Auron - Ive been singing that tune for the last week now. I say he shouldn't change ANY numbers and only go by his samples. OR if he changes numbers change them for ALL the other tracking sites.

I've been arguing with morons for a while about it. You won't win Auron. It's a lost cause.

When you say "If he changes them for Jap and Others he should change them for NPD"

They respond with "Why? he might not think NPD's numbers are accurate and if he goes by NPD he shouldn't do what he does. He wants to go by his own samples"

Then you respond with "then dont change them at all. He should go by his own samples and not change them for Japan or Others"

Them - "But his numbers aren't accurate in Japan or Others since he doesn't have big enough coverage there"

Me - "Then he should change them to a groups numbers in NA who has bigger coverage and are more accurate"

Back to the begining.

It's really too bad that the NPD thread got so long, because I actually covered all this a few times.

He doesn't adjust numbers for Japan ever as far as I know, and since Famitsu and MediaCreate both come out with weekly data, he doesn't adjust to them. He'll adjust to the big threes financial reports, but NPD does it too (you just don't see it).

As for Pal region, I have a few points, which will hopefully make what I'm trying to say clear.

1. The market size in each PAL region is much smaller than the market size of the US, and most are smaller than Japan (I would think only UK can rival Japan). This presents two problems, one of which PooperScooper mentioned, though he doesn't realize why it's not good.
2. The first problem this presents is that 10% sample of a smaller market might not be as representative of that market as a 10% sample is of a larger market. Meaning, with 10% of the US data, you could make very accurate predictions, but 10% of the German market (which would only be like, 1/100th of the US market), you might not get accurate sales trends. So saying "if he doesn't have a good sample size in Germany than his sample size isn't good enough in the US either" is not correct. If he has 10% of the US market that's more than enough to get fairly close estimates of what the real numbers are (which he proves week in and week out), even though compared to 60% NPD has doesn't seem like a lot. Without Toys R Us and Wal-Mart, there's going to be guesswork, and estimation, so neither is going to be 100% accurate, so how is it to say that NPD is more accurate? As for PAL, he might not even have 10%, as it seems that PAL is the "weakest" market when it comes to tracking. Also, ChartTrack or whomever had those Germany and UK numbers said they track 80% of the market, which is probably a lot better than ioi's probable 5% (which, like I said, probably wouldn't be a representative sample of the market).
3. The second problem is scope. With a smaller market, if you're off by 100,000, you might be off by more than 25% of that consoles real sales. In bigger markets, 100k is nothing. Overall, 100k here and there would make maybe half a million difference, but since he adjusts to manufacturing data, it's never going to get bad enough to be a big deal.

I hope that makes sense. Take a statistics course (i'm taking one right now), and representative samples make all the difference, which may be a problem for him in PAL regions, while not so in the US/NA. 

The reason they don't blindly adjust for NPD is because the last two times Ioi did adjust for NPD... NPD turned out to be wrong and screwed up VGchartz numbers.

While this isn't true with the other tracking sites.

In other words, NPD is much less accurate tracking america then Charttrack, Mediacreate and Famitsu are at tracking their respective regions.

Of course you need to wait about a month or so before you adjust for mediacrate and famitsu. Since if they make a mistake one week they "make up" for it the next week by adding sales, rather then adjusting figures.

If you were an investor who does it yourself and had an advisor for Stocks, one for bonds and one for mutal funds are you a hypocrite if you take some advice from the stocks guy and the mutal fund guy but not the bonds guy... eve if the bonds guy consistantly steers you wrong?

 Ah see you make the mistake of thinking he adjusts to their numbers directly, he doesn't he uses their numbers as a signal to check his data and contact his sources and the companies for their shipment data, and based on that information he adjusts

uggh!!! He did not adjust the DS at all to NPD data, if he does, it will be tomorrow. ioi had DS selling 200k nearly every week now go back and look.. it starts at 180k in mid jan and goes slowly down. Last week he had the DS at 201k now it is 160k... THAT WAS THE ADJUSTMENT!!!!!

 Ah cool, thanks for double checking