Funny story behind Mario Kart 8 Deluxe's attach ratio "numbers" ... 

From the article itself, Nintendo only claimed that Mario Kart 8 Deluxe sold nearly half a million units in it's first day at which the 459k figure corresponded to which was April 28th last year but NPD tracked the month up until the 29th so the sales for the game's second day was a total mystery ... 

This is fundamentally key to our understanding of why the sales projection of ~115k for the Switch during the April month failed since the attach ratio was only reported for the total games sales relative to the total system sales for which it matched the month tracked by NPD itself! If we naively extrapolated the numbers as they were we would've obtained the ~115k figure if we subtracted the extrapolated numbers with the launch numbers ... 

It wasn't until later in may that Nintendo had truly revealed the real numbers that Mario Kart 8 Deluxe sold in April simultaneously with the NPD data as well that day. Had we extrapolated from 550k figure instead of the instead of the 459k figure we would've gotten ~1.222M units compared to the ~1.020M units so it was no coincidence that the former figure ended up being the closest to the compiled NPD data (1.186M units) which brought back consistency to Nintendo's statement at the time they've released their PR statement ... 

Not only was Nintendo's statement regarding Mario Kart 8 Deluxe's attach ratio was soon quantifiable in the near future that time but the recent statements that Nintendo made such as "115% higher sales during the same period" and "best 5 day period" are simply intractable garbage that will stay as such in the foreseeable future but those people should feel bad for themselves for even trying create misinformation out of them when we have absolutely no specific data for those cases ... 

All in all, rol's attempt to deface Nintendo's internal data regarding the Switch ended up in failure due to his lack of proper care in interpreting the data which means Nintendo have yet to release highly inaccurate or highly imprecise data for the Switch but his posts in this thread at which he isn't the only guilty one in here just highlights a deflection from reality in plain sight ...

The beautiful thing about numbers is that they aren't biased ...  

You  should know by now, that he doesn't stick to his predictions, he moves them to cover his own obvious bias. This is why people like him and others gets called as well. Like I have prediction of 1-1.1 million for the switch. I also had prediction of 100,000 for Japan this week. I'm not going to live and die by these predictions, but there is a reason, why barrage of Nintendo fans are not attacking me.

It is also amusing the bias talk I had this week, and how it reflects users posts (It is double amusing when they claim not have said bias). If one user says a prediction, that should be the basis of the bet. Not pushing goal post higher. This happened in Feb-April period as well, he made a claim, someone said back it up with a bet. He moved the goal post, and that person was like meh.

Not only that but NPD hasn't really been giving unit numbers for many months now.  All we have gotten are "leaks" and estimates from "aggregate Predictions" (which I fully understand is the best way for the insiders to do this without stepping on any NDAs) but at the end of the day, these numbers haven't been officially confirmed by any of the three companies, so we don't really know exactly how accurate these numbers are.

Wow, are Nintendo fans eager for Nintendo to be wrong about their numbers. Not sure how they think the 8.2M+ was confusing, since that was one of the hard facts in their statement, with none of the other PR foolery with the numbers. Let me just post what I wrote in the BF thread.

I also want to clear up something about the 5 day vs BF week. There's a reason Nintendo used that period. It allowed the Switch to beat the Wii. My guess is that if they used the BF week, Wii would have won. You have to remember that Wii was a hot item in 2008. It was leaving the shelves almost as soon as it was put up. And for that BF Nintendo offered no deals on the system. It's not hard to imagine quite a few of its sales came from the Mon-Thu period, as there was no reason to wait til Black Friday, which at the time happened in the early hours of Fri morning, not Thu night, like now. Cyber Monday was also a MUCH smaller event than what it is today.

Edit:  Also, according to this list of deals, Best Buy, Circuit City, Radio Shack, Sears, Target, and Toys R' Us didn't even advertise the Wii console for Black Friday.  Not advertising the hottest gaming item for the craziest shopping day?  My guess is they didn't advertise it because of low stock.  It was also sold out by the end of BF, so it basically sold little to nothing Sat and Sun.  And like I said, CM was nothing like it was today, so if stores were reserving stock for it, it wouldn't have been much.

I wonder how many they sold on pokemon week.
maybe 1st week of November was 60k, pokemon week more like 220k, and BF week 700k. adds to 980k. A total of 8.26 million, adding from the 7.3 figure from October.
Maybe on the 800k from wii week 2008, it sold a lot on the 2 days not counted, letting it bellow 700k on the 5 days.
From, the vgchart week for switch last year, taking 2 of the 7 days could also have brought the number a little down, in order to 115% increase be 700k...

It's great to see the Switch doing well.

That's an amusing post. I disagree big time. predictions are for fun, i could think switch has a 60% chance of selling 850k-950k and make that prediction, for bets i like to feel 80-90% sure, so i raise the number a bit, nothing wrong with that, and people complaining about that just wanna complain.

I wouldn't say that it's because people like to complain. I mean you are certain that the Switch will sell 850-950k units so why not make that your bet?

many people are certain that switch's 8.2 million LTD is just counting thanksgiving through black friday weekend, but they are not taking the bet. aside from that nintendo pr numbers put it 850-900k with 5 days tracking left it's not a bet i'm fully confident in, it will be very close, ill bet it won't do over 1050000 though.

Why blame Nintendo's PR for your own ineptitude in not realizing that they used two different numbers ? The so called "mistake" that you keep touting is nothing more than a mirage you created ... 

Is there any justification to your claim of "wrong number" despite the margin of error being less then 30k and a little bit above 3% ? Your accusation of bias towards me makes no difference when you're the one that's off base trying as much as possible to deviate from the numbers that you don't like ... 

FWIW as far as the past comes, at least I based my thoughts on a real figure which were shipment projections while your scrambling about lost in finding a justification as to why Nintendo's internally collected data for the Switch is somehow "wrong" and you're back to having zero precedent as well ... 

I don't think I need to highlight the irony in how Nintendo data is good enough for you guys to throw the detractors in the mud but somehow the detractors can't use Nintendo against their fans so it's a one way street ... 

I dismissed the two statements because they provided no value in our in depth understanding of the Switch's sales, not because they were inaccurate or imprecise but comprehension issues abound anyways since that's all I've gotten out of you from this thread ...