Mid-Game Structures – Player Strategy

Written by Alex Harkey

Earlier this year we launched our Mid-Game Structures series; a few perspectives of how games change their environments to keep the experience engaging. This article is part of our ongoing series titled Game Structures in which we continue to build on a foundation of game design concepts.

Building on Concepts: A Quick Review of Player Interaction

When we last left off, Matt looked at Player Interaction in games, which used a multi-category approach to help define the type and significance of interaction between players:

thedegreeofinteractionDegree of Interaction: This is an assessment of the degree of overlap or intensity players have with one another within the game environment. It creates spectrum ranging from No Interaction (which is akin to individual players quietly working on puzzles in opposite corners of a room) to Direction Interaction (which would be more like if the players needed to steal pieces from one another in order to finish their own puzzle).

Our topic today is Player Strategies and we’ll be pairing this axis with a second criteria to help us categorize how games influence the starting strategies of players. Let’s look at that second criteria:

What is a player’s time horizon to plan actions?

The next thing we want to examine is how short or long-term oriented player decisions tend to be in a given game, or what we call our Time Horizon. If you’re more comfortable with it, you can think of this as our comparison of Tactics vs Strategy; strategy being the large scale focus or objective(s) needed to achieve success, and tactics being the specific steps or tasks you need to perform to implement your strategy.

I’m deliberately going to gravitate away from using these terms, because we can’t really plot games using a Tactics vs Strategy scale; games can have a significant presence of each and they don’t necessarily come at the cost of one another. These terms are also very player-centric, while we we want to classify in relation to the game as a whole: how players respond to one another, deal with variance and foresee future needs. We’ll group these and several other factors together and categorize games based on their time horizon in which players plan decisions; a continuous theoretical spectrum.

Time Horizon

Time Horizon: How far in advance can you plan your approach?

In this spectrum, you can think of “short-term” as games in which typically make decisions on a turn-by-turn basis; perhaps players have relatively little information available between turns, or there is a tremendous need to be reactive to a rapidly changing board, other players or new information that becomes available on their turn.

Love LetterGames like Love Letter and Yahtzee have a short-term time horizon in which players can plan ahead. These games consist of drawing cards or rolling dice at the beginning of your turn, then basing your decisions off of the resulting outcome. Player’s simply can’t anticipate what their precise decisions might be four or five turns ahead, as the possibilities are too great to garner much benefit.

This isn’t to say either game lacks long-term decisions; You may plan to hold a certain card in Love Letter as long as possible and your choice of where to score in Yahtzee can be a strong contributing factor to success. But as a whole, these games are structured for a short-term, turn-by-turn basis of making decisions rather than stringing together multiple turns in a row into a more intricate plan.

On the other end of our spectrum, “long-term” describes games in which players will be rewarded for using a proactive approach to their turns. In this case, we aim to coordinate our turns or choices in a way to maximize our final outcome. These games also allow players to have the vast majority of the information we need to plan ahead before we ever get to our next turn, so you have the ability to mastermind that multi-step ladder to success.

Twilight StruggleGames like Concordia and Twilight Struggle reward player decisions on a long-term time horizon. In each game, a player has a hand of cards that they will play one-per-turn in some order. The surest path to success is to prioritize the relative order of cards played to achieve the greatest total outcome. Obviously, not as easy as I make it sound, but as a comparison, someone who is playing one card at a time without considering their next few turns can miss opportunities and beneficial combinations that the cards offer when played in a specific sequence.

Putting It All Together

We now have our two axis graph: Player Interaction (High vs Low) and Time Horizon (Short vs Long). Let’s take a moment and identify a few examples before we move forward.


We can see a few examples whose characteristics would line up somewhere in each quadrant of the graph. For instance, Chess (Quandrant I) rewards long-term planning the ability to anticipate many turns ahead in competitive environments. In addition, you are capturing the pieces of your opponent in a tightly constrained space, a fitting example of high player interaction.

Bohnanza (Quadrant II) also has high player interaction as the game drives negotiation and trading between players. Turns generally can’t be anticipated as players are constantly flipping over new cards and information, opponents are trying to make deals and you may be planting beans only to immediately pull them up as your objectives quickly change.

Yahtzee (Quadrant III) is a game of short-term planning; the outcome of your dice each roll strongly determine your course of action each turn. While you can play against competition, you have no opportunity to affect the approach your opponents use in the game other than to distract them with your incredibly good looks.

Dominion (Quadrant IV) is a prototypical example of a game with a long-term time horizon. There is a consequence to players who are not considering the synergy of the cards purchased between turns, as deck efficiency and composition are two driving factors of success. While players do have access to cards that can affect opponents, the vast majority of the game takes place independent from other players, so it is often quite low in terms of player interaction.

