Upon further investigation, it appears that no Japanese sales tracker accounts for sales from the My Nintendo Store, which has been a thing since 2016. Still, even before that existed, sell-through figures have long been less than shipped figures.

For example, according to both Media Create and Famitsu, at the end of 2013 right at the end of its big boom years the 3DS has sold a bit under 14.7M LTD. Nintendo's shipped figure was 15.76M, so there were about 1.1M units unsold units in Japan alone. Combined U.S.+Canada sales at the end of 2013 was at 12.4M (11.49M U.S. according to NPD, 920k Canada according to CESA). Assuming that they make up 95% of the "Americas" region (and it's probably more than that), that gives us ~13M for the whole region. Total shipments to the region at that time? 14.36M. That means probably 1.4M were unsold. Add in Japan, that brings to the total to about 2.5M. Add in Europe & RoW, and that figure probably rises to ~3.5M.

Either all the trackers in the world are significantly undertracking everything and even Sony & Nintendo are wrong about their own sell-through figures, in which case reported sell-through is always less than actual sell-through regardless of the region or who's doing the tracking, or there's always a lot of unsold hardware leftover at the end of the year.

I'm gonna need a source on that claim regarding Nintendo's sell-through. They seem plausible enough. We know that in 2020 the Switch sold 9M in the U.S., likely another ~1M in Canada given the usual U.S./Canada split, and 6M in Japan (using the NPD-equivalent 52-week sales period, not Famitsu's 2020 period). That's 16M right there. While there is often a decent margin of error associate with VGC's sales figures, their 7.4M figure probably isn't grossly off the mark, so that brings the total for the three main regions up to about 23.4M. Once we include RoW, that ~26M figure is probably pretty close.

As for catching the DS, the Switch's surplus over it in the U.S. was because the DS had a slower start than the Switch, which was posting good sales right out the gate. But that surplus started dwindling rapidly in 2019, and that decline was only temporarily halted by last year's gains, but picked back up again over the holidays, and now the Switch is running a deficit, one that's likely to keep growing throughout the year, considering the DS's fifth year (2009) was its peak year, selling a record 11.2M units. The Switch's deficit in the U.S. will likely be well over two million by the end of the year. And to further make things difficult, the Switch will have to sell another 8.56M or more next year just to keep the deficit from growing even further. And this is excluding the 1.23M the DS sold in its 2004 launch holiday. I really doubt the Switch will pass 50M in the U.S., though it may come close, but the odds of it matching or beating the DS are rapidly approaching nil.

Japan is possible, but unlikely. The Switch has recently pulled ahead of the 3DS in Japan in LTD sales. The 3DS has sold 24.56M to date. The DS however sold over 33M. It's going to be closer in Japan than anywhere else, and 30M is definitely possible, but actually narrowing that massive deficit with the DS is an uphill battle.

As for Europe, forget about it. That's the biggest deficit of all. The Switch isn't even keeping pace with the PS4 in Europe, much less the DS. If the Switch gets to 50M in North America (45M U.S. + 5M Canada), then assuming their current ratio is about the same the Switch will probably looking at 30M in Europe, 35M tops.

That's 115M, give or take, across the three main regions. Add in another 15-20 for RoW, and you get 130-135M. Unless there's some massive boom outside NA, Europe, and Japan and the Switch pulls off record-shattering RoW sales, I really don't see any scenario where the Switch matches or beats the DS. I think it'll land at a still-impressive third place ranking among the list of all-time best-selling consoles, behind the DS and PS2.

@Bolded. I don't believe I did. In fact, I pointed out that it had a better start than the DS in the U.S., but started losing its LTD lead in 2019, the gains of 2020 only partially slowed down the losses, and that the Switch is now selling more slowly than the DS.

As for the Game Boy, its life cycle was an anomaly. Nothing ever persisted as long as it did. By time the Color came out, the Game Boy was over 9 years old. Because of its age, complete sales data for the Game Boy is hard to come across, but what data there is indicates that in the mid 90s the Game Boy was already in a period of decline, barely maintaining relatively low-level sales. When the Color came out, the market responded to it as if it were a next-gen system rather than the mere spec upgrade it was. And the Color wasn't even out three years before the GBA was released, and when the GBA was released support for the Color quickly dried up.

If you want to think the Switch is going to have a 12-year primary life cycle including all revisions, be my guest. But every system is replaced at some point. Except for the Game Boy, nothing has lasted longer than 8 years before being replaced by a next-gen system, and only two (the 360 & Famicom) lasted longer than 7. Even systems that were doing very well had replacements announced and released when they were doing well. Nobody simply leaves a system on the market for ages anymore without planning on replacing it in 6-8 years. Nintendo is probably already drawing up plans for the Switch's replacement as we speak. The Switch isn't going to last until 2027 or later. It certainly isn't going to sell at a high level forever. As with every other system, most people that want one will eventually have one, sales will decline, and Nintendo will replace it to restart the cycle.

