While researching SETI, I noticed that about 70% of the BGG community thinks the game is best with 3 players, while only 30% think it is best with 4 players. The poll, as always, doesn't say why people think so. But I would guess that it is a very typical reason: With more players, the game takes longer. That affects both the overall length of the game, and the time a player has to wait between turns. SETI is not a very fast game, it is estimated to take about 40 minutes per player, and for a first game with 4 players we might well need 4 hours, including rules explanation. I can see why some people would prefer a 3-hour game for 3 people to a 4-hour game for 4 people.

I do prefer board games with at least some interaction between players, and SETI has that. But even by just studying the rules, I can see that there are two major ways of player interaction: Moon landings and scanning. Moon landings bring a lot of points, but every moon can be landed on only once. The more players are in the game, the more likely it becomes that somebody else lands on a moon you wanted to go to. The number of moons is the same regardless of player count, so competition is obviously greater the higher the player count is. On the other side, scanning has a small effect when you do it, but then gives additional points once a system has been scanned a certain number of times (usually 4 to 5). Once the system is fully scanned, the person who contributed the most scans gets additional effects and points. Again, the number of systems to scan and the number of times you need to scan them are the same, regardless of player count. But with more players, scanning is more likely to fill up a system, and thus more likely to score points. It is obvious that the relative value of going for a moon landing or going scanning is different with different player counts. And it can't possibly be balanced well for all player counts. From reading between the lines on what strategy is "OP", I think that the game is balanced for 4 players, even if that takes longer to play.

It is thus possible that a game is best balanced at one player count, while the "flow" and time needed is better at another player count. Which makes the game suboptimal at any player count.