• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

Revised Nov 2005 NPD Data Expected Tomorrow

donny2112

Member
http://biz.gamedaily.com/articles.asp?article_id=11330#11330

Revised NPD Data Expected Tomorrow

Following the news that it had retracted its own November data last week, The NPD Group has issued an alert stating that they are "targeting release of the November data by close of business tomorrow (Tuesday, December 13, 2005)."

"The entire Games group is still conducting extensive reviews to ensure that tomorrow's data release happens smoothly," said NPD.

After analysts predicted a poor month, it will be interesting to see how much sales were down for November.

Monday, December 12, 2005

Who knows what data (if any) will make its way onto the boards, but at least it's coming. :)
 
Just an FYI, those NPD numbers are not very accurate. They are based on numbers from participating retailers (Target, Best Buy, Gamestop, EB, Toys R Us, ...) which account to 63% of the U.S. overall retail market. Walmart, the biggest retailers with over 22% of the market is not accounted for, nor are the 15% of the smaller retailers. No one knows what the exact formula for extrapolating the rest of the market. NPD charges thousands for its subscription and it is not very transparent service. Hopefully this episode will force them to clean up their service.
 
Amused_To_Death said:
Just an FYI, those NPD numbers are not very accurate. They are based on numbers from participating retailers (Target, Best Buy, Gamestop, EB, Toys R Us, ...) which account to 63% of the U.S. overall retail market. Walmart, the biggest retailers with over 22% of the market is not accounted for, nor are the 15% of the smaller retailers. No one knows what the exact formula for extrapolating the rest of the market. NPD charges thousands for its subscription and it is not very transparent service. Hopefully this episode will force them to clean up their service.

63% is more then a good enough sample to make a calculated guess on the markets it doesn't cover.
 
Actually it used to be 63%. This month though they only covered 53% directly and had to extrapolate the rest. That is the reason why the NPD was wrong this month, their adjusted calculations were fucked.
 
NPD is only accurate when your system of choice is doing better, otherwise the numbers are inaccurate. This also applies to systems you hate, only vice versa.
 
Amused_To_Death said:
Just an FYI, those NPD numbers are not very accurate. They are based on numbers from participating retailers (Target, Best Buy, Gamestop, EB, Toys R Us, ...) which account to 63% of the U.S. overall retail market. Walmart, the biggest retailers with over 22% of the market is not accounted for, nor are the 15% of the smaller retailers. No one knows what the exact formula for extrapolating the rest of the market. NPD charges thousands for its subscription and it is not very transparent service. Hopefully this episode will force them to clean up their service.

Holy shit tell me you did not just post the above like it was some new piece of info(incorrect as you are with some of your %'s).
 
AbeFroman said:
I expect 360 hardware numbers will greater than previously released 332K.

Most likely... it is nearly impossible for NPD to extrapolate a launch product. There is no way they can know how many units WalMart got, especially since MS partnered up with BestBuy and WalMart.

53% or 63% of a market is not a very good sample when you omit 47% of the source. If NPD was to sample 53% of all retailer stores, the sample would be good. However, when you only sample 53% of the prospective retailers, then you can't extrapolate the other 47% without knowing what % of the market those 47% hold. Even then, it makes it improbable when MS is cutting deals to get more units to those retailers.
 
jedimike said:
Most likely... it is nearly impossible for NPD to extrapolate a launch product. There is no way they can know how many units WalMart got, especially since MS partnered up with BestBuy and WalMart.

53% or 63% of a market is not a very good sample when you omit 47% of the source. If NPD was to sample 53% of all retailer stores, the sample would be good. However, when you only sample 53% of the prospective retailers, then you can't extrapolate the other 47% without knowing what % of the market those 47% hold. Even then, it makes it improbable when MS is cutting deals to get more units to those retailers.

..pre...emptive... dc? ;) They're gonna sell everything they can get in to the US stores no doubt.
 
Let's follow a little logic:

1. NPD says they normally track 63% of the market, but this month only 53%.

2. What's happened that could've changed that? X360 launch is the obvious thing. Black Friday sales might be another candidate. Anyone know differently? Did they say why?

