When the group doesn't or shouldn't form

This consideration is actually born from games, during which the player characters don't make a whole lot of sense together as a group, and to me, this really does seem to encompass most games.  Questions then arise: what contrivance should the GM do to bring everyone together?  Should the players design their characters with the group in mind?  Is it okay to have sessions where the characters are separated, and if so, how do you handle that?  Do you need multiple GMs?  Should you employ cut scenes or hand outs providing synopsis?  And finally, what do you do when one character learns something that the rest of characters couldn't: do you tell them anyway, keep it secret, or something else?

The problem here actually begins at that long list of questions because a good game is about story and stories do not always have the main characters working as a team standing next to each other the whole time.  In fact, in most stories this just doesn't happen at all.  People split up or they begin not knowing each other and it's only by peeling back the first few layers of the onion that they even begin to become aware of the other players in the drama.  I'm currently watching The Expanse, which is a great show, and it's clear who the main characters are.  In the episode I'm on, which is pretty far on, two of the main characters have finally met.  This is true also of Stephen King novels.  Something like Stranger Things presents a similar problem in that everyone knows each other, but they're all facing different parts of the mystery and they're always working in teams, rather than together.

The Team conception, to me presents a similar difficulty, in that I just don't picture every story involving a Scooby-Doo-like set of investigators showing up already invested, as a team, in solving a mystery or a problem.  Games designed around this tend to work, but there's only so many designs where this tactic can be employed.  If you play a hit team in a cyberpunky kind of game, you eventually get double-crossed and then you're going to need to split up to investigate.

So, what's to be done?

The first thing I would say is that an RPG is already a game of individual action, it just appears to be team action.  As the person running the game, I explain the situation and then ask, "what do you do?"  It's really not so hard to imagine a kind of game where the GM turns to one player, explains their situation and then asks them what they do, followed by the next player and the next player and the next.  Time and distance are really only functions of the imagination here.  Are there problems with this tactic?

Well... yes.  Generally describing the situation takes time.  We like to play through an entire scenario before moving on to the next.  If you have four players, all in different places, you have four different scenarios to describe.  To play this way, you have to be cognizant, ultimately, of what you're trying to pull off.  But, if cognizant, this tactic is actually easier done than you might think.  You may find that player's "turns" are 2-3 minutes long rather than 30 seconds, but if you're players are patient enough it works.  A key to this is obviously that the players who aren't actively playing at the time need to be interested in that story.

Being interested in a story in which your character is not involved, though, is kind of hard.  An RPG isn't a movie.  We can't passively pay attention and just watch while munching popcorn.  When it's not our turn, the tendency is to tune out.  I would say that this is already dangerous in a PbtA game where there really aren't turns, but still it happens.

My remedy for this problem is actually kind of simple.  Let the players act on what they hear happening in the other scenes that they are listening to.  I don't mean this in an obvious game-y or even Fiction kind of way.  I'm talking of metagaming, and I know that's a dirty word.  So a defense...

First of all metagaming or metanarrative happens in fiction all the time.  Without getting into deeper possibilities here, which really only matter to French linguists, when character A makes a discovery and explains it to the audience, only to have, in the very next scene, that discovery matter to character B half-a-world away.  That's a kind of metanarrative structure.  People do it all the time in film, and it works.  Why not in your game?

Now, an advantage, if, in this scenario, the player controlling character B is acting on information from character A then they are being drawn together.  Recall the original point of this post, how do you deal with groups that maybe aren't designed initially as groups.  As the characters follow leads, and listen to other leads which they accidentally stumble upon or act upon without the character knowing what they're doing (even if the player does), the characters are being drawn together as fellow investigators.

I should point out that this has to still be plausible.  If character A learns that the oil tycoon is actually a werewolf without character B being present, B cannot know to go out to buy silver bullets, but B can be suspicious and might act upon that suspicion to investigate more.

So, takeaways... the first is just to run characters on opposite sides of the world as if they were in the same room.  The second is that if the players of these characters metagame it a bit, they will start moving their characters closer to each other and closer to being on the same plot line.  Then they really will be in the same room.