Thursday, 14 July 2011

What Are the Odds?

The other night we were playing a game at Walt's and this freaky roll of the dice stopped us all dead in our tracks....
We waited to see how the die was going to fall, but it didn't. Weird.
The event changed the focus of our game. We started talking about The Odds.

These are not trick dice!
The Dropship Crew has been busy looking at (and even playing some!) of this season's new game releases. In short order we have been tempted by Galactic War One, Gruntz, 5150: Star Army and the preview of Tomorrow's War. 
Wow! You got some choosin' to do! So many flavours, so much delicious detail & crunchy goodness. How to decide? What to play? I always look at the dice under the hood.

How Games Get Stuff Done
Since the day I first opened my blue-box edition of D&D I have remained curious about how and why a game designer chooses and uses the dice mechanics for their game.

Consider for a moment how games get things done.
Dice are tools, they do the job of randomising numbers. We can do a lot of work with just d6. If making straight rolls of 1-6 with a d6 is like using a hammer, rolling a pair of them must be the hammer and chisel combo. Entire game families are built on the solid foundation of the bell-curve running from 2 -12 on 2d6.
If d6 are so darn good, why bother with all the other types of dice? If we can agree that dice are tools, lets accept also that there are some jobs that won't get done very well using just a hammer, or even a hammer and chisel.
Nothing wrong with a hammer, I like hammering, but there is a lot to be said for other types of work as well....

But I want more granularity? Why not! There is a 16.67% chance of rolling any one single number on a d6. On a d8 it's 12.5% and for d10 it's a 10% to roll any one number, all the way. The more sides to any die, the smaller the percent chance of rolling any one number and the finer the 'granularity' or greater potential for detail in your game system.

I was an Early Adopter of Multi Die-type Games
When you roll a pair of dice the the percentage chance of rolling any number along the range (say between 2-12 on 2d6) is distributed; you are more likely to roll some results and less likely to roll others. That's the bell-curve at work.
Now, lots of detail is great for some people and pure hell for others. The designers' holy-grail is to find the balance point for fun and detail. They need to choose dice mechanics that work well.

Yes, I find the mechanics inside the games we play fascinating and mechanics will definitely influence my choice when it comes to buying a new game. The rich coating of fluff may be inspiring and also important, but not as much as how the game actually plays.

More Tools
I have found a truly excellent tool that allows us to explore dice mechanics for ourselves. Check out SmallRoller, it's free, freely distributable and fascinating. It's been a long time since the Fnordistan Dept. of Software Engineering blog has been updated, but this bit of code is worth the visit.
With SmallRoller you can recreate pretty much any dice-roll called for in a game, displaying the probabilities as percentages well as a cool Probability Chart. Highly enlightening and useful for players as well as designers of games!


  1. If your dice can land on their corners, you should be using different dice!

    Not all dice are created equal. If you go to a casino you'll see that the craps dice have nice crisp, sharp edges. This is because they're more random: dice which have had their edges ground off are less random and can also be "forced" by someone who has the knack.

    A couple of years ago I found some very informative videos on YouTube from a dice maker talking about his art. They're on this page - worth taking a look I think!

  2. But note that casino dice are meant to be bounced off a wall. Rolled straight from the hand, they tend to fall flat with no roll/bounce.

    But that's an awesome pic of that dice roll. In high school, we flipped coins in an experiment on determining hereditary traits, and the coin landed on edge. We, of course, immediately claimed a mutation for our experimental creature.

  3. We had that happen once. For a moment everyone looked at each other as if wondering what to do, then everyone started blowing on itto try and topple it in a direction that would most favour them.

    I can't remember the ultimate outcome... ?

  4. Then of course there are the dice without numbers or a mixture of numbers and symbols i'm not sure how roling a d6 + d4 and an "arrow" die would work when looking at the statistical curve? or a D20 that went 1-10 and then 5 skulls and 5 "!" but it's been a while since i picked up a game with it's own custom dice set. Ever since wargames and rpg started adopting the single die systems only d6, d10 and d20 seem mainstream these days.

  5. That is the most bizarre thing I've seen this week. :)

    @Jodrell - Cool link. I remember seeing a debate about rounded vs sharp cornered dice on a Warhammer forum years ago. I seem to remember that the sharp-corner claim was debunked by several sources. I even think someone said it went the other way, and you'll average more bounce and revolution with slightly rounded corners.

  6. @ Jodrell that's a great post thanks!

  7. That actually happened to me the other week too. There was a divot in the foam board that we were using on our game table, and a die managed to land right on its corner in the tiny divot. I'll have to post a photo...

  8. Chris, I don't think getting more bounce and revolution matters for getting a fairer, non-skewed result. Lou Zocchi's contention, as I recall, is that sharp edge dice stop sooner after hitting the table and are as likely to stop on any one side as another. The rounded dice roll more on the table and due possibly to uneven rounding of corners, might favor stopping on certain sides than others. This makes sense to me as long as you get the die spinning before it hits the table, but if you can remember the sources debunking this, I'd be interested in reading it.