Achieving 7 ± 2 Complexity
  • BrendanBrendan March 2009
    One of the things I most want to achieve in any game I design is a level of runtime complexity such that I personally can't keep track of all the factors that should influence my next decision. If I can look at the board and cards in play and form a decision statement like "if opponent plays X, I will take action Y; if opponent plays Z, I will take action A," then I quickly become impatient, even if I don't have the full gamestate in my head. I want to be forced to juggle tactical advantage against overall strategy at every turn. I want to be consistently having one factor or another pushed out of working memory by the mythical 7 ± 2 queue.

    Poker may be a good example of the classic principle here; it takes someone with a great deal of ability and a lot of rote math to be able to judge the odds and simultaneously play the other players' perceptions. I think poker-style bluffing is overused, though, largely because nobody takes it seriously in any game besides poker. Citadels (which has its faults) is closer to what I'm talking about: every turn demands that you make an unpredictable but useful choice of role, build something that gains you either victory points or in-game abilities, manage your gold supply and watch to see who's in the lead. I understand Puerto Rico gives some people the same satisfaction, although the few times I've played it, I felt too constrained in my choices.

    Am I expressing a concept you enjoy in games? If so, I'd like you to brainstorm in this thread: what are some good, reliable mechanical methods for generating this level of complexity while a) not being Magic and b) not depending too heavily on randomness as a factor?
  • Holly March 2009
    I suppose Citadels, Magic and others tend to do it at least partly through hidden information: different decisions would be the "right" one depending on what position your opponent is in, or what's going to happen to you next, but you aren't sure of any of these; you can only make a best guess based on your understanding of probability, psychology or something else. (I had one Citadels opponent who was so good at the psychology part that I would end up shuffling the cards I was considering and choosing one without looking at it.)

    I like this sort of thing in principle - I am lazy and can't usually be bothered to think through the implications of my decisions properly, so being forced not to do so by the complexity of the game itself is pretty nice. However, I also find it a bit alarming, because I would expect some players to try to take all elements into account, and take ages over their turns and/or be anxious about how they don't know what to do, both of behaviours which make me bored and stressed and have caused me to play board games less often now than I used to.

    Limiting the time available for a turn helps, I suppose, both with the "people take ages for their turn" thing and by preventing you from having time to marshal all the factors at your disposal in making a decision. Having several different ways to win (I like Rithmomachia's "little victory, medium-sized victory, big victory"; Attika's trail-of-temples versus building-whole-city). And thinking of Attika, tragedy-of-the-commons stuff, where you're better off doing your own thing, but if everyone does that then disaster ensues (where "disaster" means "Josh manages to join two temples"); having to balance the two makes it much less possible to have a clear best path.

    And then there are games where there isn't really a "make a decision", there's a "this game tests one skill, get better at that skill [you may wish to develop your own techniques to do this]". Set and Ricochet Robots, mostly. I love both of them, but understand why not everyone does; but there must be space to apply this sort of "you have to do X really well; feel free to figure out your own way to do so" to less mechanical games. (I suppose many negotiation games come under this heading as well.)
  • BrendanBrendan March 2009
    I had completely forgotten about Attika! I loved Attika! I am going to order my own copy of Attika right now.

    Posted By: HollyI am lazy and can't usually be bothered to think through the implications of my decisions properly, so being forced not to do so by the complexity of the game itself is pretty nice.

    Obviously this is a nontrivial part of my desire for complex games as well. When I lose at Attika, I can say "oh, that was a well-designed and well-played game!" Whereas if I lose at Tic-Tac-Toe, all I can do is silently flog myself.

    I think many, many games would benefit from a chess-style timer, although they're expensive, plus you'd need one that could expand to handle more players. I guess an egg timer would work. Also, the kind of people who need to be timed would likely refuse to play with it, and of course it's basically prioritizing my boredom over my opponents' desire to play thoughtfully. This is why I'm struggling to find a good way to eliminate turns altogether in the other thread; your Set / Ricochet Robots example actually does that quite nicely, except then you (well, I) play in constant terror of not getting any turns.
  • TristanTristan March 2009
    What about a simple sandclock? It might provide the needed time limit without feeling all that stressing...
  • Holly March 2009
    My experience is that the people who "need" a sandclock (by the standards of the lazily impatient) are not happy to play with one, and that even if they do then it's still not enough for the lazy impatient people, who sit there watching the seconds drain away. I suspect people at either extreme should just not play games with each other regularly, really.
  • BrendanBrendan March 2009
    Holly, don't be absurd! ALL SOCIAL PROBLEMS CAN BE FIXED BY CAREFUL DESIGN.
  • Wolfe March 2009
    A boardgame that I like that does something similar is Pente.

    As a point of literal fact, you *can* take everything into account. As a point of practice, as the game goes on and the patterns on the board become more complex, it becomes increasingly more complicated. Much of the late game is noticing your mistakes, hoping your opponent doesn't, and trying to notice his whilst planning your placements to win. Usually the winning tactic involves pressing to win in multiple places on the board simultaneously, and occasionally a win comes of one player or the other noticing a mistake the other player made but didn't catch in time to fix it.

    The game is designed for 2 players, but when I was growing up, I learned that it played as well with 3 or even 4 players. The winning strategies change depending on how many players there are, and the complexity grows when you add more players.
  • BrendanBrendan March 2009
    Wolfe, I think that's a great suggestion, although (as with go, chess, Terrace, etc) I tend to dislike pure strategy games at that level of abstraction. But that does provide a useful principle for the topic: one way to achieve high complexity factor without randomness is to induce fractal-style patterns on an increasingly large field of play. I'm willing to bet this would also ameliorate long-calculation turns.

    Another way to go might be a design that reduced the impact of any individual turn: no single action could win or lose you the game, only the cumulative output of many small decisions. Maybe you have a limited supply of "takebacks," although actually that might make things worse. I'm thinking here of an investment-style game where the bonuses you gained for buying into any given project could be scaled to reflect your stake: basically, a game that allowed for fine-grained risk management.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Apply for Membership