Thoughts on User-Friendly Mechanics

As game designers, we have to be careful to not let our own sense of “cleverness” turn into a poor experience for players. Often, we approach this line by making fiddly dice mechanics. When it comes to making games, here’s my philosophy: what’s easy for you to do isn’t inherently easy for everything, especially when you’ve gained plenty of practice at making your mechanic work because you’ve tested it over and over.

What I consider reasonable to have in a mechanic:

  • Finding the highest (or lowest) die in up to ten dice
  • Finding whichever dice in a die pool are of a certain number or higher, generally not more than ten
  • Adding two single-digit numbers together
  • Adding a single-digit number to a double-digit number
  • Subtracting up to 5 from a number, or subtracting whole tens (10, 20, etc) from a number

What I start eyeballing (but don’t just automatically toss out):

  • Die pools that involve more than ten dice[1]
  • Adding three single-digit numbers together[2]
  • Adding two double-digit numbers together
  • Other subtraction
  • Subtracting one die from another[3]
  • Multiplication involving a single-digit number
  • Moving dice around
  • Rotating pips/sides (rotating a 3 to a 4, or a 4 to a 3)

What I try to avoid entirely:

  • Division
  • Multiplication of two double-digit (or higher) numbers
  • Applying individual rules on dice in a pool

[1] Mythender breaks this rule, but I was conscious of this while designing it. Because you earn those dice, and because you learn the rules when you have fewer dice, those times you’re rolling 30 dice have meaning (which don’t happen often), and  in practice it’s worked out alright.

[2] Like GURPS’ 3d6. I can do that in my sleep, but I have watched over the years players stumble on adding three dice together.

[3] Like Feng Shui. This is less about arithmetic than it is about confusion over which die to subtract — something else I’ve seen over the years.

The Test & the Reason

When you’re playtesting your game, refrain from (beyond a demo or two) interpreting the dice to generate the result. Have the players do it. Watch those who do it quickly. Watch those who struggle, and if others have to bail them out.

This is important for two reasons: (1) constantly having to pause to interpret the post-roll beat can break the mood; (2) when you can’t figure out how to interpret the dice and always rely on others to tell you what you rolled, you don’t feel like an equal participant.

– Ryan


21 Responses to Thoughts on User-Friendly Mechanics

  1. Dave T. Game says:

    Non-numbered dice (or even dice with alternate number distribution) also fall into the “Start eyeballing” category for me. If the icons are obvious what they mean, it works, but it’s tricky to pull off and often gets muddied by non-intuitive icon sets which requires another set of lookups. Likewise, if the dice have an alternate numbering scheme, you can get a drag on analysis when players pick them up and have to examine all sides.

    • Ryan Macklin says:

      True. If you can make that sort of mechanic around just simple pattern matching, like matching it to an icon on the character sheet, then you’re cutting down on the interpretation beat.

      – Ryan

  2. So would dice systems with multiple colors of dice run afoul of “Applying individual rules”? Take “Don’t Rest your Head” as an example.

    • Ryan Macklin says:

      Good catch. No, I mean awkward formulas based on the individual die faces, like “if it comes of a +, treat as a +2 instead” or anything else that’s cumbersome.

      – Ryan

    • Wayne Zombie says:

      DRYH was the first thing that popped in to mind for me. I absolutely love the concept of the game, but I find the dice system kind of cumbersome, plus I’d have to buy a bunch of dice to play or run the game.

    • Ryan Macklin says:

      I’ve heard that criticism of it, and to be fair that’s part of what amused me about building Mythender.

      The die dynamic of DRYH is enough for me, given that it’s pattern recognition, but I have watched others struggle with it in play, so similar is on my “eyeball” list, but not something I’ll totally discount.

      – Ryan

  3. Judson says:

    “…when you can’t figure out how to interpret the dice and always rely on others to tell you what you rolled, you don’t feel like an equal participant,” is hot. I think there’s a whole social dynamic around rulesets, games, tables and players where one player is interpreting the rules for another that deserves more discussion.

    • Ryan Macklin says:

      Totally! I have a lot of thoughts regarding how we accidentally (or intentionally) make each other feel equal and unequal at the table.

      – Ryan

    • Nancy says:

      I agree! This is a topic I would love to see explored in more detail

  4. Andy says:

    I’m suspecting that this is also tied into the frequency of the dice-rolling. For instance, Fiasco’s dice strongly break a couple of these rules (the eyeball of “adding multiple numbers” and the “subtracting some dice from others”), but you only roll them twice (when totaling them, that is) and ruminate over the result, so you can afford to spend time figuring it out.

    That being said, a game where you roll dice that infrequently is rare.

  5. Alan Kellogg says:

    What about before the roll? The part where you determine the number or numbers to roll,

    In DJ:
    1. Find base number
    2. Modify base number by multiplication (may involve percentages)
    3. Modify base number by addition or subtraction.
    Roll modified number or less on percentiles.

    • Ryan Macklin says:

      Yup! The post I linked to the last paragraph talks about that beat, and while I was thinking today mostly about post-roll, you can do a lot of bad UX in the pre-roll.

      What is “DJ?”