Mapping Player Strategy Categories

The quadrants we’ve created (unsurprisingly) have some common traits between the population of games within them. Our approach focuses on three broad strategic categories: “Initially Reactive“, “Continuously Reactive” and “Proactive“. Also included are a few game mechanic categories that are often concentrated (but not exclusive) to certain areas of our graph.


Strategy Category – Initially Reactive

Initially ReactiveGames that fall within the Initially Reactive area often give players different sets of initial conditions which can provide strong direction as to which paths players should or should not pursue.

These starting conditions can appear in the form of your opening destination tickets in Ticket to Ride, your lord card in Lords of Waterdeep or the country you begin the game with in Diplomacy. In these games players are initially rewarded for “doing what the game tells them to do” based on their unique situation and by incorporating a game-endorsed strategy. There is rarely a single “correct” strategy applicable to all players, but there may be an optimal way to begin the game from your position.

7 WondersThe replay value in these games often comes from tapping into a player’s creativity and the need to adapt to throw something together on the fly. While you can have an idea The intrinsic reward of games in this category like Terra Mystica and 7 Wonders is that each game begins with vastly different starting conditions and there is enjoyment in solving the puzzle of how to be successful in any of them.

Strategy Category – Continuously Reactive

Continuously ReactiveThe Continuously Reactive category consists of games where the set of initial conditions in the game matter less than the conditions that exist on each subsequent turn, due to a large degree of variance. Since so much can change between your last turn and your next turn, players are rewarded for maximizing the value on each of their turns, since a long-term strategy is often not feasible.

The setting for these games might be the constant inflow of new cards and information in Love Letter, the fluctuating stock-market of Speculation or the inability to anticipate your options in Five Tribes until it actually becomes your turn. What you do on your turn in Hanabi may be dictated based on a clue given by the previous player and you can’t foresee some of your decisions in Bohnanza because they can arise unexpectedly during someone else’s turn. All of these might be reasons why you need to be on your toes constantly and watching what other players are doing on their turns.

carcassonne2One of the common traits in this category is that these games are often very approachable for new players. The highly tactical nature of these games is beneficial to a new player who may not have the ins-and-outs of the game quite yet, as the first few turns of a game like Carcassonne can serve as a quick introduction to the sequence of ideas that are repeated throughout. Since a long-term strategy isn’t always feasible in this category, a new player who has experience with similar mechanisms can try to wing it and come out alright at the end.

Replay value in this category comes entirely from the variance and the idea that you’ll probably never going to see anything close to the same game play out twice. There are generally a lot of moving parts; perhaps randomness generated by cards or dice and additional variance caused by the reactions of other players to these variables. One benefit of all these factors is that it is quite difficult to become estranged from what other players are doing, as one player’s actions might completely shuffle the priorities of all the other players on their following turns.

Strategy Category – Proactive

ProactiveThe Proactive category consists of games where each player has the same set of initial conditions and the board is relatively constant throughout the game. Players are rewarded for having a consistent strategy that they execute throughout the game and there are a limited number of optimal approaches for any given board configuration.

In games like Dominion and Puerto Rico, you’re at a tremendous advantage by planning a strategy based on the initial set-up before the game even begins. To some extent, the success of your strategy may swing based on the actions of your opponents, but players who can quickly prioritize and consistently apply a strategy will have more success long-term than those who dabble in all areas of the game.

DominionA lot of the replay value in this category comes from these games being very structured and almost methodical in execution. These are many of the “Gamer’s Games” that should almost be renamed “Expert’s Games” because they reward the experience of a player who can execute an well-orchestrated big money deck in Dominion or a player who can spot an opportunity for a double build strategy during In The Year of the Dragon. Dedicated enthusiasts of these games will find a very strong replay value as they hold the knowledge of the various strategies and the experience to maneuver the hurdles that arise in their execution.

There is a bit of a correlation we can see in the games of this category; the less interactive a game is, the more long-term oriented a game often is. The player interaction of these games is frequently of the “indirect interaction” nature, so while you might consider the types and counts of cards your opponent is building in San Juan, their decisions are usually not deeply intrusive to your personal goals.

The Last Category – Multi-Player Solitaire

multiplayersolitaireI thought our the last category would be worth addressing even if it isn’t one of our strategy categories. This category of games is really the quintessential group to be called “multi-player solitaire”. Critics frequently use the term as a pejorative for “anything with less direct competition than I prefer in my games”, but that doesn’t really serve us well as a useful function for classifying games.

When we look at games in this quadrant, they have the characteristics of being very short-term oriented and low in player interaction. As a dramatic example, Bingo would end up in this quadrant as you really don’t do anything until a number is called and you can’t interfere with any of the other players during the game.

