes $400 off the price of the tv and adds the Blu-ray player>

The same goes for Toshiba HD DVD players. But it's much harder to match a Toshiba tv with the player since Best Buy sells Sonys to Toshiba at a 20:1 ratio almost.

Kosty,

While trying to mail those documents to you, I ran into difficulties with the mail server!!! They are only about 5MB each, word documents. No wonder MS went XML in Office 2007.

Anyway, I decided to convert them into PDF (I wanted to keep them in the form I received, but I have no choice right now). Instead of sending them to you, I am posting them here. I will try again later to send the original word documents to you.

Here they are: I got the original word documents you sent through Gmail.

Yep those are their marketing documents.

No breach of confidentiality. Very good job sir. :)

What is the difference between census and sample?

census is raw data 100% off their Point of Sale system, usually transferred electronically via EDI.

sample is just that , their POS system doesn't track or can't transfer real time data so the information is gathered from a random sample to accurately represent their sales volume.

Your the one who asked the same question four times in two hours (three times in one hour). If you want to aggrevate with the same question asked repeatedly over and over, expect condescension.
I kept asking because I kept getting answers that were, at best incomplete and more like dodges.

Many here are saying that cancellations are seriously affecting the numbers, yet no one seems to have any idea how many cancellations occur. Maybe you do. I'm all ears.

The standlones sell much better at B&M stores. In fact my friend's best buy (he's a manager) received 11 Sony Blu-ray players in Jan.


Since Amazon has players for an average of 40% off compared to Best Buys 20-25%, I wonder why people would throw their $ away?

Then, all my DVD players are region free and have PAL converters, so I'm used to going on line.

I have no problem with that! :)



1. I don't work for ACNeilsen / VideoScan. (contrary to my interest in this thread, I am not good with numbers. :o )
2. I don't know what contracts they have with their partners. Please provide a link, or quote such contract so that I can believe you! :)

To help you believe me, here is my story: I sent an email to David.S.Anderson@nielsen.com (that I found on their website). David forwarded it to someone else within the organization (I won't publish that email id, of course). Finally, I received an email back with two documents. One document had full list of their retail partners (over 900 in total). The other document was an overview of their data collection methods and services. In both these documents, Amazon.com was mentioned as an example of a channel (they have combined reports as well as reports of individual channels).

The only reason I could not post those documents was because the server refused to upload them due to size (in hindsight, maybe it was for better!). If anyone is interested in these documents, please let me know how I can make them available.

Both these documents are (apparently) designed to give prospective customers a glimpse into the methods and resources used by Neilsen/VideoScan. It is quite impressive, and I can understand why people rely on their data.

I don't think this violates any confidentiality agreements. I remind you again to link/quote such a contract that you mentioned.

The purpose of my post was to put the misinformation regarding Amazon.com to rest after what I found out from Neilsen. If you have information to the contrary, please do post it here.

You do realize that not everything exists in cyberspace and can be linked. I already put a direct quote of what I was told in a previous post. Look back through if you missed it. Maybe someone wasn't doing their job and did send you a list, but I still find it hard to believe they would violate their confidentality agreement to a stranger off the street. I'll believe it when I see it.

However, Bookscan does mention Amazon.com as a partner, so it isn't a stretch that they would contribute to Videoscan too.

I'm afraid I can't believe what you are saying. Nielsen has a confidentiality agreement with their partners that states they will not publish a list of participating members.
wnoris. I guess what you are saying is that no one can have any idea of the members because VS won't tell anyone. If that is true, how did you know and state as a fact many pages back that when they say on line they don't really have any but put that they track online sales because of stuff like bestbuy.com? And then you ask if others are about FUD

Not sure what happened here .. but one way the stock can go up without restocking is if there are cancellations.

Nataraj, in theory I would agree with you, but the issue is that it jumped by 30 (one title) in an a 1/2 hour from 16:15-16:45. what is the coincidence that 30+ people will cancel a the same movie in 1/2 an hour?

Nataraj, in theory I would agree with you, but the issue is that it jumped by 30 (one title) in an a 1/2 hour from 16:15-16:45. what is the coincidence that 30+ people will cancel a the same movie in 1/2 an hour?
Actually none.

However, what IS the most obvious explanation is that they just received 30 MORE discs of that title into stock... not 30 cancellations...

Plazman, you've done some interesting analysis but your premise is flawed. Sales of pre-release discs won't show as a drop in remaining stock, so they'll be invisible based on your method. Given that Blu-ray has many more titles available for pre-order than HD DVD you would expect there are many more pre-orders being booked for Blu-ray, yet these won't show up in your calculations.

- Talk
Unless Amazon is getting new stock slower than new orders...