      – Ryan

  6. Ken Burnside says:

    I use 2d10- in a lot of my games. It came out of the appreciation I had for Feng Shui’s mechanic (which I stole and re-used with d8s). Feng Shui also had exploding dice as both positives and negatives, and that proved to be the “roll to clever” – the one die is negative, the other is positive isn’t the problem – it’s combining that with “and sixes are explosive.”

    Roll 2d10. Subtract the lower number from the higher number. This, like 3d6+, is a roll where “practice” (Doing it about a dozen times) turns it into pattern matching. There is a minor reduction in the number of reps to pattern recognition by treating “0” as “0” rather than 10, but by the time I’d figured that out, I had too much established work using “0” as “10”

    I experimented with 2d12-. The difference in reps until subtract 3 from 11 becomes automatic versus 3 from 10 is surprising. I’ve also experimented with different die sizes in this methodology (d10/d6, subtract smaller from larger) because it makes for interesting curves, and found that players fell into the expectation that there would be special rules for special dice.

    When in doubt, batch math. One of the major churn points in Pathfinder is “And I get this modifier for flanking, this modifier for charging, this penalty for this condition, and the target is a thus, so I get this modifier from my class, and this modifier from my weapon…”

    Lots of players never learn to batch math, or even to write the results of the math down for future reference – so the entire process gets repeated multiple times.

    • Ryan Macklin says:

      The thing is that lots of players are also innumerate, or struggle with math, or are embarrassed when they try. So things like batch-doing the math is concept alien to them.

      Yeah, I think it’s worth adding “regularly adding/subtracting more than three numbers in a given moment” to my eyeball list.

      – Ryan

  7. Martin says:

    Amen to that.
    This sums up many problems I’ve got with %-based systems, which often tend to end up in the eyeballing area far to often (“You’ve got a skill of 72 and in this case, a situational modifier of -17 applies… remember that you score a crit if you beat the roll by 30 or more.”)

    Another point that always makes me skeptical is any rule which requires adding up results from several rolls (I think it is called ‘extented tests’ in Shadowrun, for example.) You know, the “You need a total of 28 net successes” rules.
    Either the rolls are executed directly after another, and than there is no reason why it shouldn’t resolve quicker with a single roll. But when they are streteched over a longer amount of time and you do other stuff in between, you always have to keep track of this unfinished calculation, which does horrible stuff to the rhythm of the game.

  8. Robert Calfee says:

    I had not thought about measuring the ease of dice mechanics until I’d been in games for most of my adult life. What brought it to my attention as something you could tinker with was playing in my frioend’s home brewed miniatures skirmish game ‘Rules Crusade’ (based off the Space Crusade game). The dice were 6 sided, white dice had a a one and a two, the red had a one a two and a three; the other faces were blank.

    If you rolled over the armor of a target you did some hurt to it. Ranges subtracted one for every foot. It was so elegant and easy and quick compared to the WH40k combat, and when I did the math it came up about the same in terms of chances (my friend did his homework). Melee was contested dice.

    There were some limited re-rolls, granted by special weapons and mostly used in melee.

    When I compared this to the effort I had to put in to combat in AD&D, Champions and the Runequest games I found myself a convert to the notion of maximizing the ease of resolution for a given game environment.

    I mostly play on line PbP, and so I have other considerations to make for ease of use, I can let the diceroller do all kinds of math for me, so tracking ongoing statuses is more a bother.

    One additional thought is that in Cortex Plus, the fiddling with the dice acts like a mini-game within the conflict, and as you stated rolling the handfuls of dice in Mythender is kinda the point. In the former, it serves as the tactical element, in the latter, to paraphrase Illuminati “If you can’t be bothered to count up your dice you have no business trying to kill a god.”

    • Ryan Macklin says:

      Specifically for Mythender, the dice serve the job of tactile narrative reinforcement, which is something we don’t do much of in making games. That’s the only reason I kept that mechanic, because it was doing the job it was meant to do so well.

      – Ryan

  9. Daniel Ream says:

    Here’s a big one: integer math, if adding positive and negative numbers in the same calculation, _especially_ if you cross the zero on the number line. I’ve watched the card game Gloom screech to a crashing halt because of this.

  10. DaveC says:

    I think a good rule of thumb is that if you can’t picture in your mind what a “good” roll is before the dice hit the table, then the mechanic is too fiddly. Ideally there ought to be a connection between what you’re imagining will happen and the very first glimpse you get after the dice have stopped moving. It’s the feeling you get when you roll a natural 20 in old school D&D. If you have to do processing steps after the fall then you’ve delayed that “yeah!” moment and distanced it from the rolling action you took to try to achieve it, which really weakens the impact.

    I like rolling a big handful of D10’s in WoD and immediately having those zeroes jump out at me as an indicator of success. But if I start digging through the pile and cancelling things out with 1’s only to find the roll wasn’t nearly as good as I first thought, then it dilutes the appreciation of it a lot.

    • Ryan Macklin says:

      Hmm. I sort of agree, if by “you” you mean a player, not the designer. At the same time, it’s not especially useful as a rubric, since I can imagine a good roll in your example quite easily, and it’s a suboptimal result that’s the problem.

      – Ryan