YahtzeeThese games often have a large social component; you can play Yahtzee or participate in traditional trivia games without having to be at odds with other players or even be concerned with their choices. As a trade-off, the lighter nature of competition appeals to our ability to socialize between turns and invest in one another rather than the game we are playing.


Game Design StructuresAs we’ve been graphing, developing and revising this categorization approach, we’ve continued to find fascinating observations beyond what I could cram into a single article. Sometime ago Matt and I each developed different category names and slightly different categorical descriptions for this whole idea, but I harmonized our thoughts last year and I present it to you in the format above.

Now we want to know what you think; do you see any trends in the games you enjoy or where would you graph the games you’re designing? Do you disagree with this approach entirely? Join us in the comments below.

This wraps our Mid-Game Structures series and we’ll be back with soon with our next series. Thank you for reading!


15 comments on “Mid-Game Structures – Player Strategy

  1. Joe Dickinson

    Just read through the early and mid-game structure articles. What you wrote in these was more practical and applicable than almost every book I’ve read on game design so far. Thank you for writing them. I look forward to seeing how these affect any of my future designs, and I also greatly look forward to the end game series of articles!

    1. Alex Harkey

      Awww shucks, you’re making me blush Joe.

      Thanks so much, it means so much to us to read your comment.

  2. Jack

    Why is the “Take That” circle in short-term planning? You’ve put Chess in long-term planning, and it revolves around take that, as it’s defined on BGG.

    “Maneuvers that directly attack an opposing player’s strength, level, life points or do something else to impede their progress, while usually providing the main engine for player interaction in the game.”

    1. Alex Harkey

      That’s a really interesting point, Jack – I never thought about it that way.

      I suppose mechanics are truly a matter of perspective, but I wouldn’t have considered Chess a prototypical example of the “Take That!” mechanic. I don’t believe the BGG database has Chess tagged as a “Take That!” game either.

      However, your observation is spot on as players do directly interfere with one another and remove opposing pieces from the board regularly. I’d argue the prototypical “Take That!” mechanic shows up frequently in card games – In Bang!, Coup, Good Cop Bad Cop and even older games like Family Business all share similar characteristics where players can attack the life points of their opponents with card abilities or special actions.

      Munchkin uses “Take That!” as players can pile on to make the active player’s turn more difficult. Star Realms starts players with a number of hit points and the goal of the game is to reduce your opponent to zero hit points by eating away at that total turn after turn.

      Your main question is a great one – “Why is the Take That circle in short-term planning?”. My justification is that almost all of the above games have traits and decisions that rely on relatively short time horizons:

    2. Most of the cards you draw/acquire have a short lifespan. Players draw cards, players use them, they’re gone. Because the cards often drive your actions, your plans can purposely be fickle as a strong card draw can change your entire outlook on your next turn.
    3. Player actions are usually based on what everyone else at the table is doing. If you ask the winner at the end of any of these games what their strategy was on the first turn, the answers would be all over the place. These games do have strategy, but you largely can’t plan for possibilities ten turns in advance as premier chess players may be able to.
    4. In trying to group games by mechanic there will always be exceptions and that is precisely where this player strategy grid theory can fall short. It’s got room to grow and I’ll continue working on it, but I’ll always be the first to admit its faults.

      Your question is a great one, thanks for commenting!

  3. David B

    Interesting analysis. My own choices don’t favour any particular quadrant, which leaves me with little room for applying it.

    I would particularly disagree about your placement of Carcassonne. While it is “reactive” in the sense that you don’t know what each turn will hold, long term strategies are absolutely critical. Thus farming is usually vital to success. Building large cities takes time and planning, especially as expansions are added. Experience makes a huge difference as you learn what tiles are available and thus what possibilities remain for completion of various features. Finally, every player does start with an even footing of 7 meeples. Although this can change based on tile draw, distribution tends to be fairly even allowing for consistent strategies and approaches to gameplay.

    1. Alex Harkey

      These are excellent points David, and I can’t nor would I want to disagree with any of them. I’ve always found the meeple management in Carcassonne as a particularly interesting source of tension and long-term resource management.

      My best argument would maintain that if we plot Carcassonne on a theoretical graph against all the other games in consideration here (and beyond), it has a smaller window for strategic planning than say Chess, Power Grid, Dominion or Castles of Burgundy.

      My purpose isn’t to say Carcassonne is any less of a game because of this, or that it lacks extensive strategic value, but I’d have a hard time arguing I can work out a masterful plan for an entire game one tile at a time. The game requires mental agility and the ability to adapt from players – and skills I’d argue are as important as managing cash in Power Grid or timing the end game in Dominion.

