Yeah, that was a bit crazy - 10 minutes after I posted the article Brett wrote me that the real figure was 1.3 million (instead of the 1.5 million we had earlier). No big deal at all but sorry for those of you who have been dissapointed.

Who would do such a thing to benefit their own agenda?

It's probably a good idea to be more clear if you are estimating heavily on a certain region. Other than that, I don't think you should do anything differently.

The hack who wrote the article said you needed to get more retail data....like you would actually turn down a company that was offering you data. There are a ton of things he was either not considering in his "article", or things he chose not to mention to further his "point". Things like: not all of your sample data arrives at the same time(which obviously leads to adjustments), often times the data evens out over time whether VGC started out high or low(even without public data to adjust to until after the fact), the often large disprepencies between longtime professional tracking firms Media Create and Famitsu on hardware and software, and the fact that in the past there has been publicly available NPD data that the NPD later adjusted.

I'd almost bet money that fella is a NeoGaf acolyte or is at least attempting to earn his stripes over there. Or maybe he's just a weasel. For the time being, we're going to have to accept that we don't have Gamasutra's blessings....we'll just have to settle for Forbes and the NY Times.

I really like how they picked out one example and repeated that constantly. Iron Man is the one flaw that they picked up and because one thing might not have been 100% accurate then they claim everything is flawed on the lower end games. Like wow.

That article is the definition of cherry picking, IMO. The article points out blatant flaws that can be found with any tracking services methods and then frowns upon Vgchartz specifically for the shortcomings of tracking services.

It turns out that the real story here is not YOU, it's all about trying to get Neogaf members to visit Gamasutra.  From what I make of it, JVM put pressure on the person from Gamasutra (his coworker or it could even possibly be himself who actually knows?) to try and write a smear campaign against this site to drive hits to their own site. They create the story under false pretense, post it on their site and then run to Neogaf to advertise their story to get hits. It's a marketing ploy.  The fact that the Gaf members are too dim witted to realize this is quite funny. 

Here read this quote taken from JVM: Disclosure: I write occasionally for Gamasutra. I reviewed a draft of the article before it was posted. I also write a monthly column for Next-Gen.biz which depends almost exclusively on NPD data.

Normal Gaf precedence would state that the entire thread should be locked and deleted.

If the MGS4 data from EU was "sketchy", then it really shouldn't have been posted. Or a disclaimer of sorts that the data was preliminary (like the Japanese sales), and that the data has a chance of being off by a few percent would have helped.

Just saying that would have negated that whole argument from them, but I guess the author felt that the news article was stated as fact, and that it had no margin of error...

ioi has already said as much.

I think something should be considered here.  These are growing pains.  That people even mention this site in the same breath as NPD and Media Create is a testament to how far it has come in its short-lived history.  But we can't expect this site to become more professional and more reliable without the poking and prodding of not only the gaming press, but also more mainstream business outlets like Forbes and CNNMoney.  You see, the reason all this matters so much is not for fanboys and trolls....it is for investors.  They need data that they can use to make smart predictions when they are putting their money on the line.

And a few articles to remind everyone that VGChartz isn't quite perfect yet is nothing to feel bad about.  In time this site will grow to include more retailers and the data will constantly be refined.

Should we expect everyone to embrace a budding web site?  We need to take our lumps, learn from them...and move on.  And the most important thing...don't make it personal.  A simple email from ioi would have sufficed to the writer of this article, rather than a public condemnation.

If you want to call it cherry picking, go ahead, but you're not accurately describing the situation. The Iron Man flaw is used to describe the systemic flaw in the methodology this site uses to track data. Yes, this site is good at giving a generally idea of sales trends, but it has significant limitations in the methodology as the article points out. The results might be good (i.e. they jive with NPD, Chart Track, etc), but the method is far more important. Just go ask any scientist doing research. He could be getting similar results to his peers, but if he is using flawed methods, his data is not valid.

Wrong again. Here's where they expand on the Simexchange comment:

"In other words, for those high-selling titles, VGChartz is checking against public data, and they will change their estimates if they are majorly off. Most of the time, they are quite close compared to the worldwide charts. That's because VGChartz is - like services such as The SimExchange - using common sense, Internet buzz, real-time data such as Amazon.com and analyst commentary to synthesize a sensible estimate."

The article never contradicts itself. It is completely coherent (if you actually read the entire thing). What you're doing is nitpicking the comparison between this site and the SimExchange, which is not the point of the article, instead of acknowledging the actual point of the article, which is to point out that the methods used aren't reliable enough to be used as a professional resource.