3. NPD has stated their numbers were inflated, not understated, and subsequent reports say they were hugely overstated, not a little bit--maybe as much as 16% of ~$800 million.

4. Tracking of regular product channels should be fairly routine, hence constant.

All this suggests that what's thrown off their numbers is the usual distribution of X360 systems and games. And since they've said their numbers were too high, it follows that they overstated the X360 impact, either the hardware (unlikely) or software/accessories.

My bet is that the hardware numbers come in pretty close to what they were, but software takes a big hit, particularly X360 software. There's also the possibility of software taking a bigger hit on dollars than units because of Black Friday sales--sales as in lower retail prices, that is.
 
Leondexter said:
2. What's happened that could've changed that? X360 launch is the obvious thing. Black Friday sales might be another candidate. Anyone know differently? Did they say why?

Wasn't the GameStop/EB merger finalized in November? I thought someone stated that that was the cause of the percentage change, though I'm not sure how they could be less together than apart (store closings/consolidations?).
 
DarienA said:
..pre...emptive... dc? ;) They're gonna sell everything they can get in to the US stores no doubt.

Not really... it's just more a question about NPD being accurate. NPD is the standard and I think that they do a good job of getting accurate numbers most of the time. Especially when they have historical data and can follow trends. I just find it difficult to believe that they can accurately track a launch product during a busy Holiday month where consumer spending has yet to be determined; given the fact that only 53% of retailers give them any info and MS was partnered with a retailer not on the list.

If I had a preference, I would use manufacturer shipped numbers over NPD sold. Unfortunately, we only get those once a quarter.
 
Leondexter said:
Let's follow a little logic:

1. NPD says they normally track 63% of the market, but this month only 53%.

2. What's happened that could've changed that? X360 launch is the obvious thing. Black Friday sales might be another candidate. Anyone know differently? Did they say why?

3. NPD has stated their numbers were inflated, not understated, and subsequent reports say they were hugely overstated, not a little bit--maybe as much as 16% of ~$800 million.

4. Tracking of regular product channels should be fairly routine, hence constant.

All this suggests that what's thrown off their numbers is the usual distribution of X360 systems and games. And since they've said their numbers were too high, it follows that they overstated the X360 impact, either the hardware (unlikely) or software/accessories.

My bet is that the hardware numbers come in pretty close to what they were, but software takes a big hit, particularly X360 software. There's also the possibility of software taking a bigger hit on dollars than units because of Black Friday sales--sales as in lower retail prices, that is.
change that opening line to lets follow a little SPECULATION
 
When they retracted it, didnt they say theyd have the new numbers out in 24 hours?

:lol

These guys run the best business.

Long delays? Check. Innacurate numbers? Check. Subscribers paying thousands of dollars? Check.
 
AbeFroman said:
I expect 360 hardware numbers will greater than previously released 332K.
Considering the error was that they overstated the sales numbers, I'm going to put this in the "not likely" stack.
 
FYI,

NPD does extensive market research across 150,000 shoppers across the nation each and every month. It's the basis of their entire business which, believe it or not, includes a hell of a lot more than just video games.

They also do work directly with Wal-Mart and while, they may not have exact Wal-Mart totals from their system, they do have a very good idea where Wal-Mart shifts.

They have been quite slow to update their historical market share numbers and from what I've heard, they feel extremely confident that they should have revised their market share numbers down. However, they seem to have placed a bit too much importance on some very strange pollings in November (which was true for all of retail, thus Wall Street and the media having a very difficult time telling how the holiday season is going) and really, really sent the analysts over the deep end. It was too much too fast and on too little data.

It's always been a projected number and I'd say the numbers have been pretty damn good and their trying to react to a changing marketplace. Wal-Mart & Best Buy in particular seem to continue to take market share at a pretty heavy pace.

Although, I suppose it's good for this forum to place less importance on NPD.

Amused_To_Death said:
Just an FYI, those NPD numbers are not very accurate. They are based on numbers from participating retailers (Target, Best Buy, Gamestop, EB, Toys R Us, ...) which account to 63% of the U.S. overall retail market. Walmart, the biggest retailers with over 22% of the market is not accounted for, nor are the 15% of the smaller retailers. No one knows what the exact formula for extrapolating the rest of the market. NPD charges thousands for its subscription and it is not very transparent service. Hopefully this episode will force them to clean up their service.

