# Worlds of Design: Too Much Dice?

Game designers: You don't need bucketsful of dice in your RPG rules!

Picture courtesy of Pixabay.
Please Note: I'm not against dice themselves. I’m not saying that using dice is bad in game design. It depends on what you want to achieve. If you want a game rather than a puzzle, some kind of chance or uncertainty is required, and dice is a common way of introducing that. What I'm saying is you should not need huge numbers of dice.

## The User Interface​

Dice are part of the user interface (UI). The UI is very important in video games, but it exists in tabletop games as well. (While video games might not overtly use dice, many use randomizers that are equivalent to dice.) The difference may be that there are usually more than one person involved in a tabletop game so mistakes will be more noticeable, and that tabletop interfaces have settled into many familiar techniques because of long use.

## It Started with a Single Die​

When RPGs originated, we already had the example of the many Avalon Hill wargames that used a single D6 and a lookup table to resolve combat. We don’t know who originally used a 20-sided die to resolve situations in games, but it certainly had the effect of avoiding the use of several dice, and did not require use of lookup tables any more than several dice do.

There is an outstanding virtue of using more than one die: you get a “normal” or Gaussian curve of the sum rather than a linear result. That is, the results in the middle are more common than the results at the “edges”—with two dice a sum of 7 is six times as likely as a 2, twice as likely as a 10. Yet with the D20 linear result you have enough distinct choices that you may not need a Gaussian curve, and in many cases you don’t want a non-linear Gaussian result in your resolution.

For example, if you need a 10 to hit, and you roll a D20, you have a 55% chance of success. If you roll 2D6 and sum, you have 1 chance in 6 (16.66%) to hit. With the sum of 3D6 the result would be different again (the average roll there is 10.5).

## Some Examples​

The best-known examples of “bucketfuls of dice” in tabletop role-playing games are Shadowrun, Champions, and GURPS, but there are many others. Some of these systems count successes instead of adding up results. Combat also involves a lot of dice checks, which might make more sense in modern setting where guns are involved and results are more likely to be lethal. Over time, these systems have been refined to be easier to use, but some designers are still including massive die rolls in their games.

Several years ago I watched an atmospheric post-apocalyptic RPG session. The campaign setting supplement that I read for a couple hours during one session contained virtually no rules references; they were descriptions of places, people, technology, etc. It was a setting that cried out for a simple set of rules so that players could savor it. The game rules, at least as the people I watched play them, were quite complex and required large numbers of 10-sided dice. Any activity check required a player to roll several D10s and add the sum.

When there was any combat (there was a lot) it was worse. In some cases there was a to-hit roll, then an avoidance roll, then a determination of hit location, then armor absorption of damage and recording how much had been absorbed by the armor at the particular place. This often required someone to roll a lot of D10s.

## Why This is a Problem​

Large pools of dice create a complex set of mechanics, which has unintended consequences that end up burdening the players.
• Adding Isn’t Easy. Adding sums quickly is easier for experienced players who are practiced in making quick calculations on their heads, but it may not be as easy for new or younger gamers. They have to think about the result, instead of just seeing it.
• Legibility. Different dice have varying levels of legibility. It’s not as easy to read a D10 as a D6, and that issue is compounded the more dice the game uses.
• You Need a Lot of Dice. Rolling a lot of dice also makes for more problems than rolling just one, such as dice falling off the table, and a shortage of dice (especially D10!). In the above example, players ended up passing D10s back and forth because only one of them had enough. This all wastes time while no one is having fun.
In the next article we’ll explore how dice shapes a game’s play.

Your turn: What RPGs minimize or even eliminate dice rolls?

### Lewis Pulsipher

Dragon, White Dwarf, Fiend Folio

#### Faolyn

##### (she/her)
I think it would be a good idea to keep neurodivergence in mind when designing dice mechanics, but I don't know enough about how different neurodivergences interact with that kind of thing.

EDIT: To expand: is it easier for a person with autism (or insert neurodivergence here) to add the results of 3 dice, or to count successes (say, 4 or higher) on those 3 dice? To use a D&D 5E example, is it easier for a person with a common neurodivergence to read the result of a d20 and add their proficiency bonus, or to roll a d20 and a proficiency die and add them? I honestly don't know the answer to these questions.
For me (autism + ADHD), those would be equally difficult (or equally easy). OTOH, Something like 3x/Pathfinder, where it's d20 + a whole bunch of different modifiers, would be more difficult because I'd forget all the possible modifiers, or potentially add one of them twice.

#### ART!

##### Deluxe Unhuman
The autistic teen that I gave my book to, his Mom said he loved math, and numbers. I think it unfair to say ND equals "can't do math".
That's an excellent point, and I didn't mean to imply that.

That said, "can't do math" and "read different kinds of dice results better than other" aren't necessarily the same thing.

#### dragoner

##### KosmicRPG.com
That's an excellent point, and I didn't mean to imply that.

That said, "can't do math" and "read different kinds of dice results better than other" aren't necessarily the same thing.
True, I wanted to make the statement, before wrestling with implications.

#### aramis erak

##### Legend
Or Forged in the Dark games, where you'll be rolling anywhere from 1 to 4 or 5 dice, but there's no maths required at all - you're only interested in the highest result you rolled.
Comparisons of numbers are part of the Mathematics Curriculum.
It's not notably faster than addition.
Multiplication is slower still
Division is slower still
Any fractions slow all of those considerably further.

