Allmost everyone on this side takes NPD numbers as facts and the tracking of VGC as good guesses, and i dont realy understand why.

Pros:

NPD:

we know the tracking methode

we roughly know the sampling size

VGC:

we are like jon snow, we know nothing

Cons:

NPD:

the sample isnt so good, its to big and it doesnt track a major retailer with a different demographic than the others(toys r us).

VGC:

we know nothing

okay, seems like NPD has the better tracking. but we cant be realy sure because we dont know if ioi is guessing or gets his sample in a shady way, or(what i think is right) that he doesnt want the sample size out because people are stupid and dont understand that a small sample could be better than a big.

npd doesnt give out the sample size, but they estimate that its 60-80% of all retail, that is a way to big sample to be good.

they need a lot of math to pretend they have a representative sample, but they have not because they dont track an important retailer for a specific and important demographic.  

i think that npd numbers of some games are bad, because they dont have a representative sample for the main demograhic for some games

skylanders comes to mind, a game heavly marketed to kids with a toyline that makes it attractiv for retailers that normaly dont sale so much video games.

what this thread needs:

an actual list of tracked retailers form npd

everything about samples from vgc.

some inside about how people buy games in the us, im from germany, we dont have the most retailers you have and that are in both countrys arent targeting the same demographic.  how many games has toys r us? are they saleing games at all? do they have regular sales on games targeted specific demographics.

clearification if npd just tracks amazon itself or amazon and the amazon market.

i couldnt find much and all i could find was older(im not even sure if npd tracks toys r us again, i just found the news that they stopped and nothing about them tracking it again so i guess they still dont track toys r us)

im not unhappy with any numbers, i just wanna learn more about it and i couldnt find usefull informations on this site or the web.

and 1% sample size is more than enough to get realy good data if the sample is represantativ.

Wat? The sample size is TOO big? How? What? The bigger the sample size, the more accurate the data.


Stop arguing against NPD, you're only making a fool out of yourself...

Discussion over Bayonetta 2 sales in the US have turned ugly.

Just stop the damage control, there's nothing to argue about NPD. They track more than 95% of the retail US market (and the top5 retailers), Toys r us is barely 1% nowadays.

NPD USA Retailer Marketshare, beginning of the 8th gen (approx. 2013):

1) GameStop - 30%
2) Wal-Mart - 25%
3) Target - 15%
4) Best Buy - 12%
5) Amazon - 6%
6) Toys R Us - 1%
----Top 6 comprise 89% of the USA market

a big sample size needs a lot more math to get to beeing close to represantive.

you need a random sample, thats realy hard to get if you track a big none random part.

source link?   btw thats the frist realy usefull post. 

do you have more? not just the overall data?

You don't need a random sample when you have a formula that isn't based on general coverage but instead is based on specific known trends, EG Toys'R Us will sell ~this much more kid friendly games than Gamestop etc.

Besides, NPD covers beetween 90-95% of retail video game marketshare, so it's not much they have to extrapolate.

So, to sum up your gripe: You believe that ToysRUs sold a incredibly large number of Bayonetta 2, and thus NPD data is 200K less than VGC, which is more accurate because they track less and extrapolate more.

Npd doenst sale just tracking to manufactures, nobody would by those numbers because they could generate those fairly accurate by themself. they buying data about costumer behaviour and other things, thats realy valuable, even if they have problems with some games from time to time.