Monday 10 May 2010

Dice on Fire!

Don't panic, I'm not dishing out Corporal Punishment to my dice today. After a very bad session a few weeks ago I vented my anger on my d20 and destroyed it. I bought myself a new set of dice and over the last few weeks they have performed admirably well. On Friday however they entered a league of their own. My new D20 is fast becoming a legend.

I rolled four natural 20's in one evening three of those being combat critical hits. I've not been keeping detailed records and conducting statistical analysis, but this d20 definitely seems to be rolling high compared to its predecessor. There are two explanations. First, my intimidation tactics worked, or second, the dice isn't as balanced as it should be.

I read somewhere that gaming dice are not made to the same high specification as Gambling dice and that small imperfections such as air bubbles or irregular edges could 'unbalance' the dice. Another possibly more plausible explanation is that my dice have swirls of colour within them. Two or more coloured plastics have been blended to create this effect and of course the density of each component may be different. The variation in balance for such dice is likely to be tiny (less than 1%) but could statistically bias towards a particular number and would not necessarily be positive.

My new d20 clearly seems to be balanced towards better rolls, but is it cheating? Well if I'd gone out and bought loaded dice then yes it would be cheating. But this was just the luck of the roll (no pun intended) when I bought this set of dice and not the set next to them on the shelf. Most regular gaming dice are probably unbalanced but not to statistically significant levels. So every gamer has an equally random chance of buying a slightly unbalanced die that rolls high. The trick is identifying the bad dice and eliminating them from the gene pool so that you will be left with the high performers.

The moral of this story is that it doesn't matter how much you paid for your dice. In the end it all comes down to luck.


  1. It's a bad workman that blames his tools (for good or ill!) ;)

    I think you are a long way (a few tens of thousands of dice rolls) from demonstrating any genuine bias in your dice.

    Besides the biggest single factor biasing almost all dice is the amount of materialremoved from each face to form recessed numbering - but that would mean that every dice was biased towards rolling whatever numbers are opposite those faces with the least material removed.

    Post your results of several hundred million dice rolls covering multiple materials, markings, orientation of faces, rolling surfaces, corner and edge profiles, and double blinded rolling techniques from different gamers and maybe ... just maybe THEN I'll believe your dice is special enough to make a difference ;)

  2. Nah, definatelty high rollers. I mean a few dozen rolls must be a statistically significant sample!

    [whispers to dice] Don't listen to the bad man, I think your special.


Thank you for leaving a comment. I always try to reply as soon as I can, so why not pop back later and continue the conversation. In the meantime, check out my YouTube channel Miniature Adventures TV