I got to play a PC in a Starfinder recently, which was nice because so far I'd only played as GM. One of the most interesting aspects of Starfinder is its game-within-a-game system of starship combat. Played on a hex grid with a unique initiative system (a lower roll is arguably better, because that means your opponent moves first, which means you get to react to their action), and entirely different rules from normal combat. Until that game, I hadn't had a chance to play starship combat, so experiencing it was quite enlightening, not only to learn the mechanics but for what it revealed about how people are playing and thinking about Starfinder's space combat system. Mostly, everyone I've seen play it out is, in my very humble opinion, playing it wrong. Here's why, and some ideas on how to mitigate it.
A starship generally requires several people to operate it, which is convenient in a game that typically has 4 player characters in it. This means, however, that when players acquire a ship, they must assume different roles as crew members (or does it? More on that question later). The roles defined in the rulebook are: captain, helm, science, engineer, and gunnery. Each role has a specific set of actions available to choose from each round.
The first problem I notice in game play is that at least one player, but quite often all players, in a group think that "captain" means the person who sits in the center of the bridge and barks orders at everyone else. That's [possibly] traditionally what a captain actually does (I have no idea, in reality), and it's certainly what captains do in sci fi shows. But Starfinder is an RPG game, and RPG games are famous for individual agency. Placing one player in the role of captain goes against everything an RPG is about. The last thing I want to do as a player is take orders from another player. I'd like to play my own character, and while I'm perfectly willing and eager to coordinate my actions with other players during combat, ultimately I want my PC choices to be my own.
Similarly, it makes no sense from a story perspective. Imagine a group of individualistic adventurers banding together as equals for some mutual advantage (wealth, exploration, a rescue mission, or whatever), only to appoint one of them as lord over all the rest. And this isn't just any ol' ship that the captain would be in charge of: this is a ship the group has earned or purchased together. I'm not even an adventurer in real life and it makes my skin crawl.
My point is, a captain in Starfinder can't be a captain in the sense of Captain Kirk or Captain Adama. And the problem with appointing a captain is that it misleads players on both sides. Because everyone's seen all the same sci fi shows, everyone has the same idea of what it means to be a captain of a starship. The player in the captain role thinks the job entails barking orders and deciding what the ship is going to do each round. The other players think they're meant to follow orders and submit to authority. It sets everyone up for disappointment: the captain gives orders that aren't welcome and may not be followed by willful players, and the crew resents being ordered around but feel it's part of the game to do as they're told.
I think that Paizo tries to mitigate this by explicitly stating in the rules that the roles can be changed every round. This means that in just 4 rounds, each player could play captain. But in practice, that would be mechanically foolish, because to be played well, each role requires a very specific set of skills (good CHA checks for the captain, computer and biology checks for the science station, and so on). So what actually ends up happening is that the players settle into a Star Trek style crew compliment and live with that for the rest of the campaign.
If the "captain" term is as problematic in your game as it is in mine, change the terminology. In my latest game as GM, I'm calling the role the "CO". The title CO refers to the Chief Officer, which also isn't accurate (because no one on Starfinder player ships are, generally, actual officers), but it's a term that most people have sort of heard before, but few are so familiar with it that it comes with any amount of expectation. If you ask a player what a captain does, they can probably tell you that the captain tells everyone else what to do. But if you ask a player what a CO does, they'll have to give it a lot of thought before hazarding a guess. That means that player's are less likely to fall into the trope trap.
A CO, in real life, coordinates the bridge crew. That's a better role for a player in a cooperative game, although the actions available to this role in Starfinder don't actually encompass that. Mechanically, the "captain" role is literally a cheerleader; the PC encourages fellow PCs and berates the opponent over comms. Through extrapolation, it can be assumed that this PC also, in the game world, helps the other PCs communicate their intentions with one another in the way the players around the table are able to do by talking. In other words, say the players decide to do a fly-by to bring the ship within close range, meaning that the player at the science station knows it's worth spending a resolve point on a weapons lock. The players are able to communicate this by talking it out, but in the game it's assumed that there's a lot happening, and that the crew members aren't just sitting around a table, so the CO must coordinate and communicate the plan among all of them to make sure that the science station isn't diverting power to shields instead of targeting weapons, and so on.
It's probably my chaotic side talking here, but I abhor everything about starships as they are generally played. I blame this entirely on the quasi-militaristic versions of space naval fleets in pretty much every sci fi show there is. Firefly is close to what a Starfinder starship ought to play like, but even in that show there's a lot of structure around how Serenity operates, partly because Mal purchased her with his own funds. As captains go, though, he's not a bad model for an RPG captain; everything he asks of his crew is but a request. He's not an authority, but he's still a leader.
It can't be assumed, in an RPG, that a starship crew are looking for leadership. I think the best example of a starship in popular media comes from what is one of the best sci fi series nobody's heard of: Lexxx. In Lexxx, the ship is sentient, the pilot is bound by what amounts to technomancy, and all crew members are basically fugitives. The only way anything gets done is through arguing, debate, and force of will. It's a true pirate crew, without all the plundering. If you haven't ever seen Lexxx, you should stop what you're doing right now and watch the mini series and subsequent TV show, and steal from it a lot.
