Originally Posted by Kosty
It kinda doesn't make sense that either Blu-ray nor HD DVD would drop in sales that much after the first of April.
Indeed. Actually, both numbers for BR and HD on Apr 1 seem overly high to me, based on past figures on big release dates like The Departed and Casino Royale. This is really why I think the 11-12K worth of "missing" sales most likely reappeared in this period. It's possible that I am wrong and that those numbers remain unaccounted for, but that seems unlikely this many weeks out unless the same retailers continue to be late with reporting.
I think the use of first alert data with the hard Jan 1st - Mar 18th is going to be an issue
There's no question about it. It's definitely a problem, and with Home Media Magazine dropping down to only full percentages, the problem gets worse. It's getting increasingly hard to read the tea leaves even with my spreadsheet.
We made be able to correct it though after the next weeks corrected SI and YTD numbers show up.
What would happened if we pretended we did not have the first alert data for the new week?
Neither of these is possible to do. First Alert data is always "contaminated", and there's no way to remove the "bad" part of the data from the "good" part.
Can we just correctly calculate last weeks weekly number from this weeks SI and YTD numbers? Assuming those numbers at least are incorporating the corrected final numbers for last week and this weeks first alert numbers?
No can do. We don't really know what the real numbers are; all we have are ratios, and while we can work some minor miracles with those ratios (as you've seen), the limitation is that we don't really know what the actual volumes are, and that we are subject to whatever error is inherent to the data.
Grubert; Any chance at all at getting the corrected final numbers from last week as an update to the first alert numbers?
This would be the best possible outcome. What we really need are the finalized ratios, with the 100:xx.xx precision. Those numbers give us the most accurate projections. We're losing a ton of detail with the new way HMM is posting their data, and I'm very concerned about the accuracy of our projections at this point (for a HT analogy, what we're looking at now is a 480i image with EE instead of the full 1080p).