Considering that metroid games are meant to be speed-ran in 3 hours or less while still being able to get 100% completion, I don't see why this briefness is a problem. If Metroid fusion has problems, it's that the directed nature of the game cuts into it's replay value.

^This is a strange contention, one that I've never actually heard articulated by anyone at Nintendo--do you have a quote to back that up?

(I'm not saying it isn't, necessarily, true, but that sounds like a modern gamer reading into Metroid's design and not something it's specifically designed to contain, since only a tiny minority of players of the series would be even remotely interested in such a 'feature' and even less given Fusion came out 12-years ago, before speed running was anything more than hyper-niche.)

Which boss battle did you see slowdown during? If it was Nightmare, then that was on purpose...

I know on the GBA you could get the game to chug a little bit at times but it wasn't very often.

I know Metroid Fusion gets knocked for not having as much freedom as other Metroid games but I still like it for its take on the Sci-Fi trope of being trapped on a Space Station with a hostile alien force. In a way, it's the most cinematic of all the Metroid games with its story happening as you play. And there is still a lot to explore as pretty much every map shown is only like half or a quarter size of the full area.

I wouldn't really say Fusion has story as you play. It's mostly reserved to save rooms and I think other Metroid titles tell story a bit better.

As for the speed running aspect, the mere fact that they give you a completion time and percentage means there's a drive to streamline your play experience with subsequent replays.

I consider Metroid Fusion to be a great example of what a sequel should be.  It isn't a Super Metroid expansion pack.  It has enough similar gameplay elements to appeal to fans of the series but it does things different enough that it stands on its own.  It doesn't replace the other Metroid games and is also unique enough that a Metroid fan would feel compelled to experience it.  Zero Mission in comparison is a little more Metroid-by-numbers.  It works entirely because there aren't that many Metroid games so two games that play like Super Metroid isn't enough to oversaturate things.  But if they continued to chug out Super Metroid clones, you would get sick of it.  A videogame franchise can ensure a longer lifespan if it takes a more Metroid Fusion approach to sequels.

Ah, but Other M is also unique but I hate that game.  Well I personally think it's a bad game and you don't get a free pass on a creative sequel if the resulting game isn't very good.  There is also a balance in keeping with the tropes of the franchise and doing things differently and I think Fusion achieved that balance and Other M did not.

Yeah, never understood the backlashing on this one. It's a bit more directed, and the boss fights are more fire-power based than other games (load missiles into that cylinder robot, stat!), but it feels like a completely legitimate tilt on the Super Metroid formula. It's good that the games area little different! Also, hunter-killer Samus in this game is genuinely scary, quite an accomplishment for a GBA game.