Prior to the pandemic there were already indicators that the Switch was not going to be experiencing further growth in 2020, or at least not anything significant. In the U.S., for example, its YoY gains in quarterly sales that began in Q4 2018 were trending downward, with a slight reversal in Q3 2019 due to the Lite. When a system experiences a multi-quarter period of growth like that that's trending downward, it tends to continue declining and then flip over into YoY loss territory, and once it enters loss territory it almost never has another period of sustained growth, at least not for a period more than 3 or 4 months (the only exceptions were the 360 & PS3, both of which saw recoveries early in life from initial mediocre-to-poor sales after some price cuts, and then another boost later on when their slimline models were released). This is exactly what we had initially started to see with the Switch in 2020. The Jan.+Feb. period was down 6.6% from 2019. Not much, and that impact on Q1 as a whole would have been negated by March because of AC (had there been no pandemic, I still think the Switch could have pulled off at least 500k or more in March 2020, putting Q1 up at least 7% YoY), but the general trend was likely to stick for the remainder of the year, with 2020 being at best roughly flat and probably at least slightly down (the average of all predictions at ResetEra was 6383k, not quite 2% down from 2020).

A system going from declining gains trending towards loss territory to suddenly changing course and posting record-setting gains for a system that's already doing well is something that's never happened before. It was clearly anomalous behavior. And anything that's going to explain the sudden surge also has to explain the size of the growth and have a plausible mechanism.

The growth experienced by the Switch in 2020 was massive. Even ignoring the initial spring spike, the Switch had strong sales for the remainder of the year. The May-October period saw a net total increase of 76.8% over the same period in 2019 in the U.S. In Japan, the April-Aug. period was up 87.4%. Never before has a system that was already selling strong posted year-over-year improvements of that size. Like, not even close. For example, after its first price cut the PS2 experienced only 42.7% YoY growth for the May 2002 to March 2003 period in the U.S., and that's with the big initial spike in May & June. If we limit it to the July '02 to Mar. '03 period, the growth was only 36.7%, though if we further limit it to the July-Oct. '02 period the growth improves to 45%, still far less than what we saw with the Switch. The only other times we saw percentage gains consistently over 50% on average over a period of that size or longer was when a system started off selling poorly and then started surging later on, e.g., the DS after the Lite was released, the PS3 after it got cut to $400, the PS1 over its first couple of years in general after two big price cuts and general increase in attention towards the system.

As for what can cause growth of that size for such a long period of time, nearly every period of growth lasting several months or more has been attributable to either price cuts or hardware revisions. But those can be ruled out since the Switch has yet to get a price cut, and the Switch Lite is unlikely to be the culprit because of what we saw from its initial effects and what we've seen from other hardware revisions in the past. In both the U.S. and Japan, the Switch's sales were on average substantially better after the start of the pandemic than they were in October 2019 or Jan. & Feb. 2020 (the non-holiday months after the Lite's launch month but before the pandemic), both in absolute terms and in terms of YoY gains. We've never seen a hardware revision have a delayed impact like that. While there have been a couple of instances of a system experiencing sales growth without a price cut or new hardware revision to explain it, such periods of growth were modest percentage-wise, meaning stock improvements or general increase in interest from casuals aren't going to cut it as an explanation for the Switch.

If we can rule out all the "normal" factors that cause sales growth, then clearly some abnormal factor is responsible. The only satisfactory explanation is the the pandemic, which drove increased demand for at-home entertainment, which was in turn further boosted by stimulus money. Yet this remains controversial, and I think that's because some people feel that attributing the vast majority of its growth over the past year to an external factor somehow diminishes the Switch's success because it's perceived as implying that it didn't fully "earn" its sales on its own merit. But the facts are what they are, and any hypothesis to explain the data has to fit with established patterns and have plausible mechanisms. It sold what it sold, and something caused it to sell that much. The preponderance of evidence indicates the pandemic was the primary driver of the massive increases seen in the past year, just as it's the only plausible explanation for the PS4 & XBO having better than expected performances (at least until stock ran out thanks to the surge in demand). I don't believe the Switch would have had an 8M+ year in the U.S. in 2020 without the pandemic.

Saying that COVID is continuing to give Switch a boost is non-sensical. U.S is not in lockdown anymore, it's just that Switch has so much momentum going on at the moment.
If COVID was still playing such a huge role then why didnt Switch smash Holiday records? I see Shadow likes to use March-Oct gains to contribute to COVID only but then conveniently leaves out the holiday period...

Here is a fun fact. 1st Stimulus payments came between March and October (many delays) . 2nd stimulus payments came after December. 3rd stimulus payments are coming in now. The only large gap in stimulus payments is during the holiday period.

