Posted on

A Model for Decision Making in Games, Part Two: Action Execution

Last month I posited a model for decision making in games that divides decisions into three parts: action selection, action execution, and action resolution. That article also discussed a variety of popular methods for choosing actions. However, that’s just the start: once you’ve chosen an action, what do you do with it?

Phase 2: Action Execution

The action-selection phase of decision making took a broad view of player actions, asking the question, “What category of actions does the player take?” However, even after a player has selected a type of action, there’s often more choice to make: how exactly does the player implement (“execute”) the action, within the scope of the current game position?

Because Anglo-American games tend to degenerate the action-selection phase, often having just one type of action (move a chess piece; move a war-game unit; play a card; decide whether to buy a property), they tend to place most of their choice in the action-execution phase. In contrast, eurogames might have a fully featured action-execution phase, or they might have a degenerated action-execution phase where there’s no choice once the initial action selection is done. One of the reason that some people don’t like worker-placement games is that the action-execution phase can be degenerated: once you’ve decided to take a specific resource, translate a specific resource, or take some other very specific action, you often don’t make many additional decisions. (But sometimes you do: the build-fences action in Agricola offers a fine example of worker-placement action selection leading to “spatial choice” action execution.)

There are two broad ways to look at action execution: as a simple and abstract choice or as a decision that links the front-end decision mechanic to the back-end game mechanic.

Model 2a: Execution Choices

The core question of an action execution is obviously “How do you apply the category of action that you’ve selected?” Often, there are multiple choices. In Tigris & Euphrates (having chosen to place a tile from your hand) you must decide both which tile to place and where to place it; while in The Settlers of Catan (having chosen to build something from the menu) you must decide both where to place it and how to pay for it.

There are a number of common sorts of execution choice:

Construction Choice. A player can choose one of several things to build. Though these building choices might have been part of the initial action selection, as was the case in The Settlers of Catan, they can also be individual execution choices, such as in Puerto Rico, where you take the Builder role, then decide what to build. This is closely related to Purchase Choice but usually has costs in resources and is themed as literal building.

Exchange Choice. A player can choose one of several things to exchange and/or they can choose one of several things to trade for. A trading game like The Settlers of Catan offers very freeform exchange choices, as player swap` resources in a free-wheeling way. However, exchange choices can also come about as a single player’s choice of what to do with a resource-management engine (such as in Century: Spice Road where players swap one resource for another) or in a formulaic goal-completion game (such as in Splendor or, again, Century: Spice Road, where players must exchange specific resources for victory points).

Expenditure Choice. A player can decide which resources to expend. This can crossover with Token Choice (if they’re expending a token) or Exchange Choice (if they’re swapping the resource for something else) or Purchase Choice or Valuation Choice (if they’re just expending currency).

Obstacle Choice. A player can choose a specific obstacle to try and overcome. For example in Near and Far a spatial choice is often dependent upon an obstacle choice: whether a player is willing to face  bandit on the road.

Opponent Choice. A player can choose a specific opponent to attack or duel in some way. This is the most common in take-that games where players directly play cards that affect other opponents or in wargames (though there, the opponent choice is often subsumed by a region choice). Because of the problems that naked aggression can cause, some games like Cosmic Encounter and Epic Spell Wars instead regulate opponent choice through random card draws or dice rolls.

Purchase Choice. A player can choose a specific thing to purchase with their in-game currency. This might be a simple purchase with a currency-themed resources, such as when a player spends Item Points to purchase equipment in D-Day Dice. More frequently, it’s a choice to bid on a specific item in an auction game, which links to a Valuation Choice. When a purchase is instead made with resources and is building-themed, that’s a Construction Choice.

Region Choice. A player can choose a general region on a board. He might move troops into the region in Risk or airdrop majority-control markers in El Grande. Alternatively, he might link a Construction Choice to a Region Choice by deciding where he’s going to build something. This is usually a subcategory of Spatial Choice.

Spatial Choice. A player can choose any space on a board to do something. This is usually more variable than a Region Choice because the options are more open, like deciding where to place a road in The Settlers of Catan or where to build a railroad track in Empire Builder.

Token Choice. A player decides between one of several tokens. This could be closely linked to a Construction Choice as in The Settlers of Catan where a player decides which Tokens (resources) to use in order to build, but it could also be the choice of a card to play, a tile to place, or a resource to expend.

Valuation Choice. A player decides the valuation of something, usually by making a bid for it or offering it for sale at a certain price.

Model 2b: Mechanic Models

Truthfully, though, players and designers don’t often think about execution choices at this granular of a level. That’s because the choices are often predefined by the type of mechanics that are being used in the back end of the game.

