Player-facing rules, and what they move

Putting the roll in the player's hands changes who carries the burden of knowing the system.

A character sheet and pencil on a table, close — A sheet and a pencil: under a player-facing system this is the whole probability model, in public.

When every roll belongs to the player, the referee stops being a probability engine — and the design question becomes what exactly fills that gap.

The mechanical shift is also an authority shift

A conventional resolution system distributes labour unevenly. The referee rolls dice, reads results, and translates them privately into fiction: the orc hits or misses, the lock holds or opens, the guards are suspicious or indifferent. Players state intentions and wait for verdicts. This is not an accident of fashion — it descends from the wargame umpire tradition, in which one person adjudicated what the rules did not cover and what the dice produced. When Dave Arneson adapted that role for Blackmoor, the adjudicatory function came with it wholesale.

Player-facing resolution inverts the distribution. The player rolls to attack, to defend, to detect, to persuade — and the referee, in the most committed versions of the pattern, rolls very little or nothing at all. The mechanical labour moves across the table. What is less obvious, because it operates a layer down, is that epistemic authority moves with it.

Consider what the referee knows when they roll behind a screen. The probability model is private. A player can observe that the referee picked up two dice, but cannot read the spread of outcomes, cannot see what number came up, and relies entirely on the referee's narration to know what happened. Even an honest referee is — structurally, not personally — a bottleneck. Every result is filtered through one person's interpretation before it reaches the fiction.

A character sheet with attribute boxes filled in pencil — The target is already on the sheet, so the player brings the difficulty to the table with them.

When a player rolls, the result is public. The number is visible. The difficulty threshold is visible — or, in systems that omit it, the player calculates against a stat they already carry on their sheet. The gap between roll and outcome is narrow and inspectable. This is not merely a transparency gain; it changes who is required to know the rules. Under a referee-mediated system, a player can sit at the table with only a partial understanding of how resolution works, because the referee handles the machinery. Under a fully player-facing system, the player must understand the system well enough to read their own results — which is a design commitment, not a neutral convenience.

What moves when the roll moves

The oldest form of player-facing resolution in the dungeon game idiom is roll-under: the player rolls dice and tries to score beneath a stat on their own sheet. Its appeal is precisely that no external number is required. The difficulty is already owned. The player is not waiting to be told a target; they bring the target to the table. Systems built on this principle — including early editions of Call of Cthulhu, where the mechanic has been present since the game's 1981 debut — hand players a stable, legible model that belongs to the character rather than to the scenario.

The move system codified in Apocalypse World, designed by D. Vincent Baker and Meguey Baker and published in 2010, takes a more structured approach. Every move begins with a player action: "when you do X, roll." The referee does not initiate rolls; they respond to results. This is significant because it means the referee's function shifts from probability engine to consequence engine. The referee's creativity is channelled into "what happens next" rather than "did this succeed." In the Apocalypse World text, the MC makes a move when everyone looks to them to see what happens, when a player gives them an opening, or when a player rolls a miss. The MC's hand is forced by the player's dice, not the other way round.

This is an authority claim with consequences for table culture. When the referee's move is a mechanical response rather than a unilateral decision, players have a stronger purchase on the frame: they can see the causal chain from their roll to the narrative event. Whether that purchase translates into more equitable creative participation or simply into a more legible power structure depends on other design variables — the framing of playbooks, the severity of hard moves — but the structural lever is real.

A dice pool that counts successes, as in the Storyteller system developed by White Wolf in the early 1990s, distributes probability differently but maintains player-facing resolution in the same sense: the player rolls, counts, and knows. The referee may set a difficulty number, but the arithmetic happens on the player's side of the table. Crucially, dice pools tend to make the shape of probability more opaque even as they make results more legible: a player can read the outcome of ten dice instantly, but modelling what adding two dice to the pool actually does to expected successes requires genuine statistical intuition that few players carry.

The residue the referee still holds

Fully player-facing systems do not eliminate referee authority; they relocate and reshape it. What survives — and this is the design problem that any player-facing system must solve — is framing authority: the referee still decides what situation the player is rolling into.

If the referee controls framing tightly, player-facing resolution can become a theatre of agency: the player rolls, the result is public and inspectable, and yet the referee's choices about stakes, about what counts as a trigger for a roll, about how results translate into fictional positions, still determine the texture of play. The OSR argument about rulings-over-rules positions the referee's improvisational judgment as a feature rather than a deficiency; a player-facing roll doesn't neutralise that judgment, it simply strips one layer of it away.

The reaction roll is an instructive case: it is structurally player-facing in the sense that it replaces a referee decision with a dice result, but the referee still decides when to roll it, what the fictional result means, and how to narrate the disposition of an encountered group. The dice remove one kind of bias — the referee's in-the-moment inclination toward a given outcome — while leaving others intact. This is not a flaw in the mechanism; it is what mechanisms do. They constrain some decisions and leave others open.

The design question for any system that moves rolls to the player side is therefore not only "what does the player now control?" but "what has the referee been quietly released from?" When a player rolls and fails, the referee is freed from having to decide to make things difficult: the dice decided. This is load-bearing. Referee decision fatigue is real, and player-facing resolution is one procedural answer to it, distributing the cognitive and narrative work that would otherwise accumulate behind a single person and a screen. What fills the gap the referee vacates — player creativity, move triggers, oracle tables, hard-coded consequences — is where each system's argument about the purpose of play actually lives.

A game master's screen seen from behind with notes taped inside — Behind the screen the probability model is private. When the player rolls, it stops being.