When looking at the data I did notice two anomalies, both on the HD side. For Batman Begins, and for the Mummy, there were large increases in stock levels (200 copies) followed by a quick drop to almost the same starting level within a few hours or days. Unless someone is buying 200 copies of something the most likely cause would be input error followed by correction.
More likely backorders and/or pre-orders being filled.

The computer would update the stock as each title is removed from stock, stuffed and mailed.

It would not surprise me in the slightest that Amazon could get a couple hundred out within hours.

Add to this the fact that Amazon has multiple, regional warehouses for fullfillment. Stock could be replenished at one site, then filled at that site moving the numbers by smaller increments of the whole, also...

of the 30 blu-ray titles i have bought from Amazon, 4 were shipped from some place other then Amazon. these 4 were separate from the other disks, so i think there is some on demand going on for slower sale titles.

So there is no incentive for Amazon to keep the arrival inventory data posted or correctly showw all inventory reductions. They may even summarize stuff in their input as long as teh end of the cycle number on han dis accurate for a consumer buying decision purpose.

I'm not so sure. It is obvious that the stock levels are done in s/w. Unless they apply a random weightage to each stock level - so as to not let people know about the actual stock level - the stock level shown should be the correct one.

Also I do mot know if any its are drop shipped form seperate inventory systems.

This is possible - and most likely too.

I'm not so sure. It is obvious that the stock levels are done in s/w. Unless they apply a random weightage to each stock level - so as to not let people know about the actual stock level - the stock level shown should be the correct one.

Natarj, it could also be lots instead of titles.

What is the difference between census and sample?
I was offline for a while.

I see that Kosty explained the terms. The email that I received didn't have anything other than the attached documents that I already posted. We are even as far as information is concerned.

You do realize that not everything exists in cyberspace and can be linked. I already put a direct quote of what I was told in a previous post. Look back through if you missed it. Maybe someone wasn't doing their job and did send you a list, but I still find it hard to believe they would violate their confidentality agreement to a stranger off the street. I'll believe it when I see it.

However, Bookscan does mention Amazon.com as a partner, so it isn't a stretch that they would contribute to Videoscan too.Well, believe it or not. That's your prerogative! You can try to email David yourself.

As Kosty said, these documents look more like marketing material than confidential documents. Anyway, I am not going to try to convince you one way or the other.

PS: I might have missed your quote about the contract. I will try to go over past few posts.

I got the original word documents you sent through Gmail.

Yep those are their marketing documents.

No breach of confidentiality. Very good job sir. :)Yes, I tried again. Luckily, it went through fine.


I sent an email to the person that sent me the documents to clarify about the confidentiality of the documents. I don't want to put that person in trouble.

I still find it hard to believe they would violate their confidentality agreement to a stranger off the street. I'll believe it when I see it.



So what do you think happened? Do you think Kdragon sat down and typed up that long list, neatly dividng the companies into sample and census groups?

Clearly not.

They made that list to market their services, and they sent it out upon request. It looks pretty simple to me, I'm not sure what there is for you to "believe it when I see it."

Now that we have the list - does anyone think it could represent only a third of the sales ? Apart from Wal*Mart (who are probably not that big a seller of hidef dvds) are there any other known majors missing ?

Otherwise we have to come back to the apparent discrepancy between 1.5M shipped figure given sometimeback and the 500K or so figure given by Videoscan.

Apart from Wal*Mart (who are probably not that big a seller of hidef dvds) are there any other known majors missing ?



I didn't see DVDempire on there, but I need to check again.

So while it might be slightly unethical wouldn't we gain lots of understanding regarding how the sales ranking works if someone ordered and then an hour later cancelled a 100 copies each of titles around say 50,200,500 and 1000 on the sales ranking. Watch what happens to stock and ranking. Should clear up a few things.

So while it might be slightly unethical wouldn't we gain lots of understanding regarding how the sales ranking works if someone ordered and then an hour later cancelled a 100 copies each of titles around say 50,200,500 and 1000 on the sales ranking. Watch what happens to stock and ranking. Should clear up a few things.
Yeah, you're all over it. Let us know how it turns out.

Maybe someone wasn't doing their job and did send you a list, but I still find it hard to believe they would violate their confidentality agreement to a stranger off the street. I'll believe it when I see it.
You're funny.

I didn't see DVDempire on there, but I need to check again.

When we start thinging about majors, we have to start looking for Fry's, Sam Goody, etc, as well as Canadian retailers like Future Shop (they are a division of Best Buy, but would they be listed separately?). Other large retailers like Meijers, Costco, etc may also be retailers to check for. It would be easier of a copy of the list was just hosted somewhere so people could check it for themselves.

