Why do you think this happens when these developers already had a winning formula?

I mean, all series are going to have some point where they dick things up, else we'd have never-ending amazing video game series. I don't think that the second game in the series is uniquely bad.

Some of it is just going to be luck. Like, hitting just the right combination of employees, market timing, consumer interest, design decisions, scoping a game's development time and so forth isn't a perfectly-understood science. Making the best game of the year probably means that a studio can make a good game, but that's not the same thing as being able to consistently make the game of the year, year after year.

Some of it is novelty. I mean, part of most outstanding games is that they're doing at least something that hasn't been done before, and doing so again

especially if other studios are trying to copy and build on the winning formula as well

may not be enough.

Some of it is that most resources don't always make a game better. I know that at least some past series have failed when a studio made a good game, (understandably) get more resources for the next game in the series, but then try to expand their scope and don't do well at that new scope.

Engine rewrites are technically-risky, can get scope wrong, and a number of games that have really badly failed have happened because a studio tries to rebuild everything from the ground up rather than to do an incremental improvement.

You mention Cities: Skylines 2, and I think that "more resources don't always help", "luck", and "engine rewrite" were all factors. When I play a city-builder, I really don't care all that much about graphics; I've played and enjoyed some city-builders with really unimpressive graphics, like the original lincity. CS2 got a lot of budget and had a dev team that tried to use a lot of resources on graphics (which I think was already not a good idea, and not just due to my own preferences; reading player comments on things like Steam, what players were upset about were that they wanted more-interesting gameplay mechanics, not fancier graphics). Basically, trying to make the world's prettiest city-builder with the money maybe wasn't a good idea. Then they made some big internal technical shifts that involved some bad bets on how well some technology that they wanted to use for those graphics would work, and found that they'd dug themselves deeply into a hole.

Sometimes it's a game trying to shift genres. To use the Fallout series as an example of both doing this what I'd call successfully and unsuccessfully, the Fallout series were originally isometric real-time-until-combat-then-turn-based games. With Fallout 3, Bethesda took the game to be a pausable 3D first-person-shooter series. That requires a whole lot of software and mechanics changes. That was, I think, successful

while the Wasteland series that the original Fallout games were based on continued the isometric turn-based model successfully, Fallout 3 became a really big hit. On the other hand, Fallout 76 was an attempt to take the series to be a live-action multiplayer game. That wasn't the only problem

the game shipped in an extremely buggy state, after the team underestimated the technical challenges in taking their single-player game multiplayer. But some of it was just that the genre change took away some of what was nice about about the earlier games

lots of plot and story and scripted content and a world that the player was the center of and could change and an immersive environment that didn't have other players acting out of character. The audience who loves a game in one genre isn't necessarily a great fit for another genre. In that situation, it's not so much that the developers don't have a winning formula as that they've decided to toss their formula out and try to write a new one that's as successful.