January 30th, 2007

games, MTGO, Niv-Mizzet

Notes towards the design of field-movement board games

A hobby of mine: inventing board games. I'm only recently starting to get any good at it; I'm happy with my most recent three designs, but all the ones I had before that basically suck. Most recently is Stacktics, an Icehouse game of variable complexity for two players. I've played it several times, and it's actually an interesting strategy game with a non-arbitrary feel to it- my other games have this annoying problem in which the strategy is non-obvious enough it's extremely difficult to figure out how to play the game reasonably even at a beginner's level and I'm still not convinced that these pure-strategy games I've made actually have any strategy. Stacktics is different; it has reasonable ways of evaluating how good a move is.

Of course, I'm very happy with it, but I can't take all the credit for it; it was heavily and directly inspired by Stack Chess, which I very much liked the idea of but found to be completely unplayable. I like the idea of a Chess-like game in which the stackability of Icehouse pieces is used to let pieces dynamically change their movement. But Stack Chess gave the pieces too much movement. It seemed that it was nearly impossible to pin anything down for a capture without putting your stack in equal danger. Chess is similar, except its trades tend to eventually lead to a material advantage when someone can't balance the trade or someone gets a better piece out of it; Stack Chess gave no such result. Because the stacks provided so much versatility, almost any stack could move in almost any way, meaning there was no way to attack a piece without being in danger from it.

I hadn't thought about things that way when I started developing Stacktics. What I wanted was a less chaotic Stack Chess, without losing any of the strategy it might have had. I wanted to encourage players to build stacks because it makes the game more interesting, but I had to make it hazardous, although losing the entire stack was a way to handle that; the real problem is that I didn't want stacks to dominate so completely that the game lost its strategy in a deadlock of unstoppable pieces that neither player wants to break because the only way to voluntarily break the deadlock is to lose.

I'm not sure how I hit on the "must be on top of a tower to attack" rule, but I'm very happy with how it turned out. Setting up stacks is the only way to attack, and the more stacked your setup is, the less movement it takes to prepare for the next attack- but putting everything in one stack is not a viable strategy because large stacks have trouble moving. The remarkable usefulness of a stacked pair of identical pieces was emergent, not intentional, but it turned out to work very well. Something that surprised me is how extremely strong just a stacked pair of Queens is in the game- it's the most versatile attack pattern there is, with the distinct disadvantage that losing both Queens at once (or, for that matter, either queen) sends you far on your way to losing the game and it's not so difficult to prevent your opponent from being able to build a stacked pair of Queens given the stacking rules in the first place.

What I learned from the game, as the most important change from Stack Chess: Chess-like board games are only interesting because pieces can't fully defend themselves. Blind spots are the only way for such games to work: it must be possible to position a piece such that it is threatening an attack without itself being attacked directly by its target. This requires pieces to move together, in careful formations, both for attack and defense; the strategy of the game comes in as disrupting your opponent's position to make captures with minimal losses of your own. A game with excessively strong pieces doesn't let that work, because pieces can always defend themselves, and it becomes prohibitively difficult for anybody but an expert who can see the entire chain of captures to actually launch a meaningful attack- and in many cases, it's not even possible to set up a meaningful chain of captures to gain a significant advantage.

As tempting as it may be to make all your pieces powerful because you know how fun your Queen is to use in Chess, it just doesn't make for a very interesting game.

But by the same token, it must be possible to cover your vulnerabilities, or the game swings too far in the other direction; instead of it being impossible to launch an attack without losing your pieces, it becomes impossible to block an attack, and the game generally goes to whoever is given the first move. It's a balance, and developing Stacktics showed me a lot about how that balance works.

Okay, off to class now. I have other stuff I need to post about, it's on my to-do list, I just haven't gotten around to it yet...
  • Current Music
    Jim's Big Ego - This Message