What? We were always joking about the double helix pattern in the PS4/Switch launch aligned sales comparison articles. Because they started at different times of the year they gained the lead at different times, but that they kept repeating that was because for the initial years Switch and PS4 basically sold on the same curve. Do you really think PS4 needed time to gain more traction? Because it was a new concept and people were burned coming off the PS3? I think both consoles, Switch and PS4 started off equally strong (and PS5 seems to follow that sales curve again).
The PS4 & Switch having the same curve was a coincidence, and at the regional level was only true in the U.S., and even then it didn't last.
The PS4 had a relatively flat sales curve in the U.S., at least for its first five years (for non-Nintendo consoles, Gen 8 was the flattest generation). Meanwhile, the Switch's baseline in 2018 was essentially flat, with that year being up as a whole from 2017 only because of a big YoY jump in holiday sales that year (deals like the then-new MK8DX bundle probably helped with that). But in 2019 the Switch experienced a decent amount of growth even prior to the Lite's release in September. Why the growth after the relative flatness in 2018 in the absence of any obvious stimulative factors? Increased attention could have been a factor. Something had to be causing sales to grow (and it wasn't games, because all its big hitters were Q4 releases). But despite that growth, it appeared to have stopped in early 2020. Then the pandemic hit.
I don't think that many were expecting it to be down, most people actually thought it could peak in 2020. Really, take a look:
This was mid 2019, and the majority was betting on 2020, the amount of VG$ is bigger than any other option combined. So I don't know how you can think that most were expecting it to be down (yes, you restricted it in parantheses to the US, which makes no sense because how would people expect a 2020 peak year without the US, seeing how europe is Nintendos weak spot and Japan was very fast to adopt the Switch and probably is first to peak).
Also we can compare the effect of the pandemic between systems, and see that it does about 10%-15%. So that is not nothing and it did slow down the decline of PS4 and Xbox One, but to explain Switch's more than 40% growth you need much more than the pandemic. It is more like the pandemic slightly supported an already existing momentum.
I also don't see your prediction of sales suddenly collapsing because people can go back to museums and bars coming true. It seems not unthinkable, that 2021 is actually Switch's peak year (and nearly everyone will lose the above-linked bet). You need to look at the games, not extrinsic factors to see the momentum. Like Salnax did.
That's global sales. And you are right that most weren't predicting downward sales globally. That was an error on my part (I was thinking of U.S. sales at the moment). Globally, the Switch sold something like 19M units globally in 2019 IIRC (actual sell-through, not shipped). The average prediction for 2020 here at VGC was I think around 20-21M. And the Switch could very well have been up slightly in 2020 globally.
For U.S. sales, the one prediction thread I've seen (over at Era) had an average prediction that was a bit less than what the Switch sold in 2019. I'd say the range of predictions there was pretty reasonable. At the rate it was going, the Switch's U.S. baseline was likely to be down, offset somewhat by AC giving it a bump in March (with possibly some residual effect in April), and had the pandemic not been a thing, March would have been a much smaller month (though still quite big; I think it still could've been a 700k month). The Switch was likely going to be about flat or slightly down in the U.S., at best maybe only slightly up.
Any potential YoY drops in the U.S. could have been offset by Japan, however. The Lite was a bigger deal there, causing much higher growth. Sales for the NPD-equivalent Sept. 2019 period were over triple what they were in 2018, compared to a relatively much smaller +66.8% in the U.S.; weekly average sales had also increased more than three times what they were in August in Japan, but by only 43.4% in the U.S. This translated into larger YoY gains in Q4 in Japan than in the U.S., and January & February were also indicating better momentum going into 2020, with actual YoY gains in both months instead of declines like in the U.S., so the 2020 baseline still might have been up at least a little bit even if 2020 was a normal year. And with AC still being a bigger deal in Japan (if AC's sales are anything more than 33% digital in Japan, then Japan represents probably around a third of global AC sales despite representing maybe a quarter of global Switch HW sales), March would have still been pretty huge. I think the Switch could have still crossed the 5M mark in 2020 in Japan, though I think it would have still fell well short of the 6M it actually sold in 2020.
Europe is a harder call, but if VGC figures are any indication the Switch was likely on track to track more similarly to the U.S. in terms of YoY changes than to Japan.
Also, I do think that once things go back to normal that demand for consoles will start to normalize as well. As 2020 sales were a story of artificially increased demand, it stands to reason that, when the external factors that drove said demand cease to be factors, demand will decline as spending habits return to where they were pre-pandemic. The Switch does not exist in a vacuum.
And while a good games library is important to the health of a system, there doesn't appear to be any evidence that the games lineup in a particular year affects the baseline sales of that year. If strong lineups year after year caused growth, then the PS4 should have continued to grow instead of exhibiting the flatness it did. Now, certain individual titles can cause short-term growth (usually for a month or so), and that can affect the overall balance of sales for the year, but that requires them to be significant system-sellers, and even then they can't cause long-term growth by themselves. Even with a better overall lineup in 2018 that included several big system-sellers in 2018 (Spider-Man, GoW, & RDR2 in the U.S., and MH World in Japan), the PS4 was still down from 2017. 2018 was, perhaps not coincidentally, the PS4's fifth year, just as 2021 is the Switch's fifth year. And the Switch's 2021 is unlikely to have anything as huge as Animal Crossing was. Unless Nintendo has some unannounced megaton release like Mario Kart 9, the only thing that could come anywhere close to that is maybe BotW2, assuming it releases this year.