The problem in Starfinder is that starships are nearly a necessity because without them adventurers can't get anywhere. But just because a player signed up for sci fi doesn't mean they signed up for military, or even mercantile, service. For me there are two possible fixes for this:
Players crew a starship in total anarchy, or they just hire one with a rapscallion captain already on board, and they claim no roles whatsoever.
The starship is sentient. It can be organic or it can be an AI, but the ship itself is its own captain.
During battle in both models, players choose to pitch in by choosing their actions from all roles, even if it means firing a gun twice in a round. The enemy ship gets to fire [or the enemy creature gets attacks] at least the same number of times the players do. I've never heard anyone complain that ship combat is too fast, so the more shots that get fired the better.
Starship initiative in the rulebook is a little upside-down. Normally, if you roll a high initiative you go before lower rolls. In starship combat, the lowest roll goes first. This makes some sense, because you want your opponent to tip their hand before you decide your own action.
However, it still feels backwards and inconsistent. It has to be done every round, so it also feels painfully repetitive.
The Dead Earth RPG had the same design problem, and its authors solved it with combat that happened simultaneously. Each round, each player wrote down their actions. When both were ready, the actions were revealed and resolved as they relate to one another. When an intended move was made impossible by an opponent's success, a new action was chosen in the moment. It was a dynamic and exciting and interactive combat system that almost felt like it was actually happening, and it had the same out-of-time feel that VATS has in Fallout 2 combat.
I think the same thing could be applied to starship combat. No initiative rolls, just rounds of comparing chosen actions to the opposition's actions. Since there is a set number of actions possible in starship combat, it would be trivial to quickly design a set of action cards, such that combat turns could be quickly and easily generated at the table.
Speaking of actions, I don't love playing an RPG by menu selection, and sometimes starship combat falls into that pattern. Each player has a page, in theory, of actions they can take during different phases of each round. It works out, mostly, but if it goes on too long, it can start to feel like a random playlist of the same three or four key phrases. Science and engineering, for instance, have a few actions, but few of these actions are independent. In fact, if the helm announces an upcoming move, then the engineer basically has no practical choice but to boost the shields, and if the gunner intends to fire then the science station should probably provide a bonus to targeting or weapons lock. Most, if not all, science and engineering actions may as well be requirements for helm and gunnery actions. I mean this literally. A helm and gunner can run a starship, rolling for the obvious bonuses that apply to their actions.
In fact, a helm and gunner is all that's really necessary. A group of players could easily collectively take helm and gunner actions with table discussion.
The advantage to Paizo's design is that a group that doesn't have enough players to man each station, starships still work in the game. However, even when a group has enough players to man each station, some stations are redundant.
I looked at Unearthed Arcana to see whether there was anything I might be able to adapt from 5e. Turns out ship combat in 5e is possibly even worse, with three roles being defined: the captain (who explicitly gets to decide what the ship does each round), the first mate, and the bosun. Interestingly, the available actions are binary: a ship can go full speed ahead or it can fire at will, so as far as I can tell, all but the captain role is meaningless.
There's a valuable lesson here, though: despite defining three roles, there's an extreme reduction of choice. It seems to me that fewer mechanics in ship combat can speed up the encounter while also heightening the risk, because the risk is absolute. You're not worried about details, like the health of your aft shields compared to your port shields, or the endurance of you power core. You're just worried about whether you're going to be blown up or get away. Simple, urgent, and easy.
One way to help starship combat, I have found in my limited testing, is to abstract the process from the players even more severely than 5e does.
Count your number of PCs and multiply that by 2: this is the number of rounds that will happen in combat. On their turn, each player makes 1 helm roll and 1 weapons roll. If playing with miniatures, that player controls the ship for their turn. All moves are considered successful, but weapons must hit as usual.
The GM moves after each player, using the same mechanic.
Score 1 point for the player if their attack hit, or if they did not attack at all. Score 1 point for the opponent if their attack hit, or if they did not attack at all.
Once each player has had two turns, combat is over. Tally the points. The team with the most points wins, with "winning" being defined by the team itself. If they were trying to disable a ship, then the ship is disabled. If they were trying to flee, then they fled.
In the event of a tie, both teams win.
It's simple, it's quick, but it involves everyone, provides equal footing, and is more a mini game than a separate one.
Another option are ship-to-ship starfighters. When starship combat begins, each player boards a Viper or TIE fighter style ship and rushes to defend their mothership, mano a mano. In their absence, the onboard computer (the GM) takes evasive action to preserve itself, or else a player can optionally stay aboard and pilot it. Combat is essentially normal RPG combat, with each player taking turns to fight against the monolithic enemy ship, or at your option against other attack ships. When I tried this, I used the simple 3-action economy system developed in Pathfinder Unchained and later adopted by Pathfinder 2.
In this system, combat returns wholesale to individual choices and strategy. The players are no longer confined uncomfortably in a monolith, with some players awkwardly serving as redundant token roles while other players get to do fun and meaningful stuff like flying and shooting. Each player takes their own turn, chooses their own actions either from a simplified helm-and-gunner list or else from all roles defined in the rulebook (assume an R2 unit onboard can fulfill complex requests like quick repairs and optimizing shields and so on).
However you play Starfinder, starship combat can be fun as long as you get a feel for what you like, what works for you, and what keeps the game interesting. If you haven't tried starship combat as written yet, give it a go. If you try it and find it underwhelming (or overwhelming), be creative and flexible until you discover what works best for your group.