EDIT: I see the pdf copy that was hosted now and I'm looking it over. Off the cuff, I don't see Future Shop. I also don't see Deep Discount, which would probably be a signifigant online source.

Kosty,

While trying to mail those documents to you, I ran into difficulties with the mail server!!! They are only about 5MB each, word documents. No wonder MS went XML in Office 2007.

Anyway, I decided to convert them into PDF (I wanted to keep them in the form I received, but I have no choice right now). Instead of sending them to you, I am posting them here. I will try again later to send the original word documents to you.

Here they are:

I guess now I wonder when Amazon.com started contributing to Nielsen, if they indeed do so. Here is a 2001 article originally from the Hollywood Reporter. The bold emphasis added by myself.

'The Mummy' vs. 'Menace': That's a wrap

Nov. 01, 2001
By Brett Sporich


An independent review of first-week DVD sales shows that 20th Century Fox Home Entertainment's "Star Wars: Episode I -- The Phantom Menace" did not break the first-week DVD sales record set by Universal Studios Home Video's "The Mummy Returns."

Despite a Fox news release and related stories in the media claiming that "Menace" had broken "Mummy's" record, VideoScan research shows that the "Menace" DVD came up short in a first-week sales comparison.

While VideoScan does not take into account Wal-Mart, Toys 'R' Us or Amazon.com, "Mummy" outsold "Menace" by 2-to-1 in their respective first weeks at Wal-Mart and Kmart, industry sources said. "Menace" DVD sales did top "Mummy" on Amazon.com.

Overall, DVD sales show that while "Mummy" continues to lead in total sales, "Menace" is gaining ground quickly, taking the top DVD sales slot for the week ending Oct. 21, and it leads VideoScan's First Alert DVD sales chart for the week ending Oct. 28.

Fox's Eddie Murphy starrer "Dr. Dolittle 2" topped the video rental chart its debut week, earning an estimated $7.3 million after five days on rental shelves, Video Store magazine research shows.

"Mummy" placed second on the rental chart, earning more than $7 million last week, for a cumulative total of $38.1 million.

Columbia TriStar Home Entertainment's debut of "Final Fantasy: The Spirits Within" was the third most-rented video last week, earning $5.2 million after five days on rental shelves.

VideoScan is a service of VNU and ACNielsen that collects VHS and DVD sales data from a sampling of all categories of retail stores. VideoScan charts do not include sales data from Wal-Mart, Toys 'R' Us or Amazon.com. Most other sell-through retailers are represented.

Video Store magazine, a leading business-to-business weekly serving the home entertainment industry, compiles and analyzes VHS and DVD rental data through an interactive methodology using a multisource predictive model with data from a statistically significant national sample of video retailers."

Other articles from 2002 and 2003 also state Amazon does not contribute to Videoscan. The reason I find it hard to believe is because I was told last fall by an Amazon employee at their Lexington, KY wharehouse facility that their sales are not reported to Videoscan. This makes me wonder if maybe they started in Jan. 2007 or something recently, if these documents are legit. If so, a big chunck of sales could have not been reported.

EDIT: Looking back through some trade publications, I found a 2003 article that stated Toys R Us was following in Wal-Mart's footsteps and discontinuing to supply data to organizations like ACNielsen and NPD. I also found 2003 and 2004 articles from the Hollywood reporter that says that Toys R Us does not supply data to Videoscan. They are on the list, so did they just do a 180 after making a big fuss about it in 2003?

Also, why does the list say updated February 2006 at the top, but then a couple of lines later, it says October 2006? Which is it? Doesn't seem like a very good marketing document with such a glaring error. Also, why is a place like Nobody Beats the Wiz still listed? The list states included until 4/20/03. If the list was updated multiple times since then, all the way through 2006, why was it not removed (shoudln't Wal-Mart be on here if everyone who has ever contributed is listed)? Even further, this list contains businesses that don't even sell products, like Ntunes.com, an online marketing service. Ntunes (a Nsyndicate company) and Nsyndicate.com used to create private label stores and sell items (including DVD's) on commission. I say "used to" because that was back in 2000. They have since went out of business and the Ntunes domain was purchased by a marketing company. The Nsyndicate domain name is still for sell. So how are they on an "updated" list? There are also several typos and a few .com domains that don't work (for example, there is no lonestarmus.com domain but, there is a lonestarmusic.com, so is it a typo or was there a lonestarmus.com site that has since went out of business). These are just a few that I caught without doing any kind of in depth checking.

So why all these errors if this is an "updated" marketing document? This document is starting to look like an old document with a few obvious errors corrected, like deleting Wal-Mart (which would have appeared in the blank space after Target on Page 1 , Column 1), change the date and pass it off as new. If this is really a marketing document for Nielsen, it is one of the worst ones I've seen from a multi-million dollar company.