So now is COVID the factor is stimulus?

Also last years first stimulus wasnt until April, not March. So that means March is all due to Animal Crossing, right? Since we're now saying its due to stimulus.

And even though some were delayed (not just til October, some were delayed until December but you conveniently left this off) the majority received the stimulus in April and May meaning they werent really a factor after that.

That's what I said, more or less. I think March 2024 is the most likely release date for the Switch's successor. IcaroRibeiro is the one wanting to compare the Switch to the Game Boy and its totally unique and bizarre life cycle.

If the primary reason is not the pandemic, then the timing is damn peculiar, as is the degree of growth.

The Switch was essentially flat from 2018 in the June-August period of 2019, got a decent bump from the Lite in September (not nearly as big as it got in Japan, though), with some residual effect that put October up about 21%. But November was barely up even with Pokemon, though December was up about 15% (maybe that Cyber Monday bundle, or maybe stock issues that weren't resolved until December?). Then January & February 2020 were down YoY. Whatever caused the modest boost early in 2019 was done by June, the Lite's effect was short-lived, Pokemon didn't seem to do much, and the system was down at the start of 2020. The PS4 & XBO were already slowing down rapidly in 2019.

The PS4 was down 27.4% YoY in 2019, while the XBO was down 29.7%. In the January-February period of 2020, they were down even more, 45.7% and 42.1%. Normal enough, considering the 360 & PS3 were down 27% and 24% respectively in 2012, the year before they were replaced. The 360 & PS3 were also down considerably through most of 2013 (-42.5% and -36.5% respectively for the year as a whole, -33.5% and 28.3% for the non-holiday period, and that's with Sept. & Oct. getting boosts from GTAV and the 360 getting holiday bundles). Again, quite normal considering it was getting closer to the releases of the PS4 & XBO.

So, no real indications that the Switch was going to experience any appreciable long-term growth any time soon, and relatively normal declines for the PS4 & XBO.

Then March happened.

The Switch experienced growth in weekly average sales of over 140% from February, making that March the second-best non-holiday month ever at the time for a system in the U.S., even in terms of weekly averages (March is a 5-week period), bested only by the DS's performance in April 2009. The PS4 & XBO both experienced relatively huge increases from February as well, with combined weekly average sales of the two being up nearly 75% from February. Even though both systems were down YoY by over 40% in the previous two months, they were both up 25% in March.

That's was... not normal. March sales in the U.S. are, at least in terms of weekly averages, almost always down from February, and on nearly ever case where March's weekly average was better than February's, it wasn't by much, maybe a few percent. The only other times prior to 2020 we've seen March be significantly up from February was with the N64 in 1997 (price cut issued that month), the PS2 in 2001 (restocks following supply issues), the GBA in 2003 (the releases of both the SP model and Pokemon Ruby & Sapphire), and the Wii in 2008 (the release of Smash Bros. Brawl).

With the Switch, at least it had the excuse of Animal Crossing, but even then that was an unusually large jump. The best month-to-month increase (excluding October-to-November), in terms of both percentage and absolute numbers, that can be attributed solely to a single game was the PS4 in Sept. 2014 when Destiny caused weekly average PS4 sales to increase 126.5%, or from an average 47.5k/week to 107.6k/week, an increase of 60k/week (that month was the best non-holiday month of the generation until March 2020, and was also the best one since Feb. 2011). The Switch completely shattered that record. The aforementioned >140% increase from February translated to an increase in weekly average sales of least 117k/week over February. That was odd considering Animal Crossing was never huge in the U.S. at that point. New Leaf sold decently (505k copies) and boosted the 3DS to 45k/week in June 2013, up from 28.5k/week in May. But that's a far cry from what New Horizons did (definitely well over 2M copies sold in March, maybe 3M or more, and likely a vastly bigger impact on hardware). Could Animal Crossing's popularity have surged that much all on its lonesome? Maybe. That can't be ruled out. But I still doubt that all of that growth we saw with the Switch in March was attributable to Animal Crossing. A lot of it? Yes. A majority of it? Possibly. All of it? No chance.

The PS4 & XBO had nothing going on to boost sales that month. No games big enough to drive growth of that scale. No price cuts. No nothing. So what's the deal? Why did every system experience a big jump in March 2020?

The answer is obvious: the pandemic was declared in March. Everything started shutting down. Movie theaters. Sports venues. Bowling alleys. People's options for entertainment were suddenly limited to things they could do at home... like video games. The pandemic perfectly explains the surge in demand for consoles. Without the pandemic, the PS4 & XBO would have had a March that was likely even worse than February. It stands to reason that if the PS4 & XBO were getting huge boosts from the pandemic, the Switch must be benefiting from it as well. There's absolutely no reason to think it was uniquely insulated. And comparing the Switch to them, it's possible that at least half of the boost in Switch sales may have been attributable solely to the effects of the pandemic. The Switch probably would have sold over 700k in March regardless, if AC's surge in popularity was wholly organic and not also itself largely attributable to the pandemic (something impossible to determine one way or the other), so it still would have likely been a very good month without the pandemic, just not a record-setting one.

