I've talked about creating science fiction tabletop games, using the example of Star Trek. One of the things that comes up a lot whenever you do this is how you can create enough MEAT.
Fantasy games have grown a kind of plot progression out of long practice. Fantasy stories you read in books don't have the same pacing at all: a fantasy game features a huge amount of content that would give an editor fits. Dozens of pointless fights, considering which sword to buy, exploring each dungeon room one by one... this is the meat of a fantasy game.
If you've ever tried to run a science fiction game, you probably noticed that it was pretty hard to get enough meat. The plot either felt full of filler, or moved too fast. That's because you were probably using the same structure as a fantasy game. Science fiction plot lines can't really support that - there's no monsters to slay, no real advantage to exploring every room.
Science fiction plot arcs require a different approach. And we haven't really figured it out, because we haven't spent forty years perfecting our craft.
But I do have a guide that might help. Here is my guide for creating science fiction arcs that last 1-3 sessions - "Star Trek" plots that play nicely and have meat.
The players always need to be under pressure in a science fiction plot. Players used to a fantasy plot know enough to move the plot forward even with relatively little pressure, but nobody is that used to science fiction plots yet. The noose is a crisp and looming deadline to force them forward.
Nooses should have a concrete time span and failing to deal with them is the lose state. Players might be able to push nooses back, but they can't ever reduce the noose's effect. The noose is all-or-nothing, high-stakes. The situation may also get worse as the noose approaches, but the noose is specifically the looming deadline.
Noose examples include: the disease will kill everyone on the planet in 2 days. The Klingons will declare war in a month if you don't find their lost ship. The vote happens tomorrow and the majority of the senators are planning to vote "no". The space wedgie will cause life support to fail in six hours.
My planned gameplay revolves around people, and therefore the NPCs need to be the terrain of the game - half like a dungeon map, half like a monster fight. To make this easy for the GM to plan out without needing a ton of advance work, I recommend "constellations" of NPCs - groups of NPCs that are related to each other around a specific topic.
For example, the "impending new order" constellation would feature characters from the old guard and the new order and some conflicted bystanders, and their relationships would be clear. You could leave most of the characters in the wings until you need them, not bothering to create any details until you need them to play a role.
The idea is that any given NPC would belong to two or more of these constellations, and their motivations would intertwine. This adds some complexity to the things the players do, and create some nice terrain for them to be wary of. It also creates a lot of hooks that are easy to turn into complications, twists, and major NPCs.
Normally, solving the noose is pretty straightforward. If there's a disease, you need to research the cure. If there's a lost ship, you need to track the warp trail. If there's a vote, you need to convince people to vote your way.
Complications are what make the situation a lot more difficult - they are reason that the actions that should solve the situation either can't be performed or produce a bad result.
Complications usually give the noose emotional resonance, and should generally be themed to give a strong, unified feeling.
For example, the disease turns out to be a nasty engineered bioweapon, which makes curing it much more difficult. The next complication might be that the weapon was created by the locals, and they don't want that revealed and certainly don't want to help you fix it. The next complication might be that the creators died of the disease and can't help. These complications are loosely themed - they're about plans going awry, smaller evils blooming into larger ones.
Other Examples: The Klingon ship was kidnapped by a space godling. The senators you need to convince are stranded on an ice planet. The space wedgie is making time judder backwards and forwards randomly.
A twist simply changes the lose state. It either replaces the noose with a new noose, or it reverses the noose so that your efforts have been working against you, or it makes everything you've done feel very different now that it's over.
Examples: the vote needs to fail to keep their society stable. The "missing" Klingon ship is an ambush. The space wedgie was trying to protect you from a warp core breech.
Twists don't need to be clever. This is not about creating clever plot lines - it's about creating meaty plot lines. The twist either adds more meat, or makes the meat you already ate taste funny.
The B-plot is actually an integral part of science fiction plots, although it may not be immediately obvious. See, the function of the B-plot is to create emotional resonance with the primary plot.
Most science fiction primary plots are about something big - a starship, a disease, a political movement. But in the end, the plots are an excuse to show how people live and act in those situations. The B-plot is the easiest way to lure the players into that kind of close emotional connection.
The B-plot is often "unrelated" to the main plot, at least in terms of causality. However, it reflects the human judgments and lives of the main plot. So, your B-plots should shadow your complications and twists. Since your complications and twists should be loosely themed, the B-plot will likely match many of them at the same time, and offer a human window into a world of big things going wrong.
Typically a B-plot is introduced before the first complication - often before the noose. It exists first even though it is an echo or window onto those larger events.
For example, if the vote is being screwed up because the senators are stranded on an ice planet, you'll start off with a B-plot that plays with that. Perhaps it's about planning a vacation, or the life support going on the fritz, or being unable to get off the ship, or having family visiting.
If the twist is that the lost Klingon warbird is actually waiting in ambush, the B-plot might be about your Klingon crewmember being proud of his culture's obsession with honorable combat, or it might be about an officer training test where everyone dies in an ambush: the Koboyashi Maru sequence. B-plots don't have to be clever or hard to connect.
However, B-plots do have to offer some meat. Think of them as a sequence that doesn't require the GM: when the GM is busy with a particular player, this is something the other players gnaw on.
Therefore, the B-plots should be things where the players can spend some time expressing themselves or their characters, and should involve some die rolling or content that A) doesn't require the GM and B) dictates how things go for them.
For example, the Koboyashi Maru B-plot. It isn't just a line you feed them. You say "this is an officer test." You walk them through it once or twice. But there's a table of results - if your leadership is this, this happens in the first segment. Roll weapons fire, check to see what happens in the second segment, and so on.
A concrete reference as to what can happen lets the players compare their performances to each other, and makes it easy for the GM to give various bonuses or penalties according to their efforts without having to spend a lot of time thinking up new things.
In another example, if the B-plot is that someone's family is visiting the ship and being annoying, assign a family member to every other player. Now whoever is free can be an annoying family member to anyone else who's free. Moreover, you can also get incidental synergy - if the same player is playing the goofy father and the lead starship engineer, then the goofy dad and the engineer get along oddly well. That's fine.
The whole point is to offload the content creation to the players. You need to seed it, of course - tables, NPC character sheets, whatever else is needed. But then the players run with it, try it in different permutations and different places. This lets them control the pace of the play rather than dumping it all on you, and is similar to how fantasy players might mull over what powers to buy on next level-up.
B-plot participation should be worth XP and/or in-game resources. The people who do best at Koboyashi Maru get a commendation point, for example.
Anyway, those are my thoughts.