ioi said:
@Just_Ben, who's being childish now?
I actually like the idea of putting together a margin of error statement for the site. My issue is that the only way to do it is experimentally using some of the data we post and basically compiling and comparing the accuracies. The second problem is that the only data we have to compare to is NPD, which I feel has reasonable errors in its own data. So unless we can somehow setup a random sample of data and find corresponding "real" data to do a study against (a third party can do this if they wish) then I struggle to see how we can do it. And it's certainly not something that can be demonstrated analytically. If a publisher wanted to partake in a study to compare their shipment and retailer feedback data to ours so that we can say "90% of the time we are within 10% of the real figures" or whatever I'd be happy!!
|
I can't really think of any good ways to do it either; if we only use titles that we have numbers for in press releases, end of year statements, etc, then we are just going to be matching data that we already match
However, if we just take a random sample of games on this site, we may easily get to the point where 50% of the games we're trying to match up to, there is no public data about. It's a conundrum to say the least; I've been trying to think of an analytical, substantative method to estimate this since this thread started, and I can't think of anything that will both be possible and unbiased, short of taking a random sample of our data and asking the publishers their shipment numbers for those titles if they haven't released them yet. Unfortunately, this would require contacts in the industry I don't have, and even with those they're likely to say 'no.'
All told, I am not sure that I can think of a way to estimate it in such a way that it will likely happen, short of broad generalizations like I posted above; 'unconfirmed by reports' 'confirmed by reports', etc.
That would at least give a very broad level of accuracy to the data; I would prefer something a bit more precise, but that may not be possible without additional help from publishers/distributors/whomever.
Please, PLEASE do NOT feed the trolls.
fksumot tag: "Sheik had to become a man to be useful. Or less useful. Might depend if you're bi."
--Predictions--
1) WiiFit will outsell the pokemans.
Current Status: 2009.01.10 70k till PKMN Yellow (Passed: Emerald, Crystal, FR/LG)