Quantcast
View Post
Markusdragon said:
stof said:

Think of it this way. Lets say he made a site counting the human population. So he sets the site to 6.8 billion and sets the numbers to increase by 32,000 every day. What's the problem there? The problem is where did he get either of those numbers from?

Essentially the stie is just some numbers pulled from who knows where, with a counter to make it look like they're actually following sales.


But you have no proof that he's not altered the update figures based on sales trends between the last official sales figures. Whilst this inaccurate simply on the basis that life never works like that, it's still a statistically viable estimate, and the mere fact that it's updated 24/7to reflect an updated estimate means that it is more accurate than the other method; leaving the values at verified sales figures for extended periods of time.

As for the whole 'pulled numbers from who knows where' issue, that's merely an issue with the lack of transparency to his sales data collection, and a problem that it could also be argued that VGChartz suffers from.

This was a while back, but I sat there and followed the numbers closely for a day or so. Writing down all the numbers, updating every few min (while I was at work), etc.

All the numbers were perfectly interpolated - i.e. there were no updates in-between (using the same start/end figures).

I then emailed the guy and asked him explicitly about this - and he admitted that this was exactly how the site worked (I may have that email still around somewhere).

As for the correctness of the numbers - I can't comment. He might have some sources that we don't - but I don't believe so. I think that our numbers are definitely more accurate.

 



Gesta Non Verba

Nocturnal is helping companies get cheaper game ratings in Australia:

Game Assessment website

Wii code: 2263 4706 2910 1099