The Math Will Fuck You

In face-to-face game design (pen & paper RPGs, card games, etc.), I have a mantra:

Don’t design around math. The math will always fuck you. It will distract you from the point: designing around experience.

(And I say that as a software engineer that almost double-majored in math.)

I first suffered the “joy” of learning this lesson years ago, when trying to design a game about Catholic special ops vs. True Evil RPG (which perhaps one day I’ll revisit, maybe). My friend Mike & I were GURPS fiends, but we weren’t big enough to get a Powered by GURPS license, so we wanted to make our own system. Plus, hey, build a system.

We decided that the d8 didn’t get enough love, so we started building a system that revolved around rolling 2d8, adding a number, and comparing. Then we added another dimension, where at times you could roll in another d8, but you still added the highest two. We did all sorts of Excel charts and figured out what point we wanted to set our “we’ll assume a competent characters will roll this amount” target numbers.

We played the game. It was meh. We went back to the math to see why. We moved target numbers around. We added and then removed dice. We kept tinkering. And eventually, after not being jazzed about it, we put it away. It had some cool moments, but they didn’t overcome the overall meh factor.

What I realized much later: I didn’t care for rolling d8s all the time. It was a “cute” experience, not one that felt natural, but needlessly novel. That’s not something the math can tell you. But it’s something looking at the experience of play — the whole damned point of game design — can. Unfortunately, many of us are math nerds, so we want to solve problems with math (not unlike how computer programmers often want to solve with software user issues that are outside of what software has contact with).

This was not the last time I looked at the math and was lead astray. It took several failed attempts at design to finally come this: I only look at math when I recognize a problem and know that’s what it needs. Until then, I look only at experience. Otherwise…

The math will distract you.

The math will tell you things that are unimportant.

The math won’t tell you how a system is experienced.

The math will fuck you, if you let it. Use math to double-check an idea, not as a design criteria. (Unless you’re making a video game, or maybe the next D&D.)

– Ryan

(Oh, and I find it kinda amusing that if you pay too much attention to math in game design, it’ll fuck you. But if you don’t pay enough attention to it in publishing, it’s also fuck you.)


14 Responses to The Math Will Fuck You

  1. J.B. Mannon says:

    I have known this fact for years. No matter when you dive into the math, it is going to fuck you. I have stopped trying to design the math too closely at the start of a project and rather go for the feel I want. I have a good grasp on what feel various mechanics bring to the table and I go for what looks obvious. I nail down the feel as much as possible and then go back to the math.

    Math still ends up fucking me every time but it lets me push that moment off for as long as I can.

  2. Codrus says:

    When I regularly did playtests for people, I focused on the math, and where it broke. I think that’s useful, but I agree with your basic premise: let the math guide you or answer questions, but don’t let it constrict the design to the point where you can’t make forward progress.

    I’d put this in the same category of editing while you are writing. Sometimes it can be helpful to reread what you wrote, and sometimes you just need to put the machete away and make forward progress on the document as a whole. There’s plenty of of time for machetes later. :D

    • Dave T. Game says:

      Sounds like it’s good to ignore the math in the design stage and pay attention to it in the development stage.

    • Ryan Macklin says:


      Not even then, but your comment’s dislodged a possible way to describe it: When you run into a problem with the experience of the game, and you recognize that math will help you solve it, go to the math. Solve. Then back away from the math. Doesn’t matter when in the process that is — initial design, playtesting, whenever.

      – Ryan

  3. Gareth says:

    My original draft of Mongoose Traveller had this funky thing where you rolled 2d6 + stat mod + skill, and then picked one of the dice you rolled to be how well you succeeded, and the other to be how quickly you did it. It worked really well in play, gave all sorts of interesting decisions, and nicely modeled a lot of stuff that’s important in sci-fi games. (‘Scotty! If you don’t fix the engines before the shields fail, we’re doomed!”)

    It got torn to shreds by some Traveller fans, who pointed out all sorts of edge cases and argued that it didn’t reflect the skill of the character enough (and to be fair, the system did need adjusting). Management got scared of the criticism (we were doing an open playtest) and at the last minute gutted the system and changed it to a very conventional stat + skill system which works ok, but doesn’t have the same flair as the playtest version did.

    It would, I am convinced, have been a better game if we’d gone for fun instead of bowing to the maths.

    • Jess says:

      I’m curious as to how the previous system worked in situations like combat. It seems like that’s the point where it might break down and become cumbersome, but I like the feel of it.

  4. blackcoat says:

    How are d8 different then d6 or 10?

    • Avaril says:

      A d8 has more sides than a d6, but fewer than a d10.

    • Ryan Macklin says:

      That’s actually the thing! On paper, not a whole hell of a lot. But games don’t take place on paper; they take place in squishy irrational places in our brains.

      (Also, there are more differences between a d8 and d6 or d10 that most people realize, because people focus on the numbers on the die rather than the die itself.)

      – Ryan

  5. How dare you – a software engineer – impugn the almighty maths!

    Actually, I agree with you here. Not sure what else I can add to the mix.

  6. Eddy Webb says:

    Something Justin said during one of our panels at Grand Masquerade:

    “No one remembers how well a system was balanced.”

    • Ryan Macklin says:

      Well, they do remember it if it’s poorly. But then, that gets back to “this feels like crap” — play experience.

      – Ryan

  7. Joseph Teller says:

    I think this depends on your audience. Some audiences are more forgiving of the math than others.

    Unfortunately I’ve had a lot of experience gaming with Accountants, folks with PHDs in Physics, and Business Majors. They get VERY hung up on the math. My solution has generally had one or two of *them* do the number crunching on designs to find the flaws and patterns and problems (as well as the sweet spots).

    • Ryan Macklin says:

      I’m talking about in the design process. And having worked on games for math-y audiences, I’ve found it rings true even still. (And for that, math in the playtest and revision process is where it’s needed, during/after it hits contact with those people.)

      – Ryan