Posted on

A Model for Decision Making in Games, Part Three: Action Resolution

In recent months, I laid out a three-part model for decision-making in games, where you first choose an action with action selection, then specify an action with action execution. Now, I’m ready to talk in brief about the third part of the process, where you determine the results of an action with action resolution.

Phase 3: Action Resolution

Action resolution forms the back end of a decision, where the decision-making mechanics have been overtaken by the game mechanics. In other words, it’s not actually about the process of making decisions, but instead about how the decisions that a player makes are translated into results within the game’s core system—whether it be resource management, area control, majority control, or something else.

The possibilities laid out below are drawn in very broad strokes, as the specifics will often depend on how the action resolution integrates into the back-end game mechanic itself (something that will be discussed more in some case studies, in a fourth and final article in this series). As usual, action resolution is a phase that can be degenerated, in this case into “Automatic Resolution”. Eurogames are somewhat more likely to have a degenerated action resolution phase, while Anglo-American games, with their increased player conflict and their usage of adventure-game mechanics are somewhat more likely to have a fully considered action-resolution phase.

Automatic Resolution. Most game have a degenerated action-resolution phase: actions just resolve automatically and successfully. When you decide to place a tile in Carcassonne (selection) and when you decide where to place it (execution), there’s no question about whether the placement is successful or not (resolution). This is the most typical sort of action resolution in games of all sorts.

There are a number of different models beyond “Automatic Resolution”. Most games focus on one model, to reduce complexity in this core model, but it’s certainly possible to pick and choose between all the possibilities.

Figure 6: Elements of action resolution

Model 3a: Resolution Uncertainty

What’s the opposite of automatic resolution? Resolution uncertainty, where you don’t know whether an execution will complete as you expected or not. One way to introduce such uncertainty is through a simple game effect that ensures the results aren’t guaranteed.

Random Resolution. The most common sort of resolution uncertainty comes through randomness: someone rolls a die. This is most frequently seen in adventure games: you decide you’re going to take a specific action, you roll some number of dice, and you try to equal or exceed a target nurmber; alternatively, different results occur based on your sum total. Arkham HorrorBetrayal at House on the Hill, and Pathfinder Adventure Card Game all have mechanics of this sort. Random Resolution can also tie into Conflicted Resolution, where you’re facing off against another player. In Risk, for example, both players roll dice to determine which opposing armies are destroyed.

Arbitrary Resolution. Some games use cards rather than dice to determine their random results. The result is largely the same, except you may be able to count cards to determine if upcoming results are likely to be good or bad.

Unknown Futures. The other major way to introduce game-based uncertainty into action resolution is to have some unknown future: you start executing your action, but as you do something changes, and this might force you to change your plans. Take Ascension as an example. You begin executing your action by buying cards, but as you do, new cards will come out into the center row. You may then decide to purchase those new cards, instead of what you’d originally planned. (Technically, these new cards are an arbitrary element, but they don’t actually affect your resolution, but instead introduce uncertainty into how you’re going to continue executing.)

Model 3b: Resolution Conflicts

There’s another way to introduce uncertainty into resolution: through other people. Opponents can mess up your resolution either purposefully or accidentally.

Chaotic Resolution. Other players’ choices may impact yours in a number of ways. SImultaneous-selection games like Basari tend to epitomize this. Players make a secret choice (selection), but before they can execute they have to learn what other players did, which may force them to bid for the right to take their own action. Similarly, in 6 Nimmt! after the players all choose their cards to play (execution), the resolution is based on how those car choices interacted. Chaotic Resolution can even occur in a cooperative game if players are making simultaneous choices, as is the case in Space Alert where one player can use an elevator or energy needed by another player.

Conflicted Resolution. Things get more personal in Conflicted Resolution, where one player is trying to take an action directly opposed by another player. If the players roll dice against each other, this really falls back to Random Resolution, as was the case in Risk. However another possibility is for a player to have an option to stop the action which he may or may not use, because that option is a limited resource. Consider a Counter Magic spell in Magic: The Gathering. Every attack against the player holding that spell is a decision whose resolution is conflicted; the player with the Counter Magic has to choose which one he really wants to stop.

Decision Resolution. Opponents may also have the option to spoil a Resolution without it being a direct conflict. This is often the case in trading or negotiation games, where resolution depends directly upon the good will of other players: a player has to offer enough that his opponent will go along with his selected action. For example in Settlers of Catan, a player might decide to try and make a trade (selection) and he may then offer some specific resources he has in exchange for specific resources he wants (execution). In order for the trade to go through, another player must agree to the trade or even offer an alternative (resolution).

Model 3c: Resolution Delays

Uncertainty isn’t the only way to introduce complexity to decision resolution. You can also introduce factors that change its timing.

Delayed Resolution. An action is usually resolved immediately, but this doesn’t have to be the case. It’s possible for the resolution to occur after some number of turns, such as the way that your resources are delayed in Macao.

Incomplete Resolution. More common is the idea of incomplete resolution. Orléans offers a great example: you can choose to take an action (selection) and choose to place some of the character disks for that action (execution), but if you don’t place all of the disks, the action can be deferred until a later turn when you complete the resolution. (In fact, the resolution can even be deferred if an action is otherwise ready to go!)


In some ways, Action Resolution as the simplest part of my decision-making model. But, that’s because this part of the model is so abstract. This article details ways that you can model action resolution at a high level, but its complexities are ultimately integrated with a game’s back-end engine, and that’s where a lot of the nuances live.

This ends my model for decision making in games, but I’m going to revisit it in the near future to show how it fits together, using some case studies of well-known games.

Liked it? Take a second to support Shannon Appelcline on Patreon!

The original article can be found on the great Mechanics & Meeples