I know that you know this and that you know NPD's reporting, but they have some damn good metrics from a statistical standpoint. They have much more information that just historical information and raw numbers from participating retailers of which I touched on above. The biggest problem is that they made a big change at an important time and caught people off guard. The second biggest problem, IMO, is the truly substantial one.

They do not do ~much accounting for monster sales (Wal-Mart having GBA @ $44, TRU having a 3 for 2 sale, etc) and they also don't account much for certain products / genres / target markets in the NPD. They just use a relatively simple market share calculation with a couple of modifiers and apply it to all of the games / systems / accessories. It doesn't make any significant statistical difference to the overall market, or honestly to 95% of products. However, there are a number of products that they can be substantially wrong about.

In any case, I hope that none of you harbor the illusion that any other tracking systems are much better. NPD goes MUCH farther than anybody else in terms of determining market shares, and their achilles heel is that Wal-Mart opted out of participation. The higher the participation, the better your results, obviously. But, just so you know, statistically, they have unbelievably accurate numbers.

Why do you think that Media Create / Dengeki / Famitsu can report such widely disparate numbers? And Europe is just a complete mess with all of the countries and selections of retailers.

I'm certainly not saying NPD is perfect and they could use to do ALOT better in terms of retailer participation, timeliness (but, actually, this is also retailers faults) and a more complex extrapolation for the spikes in their data.

jamesinclair said:
When they retracted it, didnt they say theyd have the new numbers out in 24 hours?

:lol

These guys run the best business.

Long delays? Check. Innacurate numbers? Check. Subscribers paying thousands of dollars? Check.

Actually, they didn't say that. They said they'd give subscribers an informational update on Friday, which they did. And it's not like you're paying a penny for the information, so please don't try to speak for those folks who do. I'm sure there's ALOT of upset people, but it's more complex than simply calling them slow and inaccurate.
 
sonycowboy said:
FYI,

NPD does extensive market research across 150,000 shoppers across the nation each and every month. It's the basis of their entire business which, believe it or not, includes a hell of a lot more than just video games.

I can confirm this since I am one of their survey "shoppers". They send me a survey maybe once or twice a week about various things. About a month ago they had one very extensive one on video games (believe it or not....from systems down to individual titles), which I was going to post about here but thought I'd get in trouble. :P
 
You know, I'm wondering if doing a phone-survey of people and asking them if the bought a video game product... what'd they buy.... and where? Would be just as reliable as the NPD survey of retailers.






Kinda reminds me of the debate between the "Household Survey (Employment Rate)" and the "Payroll Survey (New/Lost Jobs)" the Labor Department makes every month.
 
sonycowboy said:
Actually, they didn't say that. They said they'd give subscribers an informational update on Friday, which they did. And it's not like you're paying a penny for the information, so please don't try to speak for those folks who do. I'm sure there's ALOT of upset people, but it's more complex than simply calling them slow and inaccurate.

Oh no, you're definitly right, Im grateful I get free information from you guys.

But you must admit theyre slow. Japan gets weekly sales, and the US movie industry gets daily sales, but videogames are stuck on a monthly basis.

And yeah, Im aware they track bra sales and seperate SKUs.
 
jamesinclair said:
Oh no, you're definitly right, Im grateful I get free information from you guys.

But you must admit theyre slow. Japan gets weekly sales, and the US movie industry gets daily sales, but videogames are stuck on a monthly basis.

And yeah, Im aware they track bra sales and seperate SKUs.

Up until April 2002, we were getting weekly NPD updates, but retailers opted and complained it was too much of an admistrative burden to produce the weekly exports (along with a some other reporting they're required to do).
 
What the heck are you guys talking about? NPD is HUGELY accurate. How big of a sample size do you guys THINK is needed to provide accurate results? It certainly is a hell of a lot less than the 50%+ that NPD actually gathers...
 
GDJustin said:
What the heck are you guys talking about? NPD is HUGELY accurate. How big of a sample size do you guys THINK is needed to provide accurate results? It certainly is a hell of a lot less than the 50%+ that NPD actually gathers...

