The Best Way to Roll Dice

We all have our dice superstitions.

Ways we jiggle the dice before throwing them. The right height to drop them from. Whether we toss them up in the air or roll them straight down the table.

We probably developed these superstitions over thousands of RPG hours or from that one D&D session where we rolled them just right and landed a massive critical hit.

But what is the best way?

At the craps table in Las Vegas, dice must be rolled from a distance and hit the back wall to be considered a valid roll. The dice are also weighted and measured perfectly so the results come out as randomly as possible.

But are dice rolls truly random? One scientific study showed that dice rolls aren’t as random as you think they are. The best predictor for the outcome of a roll of the dice is the initial position of the die. Rolled with the 1 face up, the most likely outcome is a 1, according to a story from Inside Science.

"[The] top face will always be more probable,” said Thomasz Kapitaniak, a Polish researcher who authored the study.

So, if you’re aiming for a crit in tonight’s Dungeons & Dragons game, keep your die 20-side up and you might just have a better chance of landing that big hit.

And what are the best dice?

We decided to look and find out.

Standard plastic dice sold by Chessex, Q Workshop and tons of others have lots of variations due to the manufacturing process.

Dice are made by casting plastic into a mold. After that, the entire die is painted to make sure the numbers are filled in.

Then the dice are put into a tumbler for polishing, which removes the paint (except in the tiny crevices showing the numbers) and leaves the dice shiny and pretty. The tumbling process also leaves the ever-so-slightly rounded edges you’re familiar with.

Because the dice are physically worn down, that process also leaves dice with small technical imperfections that could influence the outcome of your rolls.

Those “evil” dice that always seem to roll low and the “lucky” set that seem to roll 20s more often? It might not be superstition. They could have worn faces from manufacturing that may cause just enough of a difference to affect your rolls.

Precision dice, on the other hand, are manufactured to keep from having those imperfections and thus roll more randomly. They have sharp edges and typically are sold without paint. (A plus to those of us who remember filling in the numbers with a crayon on old school dice sets.)

The dice have not been painted or put in a tumbler, so they have neat, sharp edges. They are designed to roll more randomly.

But do they?

We decided to test them out.

We took a d20 from a set of regular RPG dice as well as one from a set of precision dice. Then we rolled them through our handy Dungeon Crate dice tower from Advanced Deployment in an attempt to get them to roll as much as possible.

Here are the results:

We rolled each die 100 times. On a d20, there are 20 possible rolls. (Obviously.) If the dice were truly random, each result would come up 5 times.

The results from our (admittedly limited) test were far from completely random, but they didn’t deviate too much.

Almost every result had a deviation less than two, which felt pretty good to us.

Only two results from the standard dice deviated more than that. We only rolled one 6, and we rolled eight 9s.

With the precision dice, there were actually more deviations. We rolled eight 3s, only two 6s and nine 16s.

If you want to get technical, our test showed the precision dice were actually less random.

But just a glance at the graph shows that both dice were pretty random. If you extended this out over 1,000 rolls or 10,000 rolls, these results would become even closer to accurate.

So, we ask again: What’s the best way to roll dice? And what are the best dice to roll?

We say it’s however and whatever you decide to roll.

Get your dice on at