Tuesday 20 April 2021

Running a Matrix Game


This is my first time. 

After last week's post I've continued to run my Sunrise Expansion game, which has largely been a success, but has already flagged up some early mistakes and lessons to take on board for next time.

So with that out of the way, how am I actually running this Matrix Game?

If you want proper expert advice then go here, but I can talk about my own early, scattered thoughts.

For a bit of context, here's the player-facing rules minus the secret briefs given to each player.

This is all happening on Discord, with a private channel for each of the Actors, and public channels for open discussion, out of character questions, and end of turn reports posted by the Referee. 

Answering Questions

With the world being painted in broad strokes, it's natural to get questions.

How does money work in this setting?

I don't have a fleet, can I still get to Mars?

Are there space pirates?

If subjects like this were left out of the brief then the answer is generally something that isn't going to cause a lot of friction or create huge opportunities to be exploited. So money works like it does today, flying to Mars is like hiring a private jet, and no there aren't space pirates. 

But I'm walking a fine line here. If you ask me something about another Actor, then that's probably going to require some actual investigation. I have a special Consultant role for players to carry out this research, but in general you're going to need to use an action to find out your rivals' secrets. 

A lot of crossover with RPGs here. If you declare an action and I think you've misunderstood the fictional grounding, I'll talk it over with you rather than taking it as your final word. 

The Single Action

Now this is an interesting one. All of the guidance I've read for Matrix games hammers home the rule that each Actor's turn consists of a single specific action. So if you want to increase the garrisons on your walls and build a new superweapon, you've got to decide which one is more important. This game has its roots in military training, and the rule exists to force players to prioritise and decide what they absolutely need to do right now.

I'm sure my players won't mind me sharing that I've had to be strict on this! The temptation is always to try and squeeze as much as you can into a single turn, but here it's all about focus. 

Besides, resolving six players all doing different actions is messy enough without adding in secondary actions for each of them. 

Arguments in Asynchronous Play 

Matrix Games are built around a foundation of players declaring actions, then the rest of the table discussing how likely it is to happen. Often it's pitched as arguing, and the action is given a modifier from -3 to +3 relating to the pros and cons drawn up next to it.

I can see why this works for a tabletop environment, but it's not a great fit for playing on Discord with fifteen people in different timezones. 

So I'm embracing a Benevolent Dictatorship for this. Straight-up "one GM, many players" that we're all so familiar with. I look at the Action and adjudicate it all by myself. 

And I actually don't feel like I'm missing out on much by having the players not have to constantly argue about chances of success. Instead it's down to me, and hopefully the players trust me enough that they don't feel cheated out of a fair process. It helps that this is all taking place in a fictional setting, so there isn't much call for Subject Matter Experts. 

I'd definitely like to try a Matrix game with this more traditional argument system, but I don't feel a great sense of loss. 


Some Matrix games require an action to have "three reasons why" it's likely to succeed. So I can send a spy into your parliament because:

  1. I have a well established spy network.
  2. We have open borders.
  3. I have sympathisers in your country.
Again, this is a gut feeling thing, but it feels like box-ticking to me. Again, in an educational environment it makes sense for players to be pushing themselves to think of factors that would benefit the action, but I'm just running this for fun.

So I shortened it to just "Leverage". Tell me why this action is likely to succeed. You might still reel off three reasons, but you could just as well give a single compelling reason. 

Then I look at any opposition to the action, which could come from another player's action or from an external factor in the established fiction. 

If there's no opposition then the action is Unopposed and you get your desired outcome. Otherwise I weigh your leverage against the opposition and grade it as Strong or Weak in comparison. Strong leverage needn't be overwhelming, it just means you've established enough to reasonably overcome the opposition. 

We'll get onto what this means for the Outcome below.


There's a concept of "Narrative Bias" in Matrix games, where most dice mechanics are weighted towards the player carrying out the action, so it's easier to impact the world than a 50/50 shot would allow.

This is fine, but I wanted every Action to have some sort of impact. Fail Forward, and so on. 

So my Narrative Bias is "Action is certain, Outcomes are unpredictable"

When an Action faces opposition, I roll 2d6. If the leverage is Strong I keep the high die, if it's Weak I keep the low die.

4+ means the player got their desired outcome, 3 or less means they get something worse. If you want to get fancy you can treat 1 and 6 as Criticals. 

So if you want to kidnap a journalist to silence them, then the kidnap operation is definitely going to happen, but a failure would mean that the action backfires on you somehow. Maybe the location you're holding them is leaked, you get the wrong journalist, or... wait, did I roll doubles on those dice?

When the dice show doubles, I use this as a prompt to insert a "Force of Nature" event if something fits. This represents those chaotic events that occur outside of the influence of any player and make things extra messy. In this case, maybe the driver is hit by a truck, killing the journalist. I'd be careful with these, as you don't want your world to be utter chaos, but it can add a little bit of spice to things.

A good rule of thumb for creating Outcomes is "RAT". 

Reasonable: It feels fair and isn't disproportionately punishing. Killing the journalist might help you in some ways, but it's also threatens to create a lot of heat for you. 
Actionable: It isn't a narrative dead-end. It creates hooks for further actions and attracts the attention of other players. 
Traceable: You can look back at the steps leading to this. Maybe the Journalist was already being targeted by another agency, or the previous turn there was talk of flooding on the roads. This is why sowing a little flavour into your reports can be really useful. 

Early on I was probably letting actions go by a little too smoothly. Just because there's no opposition to an Action doesn't mean that it should be Free. My mantra of the moment is "No Action without Friction". Every decision should leave some sort of mark on the world, and it should never be wholly positive. 

I much prefer Friction to Hurdles. So for the most part I'll give players a way to achieve their action, but give them messy consequences. Feels more fun than saying "No you can't do that" or "You can do X but only after you've done Y and Z to prepare."


I try to keep these brief, written in the style of a news roundup, and it's simple enough.

If there's a secret action that I need to report privately ("okay, you got your spy into the opposing parliament") then I always give a clue in the public report. No perfect secrets here.

"Newly appointed Senator Grey commented on the attack as an affront to Democracy"

It can be small, like that, but remember that when Senator Grey reveals himself as a traitor in three turns' time we at least want to be able to point back at his existence. Remember: Reasonable, Actionable, Traceable.

Actionable is the one that I still need to work on. Think of this Report like you're writing the initial brief again, trying to draw the players in. Fill it with hooks. The downside of having a game where the players can try anything is that they can become overwhelmed and feel directionless, even with their objectives in mind. Put something right in front of them, and even if they don't bite it'll at least give them a starting point for their strategy. 

Is this still a Matrix Game?

When I've tried to talk to friends about this, the term Matrix Game is really unhelpful. "No it's not The Matrix. No, it's not the Matrix Wargame publisher that make computer wargames."

Of course it's just one step removed, but I'm at least trying to think of a different name to use to pitch this to potential players. Open Strategy Game? This sort of thing is weird enough that it will always require some explanation, but I'll see how it feels in conversation.

Later this week I'm trying something else that's adjacent to this style of play, and hopefully I'll be able to talk about it next time.


  1. In the UK they call them Megagames, because there are so many people.

  2. I typically refer to them as Engle Matrix games to differentiate them from the publisher. Fun fact, Engle Matrix games are robust enough that you can, IME, play them solo much as you would and hex-and-counter wargame.

  3. It's not always going to be open war, so wargame doesn't apply as much. Politigame? Something of that sort?