I mostly agree. The question is knowing what number they have. Having actual numbers only helps if you KNOW the sample size. 63% vs 53% makes for a huge difference.

IMO, NPD does know pretty well, but got a tad sloppy and moved too fast for the analysts liking.
 
sonycowboy said:
I mostly agree. The question is knowing what number they have. Having actual numbers only helps if you KNOW the sample size. 63% vs 53% makes for a huge difference.

IMO, NPD does know pretty well, but got a tad sloppy and moved too fast for the analysts liking.

Well I haven't heard anything to this effect so keep in mind that this is STRCTLY my speculation, but my guess would be their sample size got smaller (dropped a retail chain?) but they didn't change their formula enough to compensate for that smaller sampling... so their end data was jacked up.

Again, strictly my guess...
 
GDJustin said:
Well I haven't heard anything to this effect so keep in mind that this is STRCTLY my speculation, but my guess would be their sample size got smaller (dropped a retail chain?) but they didn't change their formula enough to compensate for that smaller sampling... so their end data was jacked up.

Again, strictly my guess...

Almost every single analyst report on Friday went over this on Friday. To my knowledge, they haven't lost any significant retailer since Wal-Mart some years ago, and they've added some of the online folks since then.

I'll quote a bit of one, although they're pretty much all the same:

Wedbush said:
In its first data release, the NPD group projected total U.S. console software sales for November of $818 million, down 4% compared to last year and significantly higher than our $715 million estimate. Unfortunately, the methodology used by the NPD to extrapolate the data from raw data captured to projected industry sales was materially different than the methodology used in the past. In past extrapolations, the NPD presumed that its raw data set captured approximately 62 – 64% of all sales, and it grossed up the raw data in arriving at its projections by the inverse of this figure. This month, the NPD presumed that its raw data set captured only 53% of all sales, and its gross up factor was materially higher than normal.
 
As a little tidbit for tomorrow, don't be suprised if LucasArts is VERY high in the publisher market share rankings beating out someone you wouldn't expect them to. That is if they don't change them based on the recalculations (just recalculating the numbers equally).
 
sonycowboy said:
As a little tidbit for tomorrow, don't be suprised if LucasArts is VERY high in the publisher market share rankings beating out someone you wouldn't expect them to. That is if they don't change them based on the recalculations (just recalculating the numbers equally).

Does that mean the numbers are gonna be posted here? :D
 
sonycowboy said:
Almost every single analyst report on Friday went over this on Friday. To my knowledge, they haven't lost any significant retailer since Wal-Mart some years ago, and they've added some of the online folks since then.

I'll quote a bit of one, although they're pretty much all the same:

Ah, alright. I didn't write our NPD story... so I didn't read any of those analyst notes. Good to see my speculation was on base :)
 
sonycowboy said:
Up until April 2002, we were getting weekly NPD updates, but retailers opted and complained it was too much of an admistrative burden to produce the weekly exports (along with a some other reporting they're required to do).

Do you think thats a valid complaint though? Dont retailers have a database keeping track of whats sold? How hard can it be to forward it to the NPD?

This is indeed a question, not sarcasm,


Also, I know you arent as liberal as Bunkum on releasing sales, and thats fully understandable, but would you be up to releasing a few key numbers?

For example, instead of the whole software chart, titles by request. Sonic Rush for example?
 
sonycowboy said:
As a little tidbit for tomorrow, don't be suprised if LucasArts is VERY high in the publisher market share rankings beating out someone you wouldn't expect them to. That is if they don't change them based on the recalculations (just recalculating the numbers equally).

D'oh! I didn't expect Battlefront II to do that well.

sonycowboy said:
Although, I suppose it's good for this forum to place less importance on NPD.

It seems like you've said this before. Is it your opinion that it is unlikely for GAF to be getting much in the way of sales numbers from here on out? Just curious.
 
In a world where companies shamelessly use lies in their PR, having NPD numbers made public should be a constitutional right. And yes I'm serious :D
 
The first numbers (sales in millions).

EA - $143.4
Activision - $97.3
THQ - $48.2
Take Two - $37.5
 
Top Bottom