Sunday, July 5, 2009

Getting games wrong

I'm in the middle of Brandon Sanderson's new novel (so far: entertaining in a geeky way, which is what I looked forward too). And I hit a bit about a game, or rather a sport, which isn't very good. Not the writing, I mean; the game isn't very good.

Tarachin was a complex game, played only by the wealthy. Lightsong had never bothered to learn the rules.

He found it more amusing to play when he had no idea what he was doing.

It was his throw next. He stood up, selecting one of the wooden spheres from the rack because it matched the color of his drink. He tossed the orange sphere up and down; then -- not paying attention to where he was throwing -- he tossed it out onto the field. [...]

"Five hundred and seven points," the priest announced.

"Now you're just showing off," Truthcall said.

Lightsong said nothing. In his opinion, it revealed an inherent flaw in the game that the one who knew least about it tended to do the best.

-- from Warbreaker, Brandon Sanderson, chapter 22

Now, it is one of my long-standing grumps that authors aren't very good at inventing fictional games -- they tend to come up with games which are silly, broken, or chess.

So what, you say, game design is hard; authors have enough on their plates. Entirely true. (I said it was a grump, not a peeve. It's not justified enough to be a peeve.)

In Warbreaker we have a game which seems to be a cross between lawn bowling and Fizzbin. That's a plausible invention, given the story setting. And the narrator is, of course, correct -- if not knowing the rules gives you an advantage, the game is broken. But is it plausibly broken?

You can name plenty of games in which not knowing the rules is not a disadvantage... start with Candyland. We say these games are "random", or "have no strategy". But Tarachin isn't like that; Lightsong really does seem to win a lot. And this doesn't make sense. If it were advantageous to not know the rules, players would avoid learning them -- or just play without thinking about the rules too much.

(Compare Dragon Poker, from Bob Asprin's Little Myth Marker. In that story, Skeeve is faced with a game so complicated that his least bad strategy is to play randomly -- and end the game fast, so that the players who do know the rules have no chance to roll him up. But this is a far cry from Skeeve having an advantage over them!)

(By the way, I love that the Wikipedia entry for Dragon Poker says "See also: Double Fanucci, Mornington Crescent.")

Anyhow, as I say, Tarachin seems like a weird case. But then, weird cases do exist in real life. So I throw the question open to the teeming Gameshelf readership.

What games are out there in which trying to win is worse than playing randomly? How does that work? Maybe there are strategies which are tempting, but have a low chance of paying off? Maybe the game is popular among people who aren't analytic gamers? Maybe "winning" isn't the reason people play? Show off your corner cases, folks.

(Last parenthetical: I haven't finished Warbreaker, and it is a fantasy novel. So maybe there's some as-yet-unrevealed reason that Lightsong wins so often. If so, the author has set me up with a sneaky plot point -- which, to be sure, is one of Sanderson's trademarks as a writer. In which case, go him! But the question stands. No book spoilers in the comments, please.)

Comments imported from Gameshelf

dfan (Jul 5, 2009 at 4:01 PM):

Roshambo (Rock Paper Scissors) leaps to mind.

Denis Moskowitz (Jul 5, 2009 at 9:02 PM):

How about a game with a rare but devastating penalty? Players that know the rules will play conservatively to avoid it, while those who don't know the rules will make wild plays that score much better but have unknown-to-them risks. I know nothing about Tarachin, but imagine this rule: distance is good for your score but a player whose ball reaches the outer edge of the court immediately loses a finger. Lightsong will generally win but be unpleasantly surprised one of these days.

Alex (Jul 6, 2009 at 3:56 PM):

In Fluxx, trying to win can be worse than playing randomly. Also, since I can be very bad at figuring out what people are going to play, any game in which the whole game is guessing that I often do better by playing randomly. But that's not true of everyone.

I'd be interested in hearing your thoughts on The Player of Games by Iain Banks. When I read it (ages ago) the grand game he comes up with (vaguely like Go) felt very compelling to me.

Jason McIntosh (Jul 6, 2009 at 10:29 PM):

What dfan said. Playing randomly in RPS will completely nullify any cunning strategy your opponent might want to use, which makes it a pretty compelling technique!

No comments:

Post a Comment