Ignoring what I assume to be sarcastic questions, the maths of Ampere is just bad.

Switch had sold through ~102m by the end of 2021 and 2022 can be expected to achieve at least 21m, resulting in an LTD of ~123m by the end of 2022. From there we have to distribute a total of 23m across 2023 and 2024.

If we applied consistent yearly drops, we would be looking at ~15m in 2023 and ~8m in 2024. Such a drop in momentum doesn't make it feasible to end up with 158m+ lifetime.

If we used a less severe drop for 2023 and had, say ~17m, we would then get only ~6m in 2024. That would be even worse momentum than in the example above.

Going in the other direction of granting 2023 only ~13m and then 2024 ~10m would allow 2024 to end on a high enough level to possibly leg it out to 158m+ lifetime, but we'd then have to ask why 2023 would have such a sharp drop and how 2024 could slow down the severe decline by so much.

Any way you look at it, it just doesn't work out to a realistic sales curve.

I've gone through a similar line of thought when I first read that article a while back. Doesn't really add up.

PS5 and switch sales are exactly as I expected. I am disappointed with the low sales of Xbox considering there is plenty stock on amazon for over a week now, both the S and X.

Haven't Ampere publically messed up before in terms of numbers? I vaguely remember an incident where a lot of people were taking their figures as gospel only for them to be disproven later. Or am I misremembering?

You are possibly thinking of another market research firm, because when it comes to the video game market, they are all pretty bad.

One of the most baffling examples I remember concerns IDC. quickrick and a few others were hellbent on taking their Switch estimates as the truth despite their PS4 estimates having an error margin of 400k compared to an official Sony announcement that had predated IDC's report by more than a month, meaning IDC wasn't even competent enough to use an official figure instead of relying on their own guesswork which was completely redundant in light of the existence of an official figure.

My favorite failure was a prediction that the Wii would literally lose sales.  I forget the analysts but IDC is likely.  It was on a chart with yearly cumulative totals for Wii, PS3 and X360 and they had the Wii in year 5 or 6 actually losing already sold units. 

How would that work? Lol, some analysts.

Statista actually has something like that in their Switch hardware sales graphs. At one point in 2019 the Switch apparently had negative sales. That's why I don't see that site as reliable.