There is something very fishy about this "official" list.

census is raw data 100% off their Point of Sale system, usually transferred electronically via EDI.

sample is just that , their POS system doesn't track or can't transfer real time data so the information is gathered from a random sample to accurately represent their sales volume.

Can a random sample accurately track the sales of something new like HD optical discs, especially when volumes are very low, or are not consistent?

It seems like if this is truely how they collect their data, the majority of Videoscan contributors do so via sample. Those samples would then be used to extrapolate a sales number. That extrapolated sales figure would then be added to census figures, and Videoscan would then analyze those combined numbers to get SI and YTD. This method seems like it could have a signifigant amount of error for a relatively low volume product.

EDIT: Nevermind. The list is organized a bit oddly and looking at it closer, it appears mostly groceries use sampling. I doubt they are a source of hi-def sales.

Any update coming for the initial post? I think that would be the first week ending of February...

Any update coming for the initial post? I think that would be the first week ending of February...Looks like Mondays are now the most likely days for these updates. I believe that we'll get some numbers for the PS3 sales in January tomorrow and hopefully for the sales of the XBOX360 add-on for January also.

--Darin

Kosty,

While trying to mail those documents to you, I ran into difficulties with the mail server!!! They are only about 5MB each, word documents. No wonder MS went XML in Office 2007.

Anyway, I decided to convert them into PDF (I wanted to keep them in the form I received, but I have no choice right now). Instead of sending them to you, I am posting them here. I will try again later to send the original word documents to you.

Here they are:

Well, after looking over the list a bit more, I'm even more convinced it is a fake. In addition to the inconsistencies mentioned in my previous post. I've found a couple more of the retailers on that list that are no longer in business, like Ntunes. They were e-commerce business that sprung up with the .com boom and then went out of business a few years later (some out since 2001).

The more I examine the list, the more I'm convinced it is an old Nielsen list that has had the dates altered (and altered inconsistantly) to make it appear more recent, or it is a complete fake. Either that, or part of Nielsen's "60% of the market" includes business that have been out of business for half a decade.

I've contacted Nielsen in regards to this list and the errors found with minimal research (I would expect to find many more errors if I went over it thouroughly). I'll post their response when one is received.

To others: sorry for this long post. I think this has now become quite OT. I won't post on this subject after this. Just one last bit of clarification. I don't like when someone doubts my integrity, so please excuse me...

There is something very fishy about this "official" list.

You need to get over this. One last time: I got the list from Neilsen. In fact I just got a confirmation email back from the same person who sent the list that I can keep the documents on this forum. Yes, I had emailed to that person that I had posted the documents on this Internet forum and wanted their permission. I don't like to live in guilt. Clearly they have no problem making this list public. I suspect, they may be happy about it. So, the "violation of confidentiality" based conspiracy theories that you proposed are now invalid! Give it up. If you don't believe me, send a reference to that post to anyone in Neilsen, and let us know what they say! By the way, I sent the email from my personal email account (with real name), just FYI.

First you said "I'll believe it when I see it". Now that you saw it (I posted it even before you said that), you think something is fishy! I like people with healthy skepticism and, in fact, I am a skeptical person myself. But, what you are going is beyond that.

I have no interest in a lesson in the history of Neilsen right now. I don't care how they reported in 2003 or 2004. Maybe, I won't mind, but not in this context. I was interested in knowing what they are reporting now (especially whether Amazon.com is included or not), and I got just such a list from them which I shared here. If you have your beliefs based on past information, then I think you know what is wrong. (Hint: things change...).

Just so that you know, when I received the documents, the first thing I noticed was the date discrepancy (well, after noticing Amazon.com). But, human errors like forgetting to update the dates at two places in the document are not that rare. I just checked: the "updated in February 2006" is in the header section. Someone forgot/didn't know/care to change the text in header section. That doesn't count as "something fishy". Even without this, Occam's Razor is quit useful here. I hope you see its value not just here but on other such occasions! I would assume October 2006 to be the date this was last updated, unless you want to assume that someone edited the document without changing the dates at all.

Please don't take offense if I don't reply to your posts on this topic anymore.

Okay, last one... :)
I've contacted Nielsen in regards to this list and the errors found with minimal research (I would expect to find many more errors if I went over it thouroughly). I'll post their response when one is received.This is the best thing you have done so far. Thank you (I think Neilsen will also thank you!).

Inaccuracies in the documents from Neilsen apart, I am getting ready to accept an apology from you when you confirm that this list indeed is from Neilsen. :) j/k (You do realize that their own mistakes don't make this list a fake, right?)