The point isn’t to take over the console market but to defend the PC gaming market from console exclusivity and Steam itself from e.g. Microsoft Store
I don’t know if I’m in the minority or not, but I have never owned a gaming PC, and have not had a console since I was a teenager with an Xbox 360. I’m now 30, and I bought the Steam Deck precisely because of its portability.
That was initially the case for SteamOS/Steam Machines, but the MS Store has been a _resounding_ failure in the gaming market. It's use is largely limited to PC Game Pass subscribers, who use it as the download backend for the Xbox app (which is the primary interface for installing/managing games).
If anything, it's a way for Valve to hedge against streaming platforms.
I guess the target audience is those previously gamers with an extensive collection of games to play "someday" who now have kids that would like something to game on. The deck fills this market very nicely and Switch has already paved the initial road.
Countless 1000+ hour per year players aren’t getting caught up in competitive-tier multiplayer content with high performance requirements and yet qualify as something besides “casual wannabe” gamers. The Steam Deck is a nice way to take their serious hobby down the hall or on vacation.
There are people who want to play games, but cannot or do not want to spare the expense for a gaming rig. Rising component prices have expanded this category. There is very little overlap between competitive gamers and people wanting to be gamers, some of whom are caught by rising component prices. Again, nothing surprising here.
Steam Deck seems to be designed mostly for this market: people who want to play games, have a leaning towards PC games (possibly due to preexisting Steam library), but have no requirement for mouse and keyboard form factor. Steam deck captures the market of people who "would start playing games if cost of the rig was $(cost-of-steam-deck)" AND the market of people "would play PC games if the rig was more portable".
I'm pretty sure that if some compatibility layer, allowing to use Steam libraries on xbox/PhlashSpeed and Switch, existed the Deck would be dead on arrival.
And if my gramma had wheels she would be a bicycle.
The hole point of the Steamdeck is that existing consoles are locked down and would never allow running your existing steam library.
I'm not sure that it's close enough for truly competitive play, but look into flick stick and gyro mouse. For first person games it's good enough that I've been able to hop online in a couple of older first person shooters without aim assist and get roughly (or at least close to) the same performance I would get from a mouse today. It's certainly good enough for the vast majority of single-player campaigns to the point where you wouldn't be missing much if anything by not using a mouse.
Gryo mouse isn't quite as good, it's fine for menus and point-and-click games, but falls down a bit more for stuff like RTS games. I've been trying to figure out control schemes to make it perform better for faster movements without losing precision. It's comfortable enough to use as a general mouse, but I don't think I could use it competitively in an RTS (although I do think I could probably get through a single-player campaign).
I'm mildly hopeful that those control schemes will get better in the future, but even today as someone who pretty much only ever played PC games with a mouse, it's switched me from thinking of the Steam Deck as a primarily 2D/3D/platformer device where I could maybe pick up an FPS if it had really good aim assist, into a device where I wouldn't hesitate to load up pretty much any FPS on it without aim assist. I'm not getting 100% of the performance I used to get with a mouse in older high-precision aiming games like Sauerbraten, but I'm not sure how much of that is that I'm generally out of practice using a sniper rifle and that I haven't put hundreds of hours into practicing with gyro controls. I am still getting kills in those games.
Gyro controls have a couple of issues to work out, mainly stability when you're trying to hold the cursor still, but there's potential there to get (very close to) mouse-like precision for the majority of players who aren't investing hundreds of hours into getting competitive -- especially if games start to take advantage of them natively instead of forcing players to hack them together with mouse emulation.