These back-end mechanics are the ones at the heart of a game’s design: the cogs and gears that define how actions are executed and what that execution does to the various components of the game. They include mechanics like auctions, majority control, resource management, and adventure elements. Whereas front-end decision-making mechanics like simultaneous selection, deck building, and card drafting are usually abstract and bloodless, back-end mechanics are more likely to be evocative and colorful.

This idea of front-end and back-end mechanics cleanly fits into the model for decision-making: front-end mechanics are focused on action selection, while back-end mechanics are focused on action resolution. Action execution thus forms the interface between the two.

Figure 5. The two sides of mechanics.

Of course, back-end mechanics might be degenerated too, particularly if one of the more advanced systems is used for action selection. Dominion theoretically has a pair of back-end mechanics: card execution (which is pretty simple reading of text) and race-track scoring, but neither is that notable. However deckbuilding wargames like Asgard’s ChosenA Few Acres of Snow, and Tyrants of the Underdark all show how games with deckbuilding front-ends can still have fully featured game back-ends — in this case using either area-control or majority-control mechanics. It’s generally possible for a game to have two sets of fully featured mechanics, supporting the beginning and the end of the decision making process.

There are obviously many dozens of different back-end mechanics available in games, but following are some of the most popular eurogame mechanics, with examples of how they link to more abstract execution choices. These links are the most common: a great way to extend an existing mechanical category is to incorporate less common execution choices.

Adventure Games. Roleplaying derived board games can cover a lot of ground, but they tend to be about individual characters questing around some area, fighting adversaries, and recovering loot, in service to some larger goal. Common Execution Choices: Obstacle Choice (what monster to fight), Spatial Choice (where to move); Token Choice (what powers to use).

Area Control. Classic wargames focus on players fighting each other to control territory. Common Execution Choices: Construction Choice (what troops to build); Region Choice (where to move troops).

Auction. The classic eurogame category of auctions is simply about paying some amount to get something. Common Execution Choices: Purchase Choice (what to buy); Valuation Choice (what to pay).

City Building. Although city-building games are often a subset of Resource Management games, they also have unique design elements, focused on building out grids of interconnected neighborhoods. Common Execution Choices: Purchase Choice (what neighborhood to buy); Spatial Choice (where to place a tile); Token Choice (which tile to place).

Civilization. Civilization games are another pretty big category, but they tend to have four major elements: resource-management, trade, technology advance, and warfare. As such you can look at the Area Control, Resource Management, and Trade categories for some of the numerous choices available in this sort of game, while the technology advancement gameplay has a view execution decisions of its own. Common Execution Choices: Construction Choice (which technology to advance), Expenditure Choice (how to pay for technologies).

Combat. Although many combat games are Area Control, you can also have personal combat games such as Magic: The Gathering or Epic Spell Wars where you’re fighting a more personal war against your opponents. Common Execution Choices: Opponent Choice (who to attack); Token Choice (what power to use).

Connection Games. The train game is the best-known sort of connection game, but the category can include anything where you’re trying to link up things on the board, including pipe games and more abstract classics like Twixt and PÜNCTCommon Execution Choices: Purchase Choice (what connections to buy); Spatial Choice (where to place connections).

Majority Control.The euro-take on Area Control usually involves markers being placed in a region, then ownership being determined by who has the most markers there. Unsurprisingly, its choices are a lot like those in Area Control, except there tends to be less of a concept of troop differentiation. Common Execution Choices: Region Choice (where to place troops).

Resource Management. Perhaps the most popular category of euro-design, resource management involves players generating resources, transforrming them, and ultimately using them to build things that generate special powers, victory points, or both. Common Execution Choices: Construction Choice (what to build); Exchange Choice (what resources to swap); Expenditure Choice (what resources to spend); Spatial Choice (where to build).

Trading. Another classic and simple mechanic, trading simply involves swapping things with other players or the game system. There are games that are entirely trading, such as Res Publica, but more often it’s part of a larger game, such as The Settlers of CatanCommon Expenditure Choices: Exchange Choices (what to swap); Opponent Choice (who to swap with).


There are certainly other ways that you could classify the player choices that occur within action execution and there are certainly other execution options for the mechanical styles listed here. There are so many different options because action execution represents the guts of decision making in games, where there’s the opportunity to have really far-ranging choices.

However, it’s important to remember that action execution is just the middle part of what can potentially be a three-part model for decisions: first, players decide upon a category of action, possibly using a sophisticated front-end choice system; then they execute that action, probably making decisions defined by the needs of a back-end game system. This all leads to the third phase, where the action is actually resolved by that back-end system … which will be touched upon very briefly in the third and final article on this series, before looking at case studies of popular games under the microscope of this model.

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

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