And this is just March, before any stimulus checks went out. Those first started coming in April, which saw an even larger surge in PS4 & XBO sales, and the Switch maintaining its weekly average instead of declining like what one would expect for the month after the release of a big system-seller. May saw relatively soft drops, and demand was still elevated. Even the PS4 & XBO were still up a bit YoY. While the XBO suffered from depleted stock probably in late June and never fully recovered, the PS4 had enough stock to last it through September.

The PS4 sold about 863,000 units in the U.S. in the May-September period of 2020, the second most about of units sold of any console in the same period in its replacement year after the PS2. This is a decline of only 8.5% for the May-September period of 2019, where the PS4 sold 943,000 units. This is the lowest decline in baseline sales of any system ever in its replacement year in the U.S. For reference, the Xbox 360 sold 626,000 units in the May-Sept. period of 2013 compared to 1,083,000 in the same period of 2012, a decline of 42.2%. Under normal circumstances, throughout all of 2020 the PS4 likely would have been down from 2019 on average by at least 30-40% given what the 360 & PS3 were doing in 2013 and where it was at itself at the start of 2020. Even the PS2, which declined slower than pretty much every other system ever, declined 19.1% in the May-Sept. period of 2006 (its replacement year) compared to the same period in 2005, far more than what we saw with the PS4.

Between the two of them, the PS4 & XBO managed to sell about 1,282,000 units in the May-September period of 2020 versus 1,528,000 units in the same period of 2019, a decline of 16.1%. Compare this to the 360 & PS3, which sold 1,186,600 units in the May-Sept. 2013 period versus 1,885,000 units in the same period in 2013, a decline of 37%. Combined PS4 & XBO sales managed considerably smaller drops than combined 360 & PS3 sales and even outperformed them in their respective replacement years despite lagging behind the prior year, and that's with the XBO's stock issues and the PS3 and (to a lesser extent) the 360 both getting a solid bump from Grand Theft Auto V in September 2013. This indicates higher demand for the PS4 & XBO than what one would have normally expected, which appears to confirm that the "COVID bump" was not limited to just March & April. That they didn't do even better was because they were old systems on the verge of replacement, well past peak demand and with insufficient stock to maintain better-than-expected sales for the entirety of 2020.

And if the PS4 & XBO over-performed in the wake of pandemic, then the Switch must have done so as well. The Switch's massive sales even after the initial March-April spike are not simply a case of "post hoc, ergo propter hoc" in relation to the pandemic. There is a clear cause-and-effect relationship, as the rest of the console market showed. The Switch was already down YoY at the start of 2020, started blowing up at the same time the PS4 & XBO, and there was clearly elevated demand for consoles in general throughout the year.

Going by its initial level of sales, had 2020 followed a more normal trajectory following that sort of start, then even if AC got a few hundred extra hardware units pushed through the doors, the Switch might have been merely flat YoY, give or take a few percentage points... like the average predictions made at the start of 2020 were assuming. There's no reason to assume that the Switch would have experienced significant growth without the pandemic "just because." That kind of growth needs a clear cause, something that could generate significant increases in demand, and every usual cause of growth has been ruled out, leaving just the pandemic. The timing fits. The impacts on its spending in general are well documented. The impact of large across-the-board windfalls for taxpayers is well documented. While AC was a definite factor for March and maybe even also April, almost all of the gains in 2020 were due to the COVID bump. It is inescapable. The only reason to deny it is if one feels the need to somehow prove the Switch fully earned those sales primarily on its own merit, with outside factors playing an at best limited role. But if you feel there's good reason to believe the Switch would have still experienced massive growth without the pandemic, that's up to you to prove that demand can grow that much absent factors like a price cut or new hardware revision (and the Lite has already been ruled out, as I've explained before).

And sorry, but I don't have to accept that the Switch is some unique phenomenon that the rules don't apply to.

So we agree that Animal Crossing's impact cannot be ruled out as a factor. Again, I never said the pandemic wasn't a factor, just that it was not the sole or majority factor.

New Horizons massively expanded the system's demographic reach by appealing to non-gamers in a way that nothing on the system did so strongly and compellingly up to that point. Suddenly millions of people who had no interest in the system had a strong reason to buy one. It's the kind of killer app that rolls around only once or twice in even a very successful system's life.

As for past AC games, no Zelda before BOTW crossed 20 million either; unprecedented things do happen.

And you are of course free to continue to view Switch only through the lens of past system's patterns, but that why your predictions on it have been off base, you're measuring an apple by the standard of oranges.