      I find Carcassonne to be a fascinating game of risk management; “Should I claim that small farm so I can join it to the main farm on the board?”, “Is it worth it to put a meeple on this road yet?”, “What if I claim this castle with my last meeple – is it possible I’ll be stranded without a meeple in hand for the next few turns?”

      You’ve made a really great point and it’s probably one I’ll be thinking about for quite some time as I keep working on this player strategy topic. Thanks so much for commenting!

  4. Joseph E. Pilkus III


    Great article…I don’t always have the time to read them immediately, but I do enjoy every word of them. I’ve recently moved from solely providing assistance as a Developer to becoming a Designer (TAU CETI: Planetary Crisis), so I’m even more conscious of the myriad topics you’ve written about these past few years. The game, by my estimation is a proactive, highly interactive one…it’s what I generally play, so it should probably come as no surprise that I designed a game with a similar feel. Again, great article!


    1. Alex Harkey

      We were missing your presence, Joe.

      Great to hear from you and congrats on the design role transition. It sounds like a great game and I’ll be looking forward to playing Tau Ceti: Planetary Crisis the first chance I get.

      1. Joseph E. Pilkus III

        Thanks, Alex! Today, I’ll delve into an older one, entitled “Mechanic Archetypes – Pool Builders” ~ your work in this area, along with your colleagues should be “must read” articles for anyone new to the industry or anyone seriously contemplating working in the field of board games. Along with Jamey Stegmaier’s Kickstarter Blog which I recommend to others on the business side, I have recommended your articles to several designers.


  5. Peter

    I would love to see an expanded graphic showing where many games would fall in your 2D system. When I first saw it it was instantly clear to me why I love games like Twilight Struggle, Terra Mystica, and 7 Wonders. It all makes sense! I also tend to enjoy games in the fourth quadrant as well when I’m not feeling as competitive or am working with an already exhausted mind. I think this would be immensely useful for people to find new games they might find enjoyable – I know that I would certainly use it – I will definitely be picking up a copy of Power Grid and Puerto Rico now that I’ve seen this!

    1. Alex Harkey

      Thank you Peter. We’ve got a rather heavily populated version I’ve continued adding to over the last year, I think the count is at around 120 games on it right now. It’s a bit busy to look at, but perhaps I’ll circle back around to this topic and post it sometime in the future.

      I’m encouraged that it has resonated so well with you. I’ve gotten a lot of positive feedback but I do worry about the practical benefits for new players. In our case we can find games we like on this graph (or another) based on what we know, but it may just end up being one more point of research for an audience who is probably going to benefit more from qualitative feedback and WSIG threads on various forums.

      It’s also an audience that is still feeling out what they like in games, so if they really only like the card drafting in 7 Wonders or the network building in Power Grid, the graph may send them in the wrong direction. They may just want to explore more games that use those specific mechanics.

      It’s certainly a concept that has some intriguing potential and some practical limitations, but one that is a fun to think about.

  6. Anonymous

    Having a bit of a mathematical background, I really like the 2-axis space you all have created here. I will think about it more, but I have one initial challenge–what about games that are not stagnant in the space? Can a game go from short term planning to more long term planning as the game goes on, for example?

    My initial thought was Through the ages: at game open the turns are relatively tactical and players start out in more-or-less the same position. As people develop their civilizations, however, the long term strategy unfolds and the player interaction tends to deepen.

    Is there room for thinking about games that are dynamic in this schema?

    1. Alex Harkey

      Great thoughts and thank you for taking the time to comment –

      Games can certainly be dynamic using this idea, although that is one of the limitations of the approach we have here. Your example of Through the Ages is a wonderful one that demonstrates not all games are locked in place to their initial time horizon.

      Even a change as subtle as drawing two cards at the beginning of each turn versus at the end of each turn could adjust a game plotted on this graph. In the first situation you might have somewhere around 50-80% of the information you need, while in the second situation you can theoretically plan your turn in advance.

      As tabletop games continue to evolve, we may see more instances comparable to video games where players start in a very constrained set of options (i.e. Choosing one of the 3 starting Pokemon options) that eventually grow into more extensive strategic choices (i.e. building a strong, balanced team to take on the Elite Four at the end of the game.)

  7. Shao-Ying Chen

    Great article! I’m a boardgame gamer and designer in Taiwan, and I’d like to translate this article into Chinese to facilitate the understanding of boardgame designing principles for boardgame designers in Taiwan. Of course, the original author and the location of this article will be subscribed in the beginning of the translation article. May I have the pleasure to translate this article and publish on the Facebook fan page about boardgame designing and testing? Thanks!

    1. Alex Harkey

      Of course, Shao-Ying, and thank you for helping grow the game design community in Taiwan.

Contribute to the conversation, please leave us your thoughts and feedback