I'm not convinced that Metroid at least really is a great data point for "Nintendo is ruining things in-house". From Wikipedia[1]:
> Nintendo announced Metroid Prime 4 with a teaser trailer during the Nintendo Direct presentation at E3 2017, and announced that Retro Studios, who developed the previous main Prime games, would not be involved.[15][16] In February 2018, Eurogamer reported that Prime 4 was being developed by Bandai Namco Studios in Japan and Singapore.
> In January 2019, the Nintendo EPD manager Shinya Takahashi announced that development had restarted under Retro with Tanabe remaining as producer. Takahashi said the previous studio had not met Nintendo's standards and that the decision to restart was not taken lightly.[21] Shortly after, Nintendo reevaluated Prime 4 after noticing changing attitudes towards open-world games, but maintained the direction as the development was already taking longer than planned. The team ignored new developments in action and shooting games to prioritize the adventure elements.
There's a perspective where this is almost the exact opposite of the problem being discussed about Microsoft. They chose to let it get developed externally, suffered delays, and by the time they moved it back in-house, the ecosystem had moved from under them. They probably could have chosen to rethink everything and delay it further, but they also arguably could have avoided having to make that call by keeping it in-house and letting the studio who made the previous entries work on it from the start and landing it in time that the original vision still fit what people wanted.
[1]: https://en.wikipedia.org/wiki/Metroid_Prime_4:_Beyond#Develo...
MP4 is what OP was talking about, an "interactive Hollywood" experience that betrays previous Metroids, adds discutable open-world design cues, and locks features behind $30 figures.
But Star Fox? Phenomenal. Such a fun game. Luckily I have the pro controller so I could map A to the back paddle or else my poor old tendons couldn’t handle the rapid fire shooting required at the high levels, but I’ve had an absolute BLAST playing the remake.
A remake (1) of a remake (2) of a remake (3)
(1) A remake (Switch 2 Starfox, a remake of StarFox 64)
(2) StarFox 64 (A remake of Super Nintendo's StarFox)
(3) ??? I don't know what the 3rd level of remake you mention is, but I'm curious!
Star Fox's development is an incredibly wild story where British teenagers argued what the SNES could do with bespoke hardware, and they ended up being shipped out to produce it because Nintendo felt they couldn't ever do it themselves. It all started with Argonaut's demo of what would eventually be released in Japan as "X". Entirely software-based 3D, on the original Game Boy.
There's actually a very humble quote by Miyamoto where he learned that someone can't just get better as a function of age and experience, after he clearly realized that these teenagers could produce something no one else in Nintendo ever had a hope of. Perhaps it's why the franchise has done so little -- Nintendo's just not in a remotely similar headspace the Argonaut lads were.
-----
Fun videos on the subject:
"The Teenagers Who Taught Nintendo How to Make Star Fox" - People Make Games, https://www.youtube.com/watch?v=to4Ekb0kXiE
"The Making of Star Fox" - Strafefox, https://www.youtube.com/watch?v=GDhNT2Qv-Mo
I assumed that's what OP meant, all of those are the exact same game with the same story and dialog, remade in 3 different game engines.
A lot of Nintendo's remakes end up being training exercises for the real deal, such as Metroid 2 remake to Dread. Meanwhile, some of the laid off devs here might have never seen a properly produced title with zero crunch and anomalies. Not every title should be an auteur title, but we have too many auteurs and we want more auteurs.
Neither of your examples fit that description. Metroid Prime 4 wasn't chasing Hollywood cinematic design; it was a highly targeted attempt by producer Kensuke Tanabe to make a tight, isolated first-person exploration formula resonate (especially in Japan where it has consistently failed). Its goals are mechanical, not cinematic. Meanwhile, Star Fox is a classic arcade rail-shooter remake with modernized cutscenes, not a prestige movie-game. Early sales data shows it's actually working well, too, having just debuted at #1 on the physical charts in Japan and nearly doubling Star Fox Zero's launch week in the UK.
Ultimately, Nintendo operates like a Consumer Packaged Goods company. They treat their library of IPs like a diversified product portfolio rather than betting the farm on individual interactive movies. They use massive, high-margin, mechanics-first games like Tomodachi Life and Pokopia to generate enormous cash reserves. They then use those profits to subsidize legacy IPs like Metroid or Star Fox to keep core fans happy and feed their broader brand ecosystem. Because Nintendo spreads its risk across a wide spectrum of lower-budget games, they can easily absorb a minor product flop. Sony's interactive Hollywood model sinks $300M into a single basket, meaning one bad miss can completely wreck a studio.
Although Nintendo is still following the path of "gaming enshittification" with lesser budgets; and I would argue that Star Fox mostly sells because there's barely anything to play on that 500$ thing...