Lots of baffling responses in this thread. Three years ago it was somewhat understandable that people willfully ignored historical sales data and had no clue that it was already a done deal that Switch is the successor to both the 3DS and Wii U, so their stupid low lifetime sales predictions for Switch could be excused to some degree. But today? Nah, people should have learned a thing or two by now, especially those who've been around for the last few years.
Through 36 months Switch is tracking 6m ahead of the PS4. The current outlook is that Switch sales will be bigger in 2020 than they were in 2019, so the console hasn't peaked yet and therefore is going to increase its lead over the PS4 further once we've completed month 48. The PS4 is expected to sell between 120-130m lifetime, so in order for Switch to sell less than 120m, people must be expecting the infamous cliff to be approaching.
The common error I see is that people cherrypick historical sales data to arrive at the conclusion of low Switch sales (read: anything below 120m lifetime). There's no interest in learning and understanding how sales materialized, because if there was, the conclusions would be very different.
Fact 1: Price cuts prolong console sales. Switch has yet to see a price cut which is unprecedented for a console that is over three years old. It hasn't even had value-added bundles yet.
Fact 2: Revisions prolong console sales. Switch has only had the Lite model so far, but if you look at the portable consoles of the past, then it's normal that there more than two models over the course of a lifecycle.
Fact 3: Software sells hardware. The reason why Switch didn't peak early is that Nintendo's top development teams don't have to go back and forth between two consoles. This results in a constant stream of killer apps that isn't going to end anytime soon. In the past we've seen a sharp drop in newly released system sellers after year 3, especially on Nintendo home consoles. On the flipside, Nintendo handhelds could always count on new Pokémon games, so if you take that into consideration, it shouldn't be so surprising anymore that Nintendo handhelds had better long term sales.
Fact 4: Successful Nintendo consoles have a lifecycle of six years minimum before their successor launches. The only exception was the GBA due to extraordinary circumstances (Sony attempting to get a headstart over Nintendo's next generation), but something like that isn't going to happen to Switch. Skim over fact 1 to 3 again and put the pieces together: It's obvious that Nintendo is doing everything to give Switch a long lifespan because they keep holding their cards close to the chest instead of playing them early.
Fact 5: Third party support isn't slowing down for Switch. The health of the software pipeline is essential for hardware sales. It's why the PS Vita tanked in America and Europe despite "PS consoles have long lifecycles and always good sales" whereas it wasn't so bad in Japan where the system saw continued support from third parties. In any case, the point is that any analysis that is based on Nintendo vs. Sony is fundamentally stupid because what's important is the state of the software pipeline. Switch is in a great position, so any holes between first party releases are filled by third parties stepping up. Also, after three years of Switch, you should be aware of the importance of AAA third party support, or rather its lack thereof. When it's clear that isn't AAA third party software that has led to the Switch sales that are outpacing the PS4 launch-aligned, then it's also clear that the upcoming PS5 and XSX can't have any damaging influence on Switch's software pipeline.
Fact 6: Switch's technology didn't get outdated as fast as the PS4 and XB1. By the end of its third year, the PS4 already had its Pro model out; the Xbox One X followed a year later. Meanwhile, nothing comparable is on the horizon for Switch. Now you probably wonder how Switch could remain up to date despite having less processing power than consoles that launched 3.5 years earlier. It's because Switch doesn't sell itself on processing power to begin with, hence why a game like Animal Crossing can become one of the biggest blockbuster games of 2020. It's for the same reason that Nintendo doesn't need to worry about upcoming 10-12 TF consoles; being outclassed in processing power can only matter when a console manufacturer defines itself over processing power. But as it is, neither Switch owners nor prospective Switch owners put much, if any, stock in processing power, but rather enjoy what Switch excels at: Quality games anytime, anywhere.
Switch has sold close to 50m units by the end of 2019 and is on track to hit ~70m by the end of 2020. It's outpacing the PS4 and it has everything that is necessary to sell well for a long time. That's why challenging the PS2 and DS sales isn't that much of a long shot. It only is for the people who are stuck in the bubble of "PS consoles sell the best" and therefore start with the conclusion and arrange the facts to fit the conclusion, rather than looking at the facts and then forming a conclusion.