I never said it was the end all, be all. I've even made the point that it wasn't in the past. No two systems have identical sales curves, or even similar curves (in the geometric sense of "same shape, different size"). For example, I have noted on more than one occasion that professional analysts were wrong to compare early PS3 sales to GameCube sales and imply the former might only sell as well as the latter. But we do see plenty of obvious patterns in historical sales data regarding how various factors impact hardware sales, and that is important.

Also, you have still yet to actually provide evidence that AC has been the primary driver of Switch sales in the U.S. over past four NPD sales periods. And you have still yet to provide any evidence of some new precedent set by either the Switch or AC. Until I see your evidence, I'm not going to spend any more time on this subject. Otherwise we're just going to be arguing in circles and repeating ourselves.

Again, your assertion that NSMB was driving sales not only has zero evidence to support it, it's outright contradicted by the data. Even the most cursory glance at the data proves this. We can clearly see that it provided only a brief boost in Japan, with no additional improvements to a baseline that has. And in the U.S., DS sales in May 2006 saw absolutely no improvement over April. It wasn't until the DS Lite came out that we saw a spike in DS sales. The sales data is right there to see, and what it tells us is clear: NSMB did not produce any kind of long-term surge in DS sales in Japan, and provided no immediate boost to DS sales in the U.S.

A game being among the best-selling titles on a system, or even being the best-selling title on a system, doesn't make it a system-seller in its own right. Not every popular game causes an appreciable improvement to the amount of hardware being sold. If Japan is any indication, most games that do cause a bump in HW sales do so very briefly, and by an amount that wouldn't even show up in a situation where sales are tracked on a monthly basis. And simply asserting that a game caused huge improvements in HW sales doesn't make it so. You need real evidence.

I'm not going spend time having a back-and-forth over NSMB as a supposed system-seller because it is obviously contradicted by the data. My reply to you Saturday provided more than sufficient evidence to debunk your point. If you can provide another example, preferably one supported by actual data, I'd be willing to discuss that, however.

Of course it's a mixture. I did state that major system-selling software does sometimes have a residual effect extending into the next month. But I believe that the available evidence suggests that the "COVID bump" has been the primary driver for the improved sales from April onward.

The Animal Crossing point is very easy to prove. Nintendo's year-end is March 30th and most of the virus scare didn't happen until mid-way through the end of the month. Stimulus checks weren't going out until later (see this article from the end of March:https://www.cnn.com/2020/03/27/politics/stimulus-payments/index.html). So this period is a good measure of how well the Switch was doing before the COVID impact. From Nintendo's earnings release;

Looking at software, Pokémon Sword and Pokémon Shield became big hits, posting sales of 17.37 million units, and Animal Crossing: New Horizons, released in March, sold 11.77 million units, which is now the best start ever for a Nintendo Switch title

So in about 10 days, Animal Crossing became one of the best selling Switch games. Since then, Animal Crossing has sold 22 million while Pokemon has sold 18 million. If the pandemic was the only thing pushing sales, why didn't Pokemon get a similar bump? In fact, Pokemon  has sold less than Animal Crossing despite being out 3 months prior. If it's the virus and not the software, we shouldn't be seeing one game sell gangbusters above the rest.  This is the problem with your assessment is you ignore the obvious: that people need something to play if they are going to spend hundreds of dollars on plastic.

The NSMB point is also dumb. New Super Mario Bros sold 30 million units which is about one-fifth the entire install base. Are you going to say that weren't people who went and bought the DS or DS Lite just to play NSMB? Your comment doesn't even pass the smell test. 

I'm sure you'll come back and say the data doesn't show it or something, in which case I'd say your looking at the data wrong (the obvious issue being you look at one point in time and not in totality. NSMB released early on. Obviously it was pushing units throughout its lifespan). I pthink you're not asking yourself why people buy games. As a result, you're missing the forest for the trees, 

The PS4 & XBO are older systems with less general demand and fewer systems in the supply channel. The XBO was obviously completely sold out early enough in the month to where only 17k units got sold. The PS4 was down YoY in June & July, as was the XBO in June, but not by much, and certainly far less than most previous systems (including the PS3 & 360) were in the months prior to being replaced. By all accounts, the PS4 really should be doing worse than it currently is, and the residual effects of the COVID bump are likely what was keeping it at less than -20% YoY as opposed to a more typical -40% or so. Simply being down YoY doesn't necessarily by itself mean the COVID bump was over. By how much is important as well, especially given the circumstances. For example, here's the YoY changes in monthly sales for the PS4 & XBO so far this year versus those for the PS3 & 360 in 2013 in the U.S.:

There are other examples as well, including for Nintendo systems, but they almost all show much worse YoY drops by this point this close to replacement. The only one to really avoid this fate was the PS2, which had phenomenal legs closing in on and past replacement, only really starting to stall after the PS3 was reduced to $400.

That the month-to-month decline after April was overall slower for the Switch does not necessarily mean that AC was the primary driver of Switch sales for the April-July period. Again, other more plausible factors already exist that we have good evidence for, as outlined in the bullet points in my post from Saturday. Had there never been a pandemic this year and we still saw the Switch experience five straight months of sales improvements from AC, then there would be no question that it was the biggest system-seller of at least the past 20 years. But that's not the 2020 we live in, and I have to assess all these other factors.

As for the rest of the world, Japan shows that from April through July the Switch was averaging about where it was before AC was released. For the NPD-equivalent sales period of April-July, it averaged about 76k per week, comparable to October 2019 and Jan. & Feb. 2020, the only other NPD-equivalent sales periods after the Lite's release month but before AC. There was an improvement in July, however (also August so far, but Obon is a factor there), but in general the Switch has done far worse in Japan than in the U.S. for the April-July period.

While the Switch experienced a relatively gradual drop from month to month in the U.S., in Japan it dropped down to a baseline of sorts very quickly. Japan, unlike the U.S., didn't show any kind of "COVID bump" effect in its sales data over the spring (the PS4's only big boost to sales in Japan this year came from FF7 Remake). Here's how the U.S. & Japan stack up over the past 19 months (note that the NPD-equivalent sales periods for Japan are offset by one day, as NPD months end on Saturdays while the sales weeks for Japanese trackers end on Sundays; it shouldn't make any real difference, though):

As for Europe, do we really even have reliable monthly/weekly sales figures for region? Supposedly they exist, and apparently some people have access to GfK/Chart Track HW sales figures, but I haven't seen any of it posted online.

I find the idea that ACNH needs to prove its system selling power to be funny since it's quite evident.
Anyway:

I find the idea that ACNH needs to prove its system selling power to be funny since it's quite evident.
Anyway:

Chart time!

Hardware continues its downward slide from the heights of March & especially April. The COVID bump did amazing things for HW sales, but now we're back to something approaching normalcy. I say "approaching" because while the PS4 & XBO were already back to being down YoY in June, the XBO hardly sold anything in July (can't sell what's not available to sell), and the Switch is still experiencing residual gains, though those are steadily declining. I suspect the Switch return to being down YoY no later than September. How much it'll be down remains to be seen, but with Switch already being up over 1.7M units YTD in 2020 from 2019, it's highly unlikely for its 2020 as a whole to be down from 2019. I think it'll end the year somewhere in the 6.5-7M range, making it one of the best years for a system ever, definitely in the Top 10 if we exclude pure handhelds. This is assuming that A) Nintendo doesn't have some major software release and/or a price cut of some kind, either permanent or temporary, and B) there isn't another round of stimulus checks by year's end. Either of those could potentially push the Switch to 7.5M or more.

More charts to come later.