Dice pools

01How the pool works

A dice pool turns the question inside out. Instead of rolling one die and checking whether the number lands above a threshold, you roll several dice and count how many of them individually clear a target number. Each die is a binary event — hit or miss — and success is a tally of those events, not a sum.

The core variables are pool size and target number. Make the target easier and more dice succeed; shrink the pool and fewer dice are in play. These two levers do different things to the probability curve even when they produce the same average hit count, and that asymmetry is what makes pools mechanically interesting.

A pool of six dice with a target of 5-or-higher on a d6 gives each die a one-in-three chance of succeeding. The expected number of hits is two. A pool of three dice with a target of 3-or-higher gives each die a two-in-three chance. The expected hits are still two. The averages match, but the distributions do not. The larger pool with the harder target is more spread out — more variance, more chance of zero hits or a full success, more drama. The smaller pool with the easier target clusters around two. Same expectation, different feel.

This is the designer's fundamental trade with pools: threshold and quantity are not interchangeable, even when the arithmetic makes them look equivalent. Adding dice at a fixed target is the same principle working in reverse: a larger pool compresses the distribution relative to its mean rather than expanding it. Adjust the target simultaneously and you can push variance in either direction.

02What grace looks like under pressure

The real argument for pools over single dice is how they handle penalties. When a character is wounded, exhausted, or working in bad conditions, a pool system typically removes dice. A single-die system usually applies a negative modifier instead — subtract two from your roll, or impose disadvantage.

Here is the difference. If you subtract a flat number from a d20, there is an absolute floor: once the penalty exceeds your bonus, you simply cannot succeed. The roll becomes theatre. A pool that shrinks never quite reaches zero in the same catastrophic way, because each remaining die still has its individual chance. At one die, you are badly compromised; you are not mechanically excluded. The probability degrades gracefully across the whole range of penalty rather than collapsing at a cliff edge.

This is also why minimum pool sizes matter in design. Most pool-based systems that strip dice under pressure build in a floor — one or two dice no matter how bad the situation — precisely because a pool of zero is not a probability event, it is a ruled impossibility. That floor is a policy decision about whether a character should ever have a guaranteed-failure outcome. The arithmetic does not make the decision; the designer does.

03The threshold as a setting dial

Because the target number controls individual die success rates, it doubles as a difficulty dial that sits outside the character's resources. A referee who raises the target from 4-or-higher to 5-or-higher has effectively cut the probability of any given die succeeding by a third — on a d6, from three-in-six to two-in-six — while leaving the player's pool size untouched. This makes difficulty adjustment feel qualitatively different from a flat modifier. The character's investment (their pool) stays intact; the environment becomes more resistant.

The trade-off is that threshold adjustments are coarse. On a d6, you have at most five meaningful thresholds. Move from one to the next and you shift success probability by a full sixth. For finer granularity, designers either increase die size — d10 pools give ten steps — or accept the coarseness and build narrative weight around it. Neither is wrong, but the choice defines how much precision the table can actually express, which is a real constraint the pool imposes on the whole system.