At least, based upon the math speed tests used in the Anchorage School District.

#### Bacon Bits

##### Hero
I'll also say that 3 is not a bucketful.

But I also remember us playing ShadowRun, and everybody had one of those cases of thirty-six 12mm d6s. The joke was, "roll 1dBrick". I also remember playing Vampire and became clear that the dice pool system didn't really work very well.

Overall I think the problem with dice pools are that it's hard to tell what's really going on. The math isn't obvious. It's really easy to understand the math of a single die roll. Two dice are not immediately obvious, but not that bad. More than that and you're down to intuition at best. If you're adding them up, it gets weird because the range of outcomes is really high, but the standard deviation is really small. If you have success-based target numbers per die, it's hard to intuit what the odds actually are. It gets worse when the same system will sometimes change the number of dice you roll and change the target number you need and you change the number of successes you need.

What's the difference between 3d6 vs TN 4+ and 5d6 vs TN 5+? What if you need 1 success? 2 successes? 3? Which is harder when? It's not even trivial to figure out.

#### loverdrive

##### Prophet of the profane (She/Her)
Comparisons of numbers are part of the Mathematics Curriculum.
It's not notably faster than addition.
Rolling dice in BitD doesn't really require comparison, though, just basic recognition. You don't need to compare 6 to 4, you just know that 6 is more than 4, and you know what 6 looks like.

#### aramis erak

##### Legend
Rolling dice in BitD doesn't really require comparison, though, just basic recognition. You don't need to compare 6 to 4, you just know that 6 is more than 4, and you know what 6 looks like.
And that action is still a comparison.

#### aramis erak

##### Legend
For me, I found capping the dice pools at 6 (with a 10 hard cap), and not doing the counting of successes thing, makes for a pretty quick and sleek approach
Most of my players have found counting successes faster than adding the dice together. Especially when using fixed TN for successes.
It's not clear which mode of dice pool you're using (other than #4, 5, 6 or 7)

See, with handfuls of dice, there are a number of ways to count...
1. Roll and Total: WEG d6 (incl. Star Wars), Space 1889, Deadlands (original system)
2. Roll and count some: EABA, L5R 5e
3. Roll and keep highest: LUG Trek², LUG Dune², Silhouette System (Heavy Gear, Jovian Chronicles, etc)²,
4. Roll and check dice individually for successes vs fixed TN: Storyteller/WoD (almost all TT versions¹)
5. Roll and check dice individually against a variable TN: VTM1.0, Shadowrun, Prime Directive, Arrowflight 1e, Modiphius' 2d20 success dice
6. Roll and count symbols: FUDGE, FATE, FFG Star Wars/Genesys. Modiphius 2d20 effect dice. With the optional dice
7. Roll, select some, count the symbols on the selected: FFG L5R 5e
8. Roll, select sets: LUG Trek & Dune², Silhouette system², One Roll Engine (Reign, Dirty World, Monsters & Other Childish Things)
9. Roll and match specific numbers: Orkworld³

¹: Excepting VTM 1E and the GURPS versions.
²: One of them, multiple 6's count as number of 6's +5; the other does similar for all faces. Don't recall which is which.
³: Roll, keep highest is the mode for skills, roll, match specific is for the armor/damage system.

#### Bedrockgames

##### Legend
Most of my players have found counting successes faster than adding the dice together. Especially when using fixed TN for successes.
It's not clear which mode of dice pool you're using (other than #4, 5, 6 or 7)

See, with handfuls of dice, there are a number of ways to count...
1. Roll and Total: WEG d6 (incl. Star Wars), Space 1889, Deadlands (original system)
2. Roll and count some: EABA, L5R 5e
3. Roll and keep highest: LUG Trek², LUG Dune², Silhouette System (Heavy Gear, Jovian Chronicles, etc)²,
4. Roll and check dice individually for successes vs fixed TN: Storyteller/WoD (almost all TT versions¹)
5. Roll and check dice individually against a variable TN: VTM1.0, Shadowrun, Prime Directive, Arrowflight 1e, Modiphius' 2d20 success dice
6. Roll and count symbols: FUDGE, FATE, FFG Star Wars/Genesys. Modiphius 2d20 effect dice. With the optional dice
7. Roll, select some, count the symbols on the selected: FFG L5R 5e
8. Roll, select sets: LUG Trek & Dune², Silhouette system², One Roll Engine (Reign, Dirty World, Monsters & Other Childish Things)
9. Roll and match specific numbers: Orkworld³

¹: Excepting VTM 1E and the GURPS versions.
²: One of them, multiple 6's count as number of 6's +5; the other does similar for all faces. Don't recall which is which.
³: Roll, keep highest is the mode for skills, roll, match specific is for the armor/damage system.

It is a roll and keep the highest approach for the most part. I use d10s, take the single highest roll and use that against a static target number. The only time you might have to count is when you get 10s on damage or when you do what we call an open damage roll (every result that equals or exceeds the target number is a wound....but that is rare).

#### loverdrive

##### Prophet of the profane (She/Her)
And that action is still a comparison.
Let's say a system uses custom D6s, with one Circle, two Squares and three Crosses.
1. If you roll at least one Circle, it's a full success.
2. If you roll at least one Square, but no Circles, it's a success at a cost.
3. Otherwise it's a failure.
Is this a comparison?

Replies
23
Views
2K
Replies
13
Views
5K
Replies
3
Views
1K
Replies
72
Views
6K
Replies
14
Views
5K