Friday, 24 May 2013
Ah, X-Box Carcassonne, you is so pretty. But why you hurt me so?
Let's leave aside the question of whose bright idea it was to emulate a turn-based game and require each player to use a separate controller, and get to the real meat of the problem; the games utter inability to even plausibly mimic random tile allocation.
For those who haven't played this game before, the idea is that each player in turn is given a tile, on which can be one or more of three features: part of a city, a length of road, or a cloister. The tile is then placed next to the tiles already on the board, so that e.g. a city piece is adjacent to another city piece. Completing cities and roads gets you points; cloisters give you points when you surround them, monks being agoraphobics, presumably. You can also place farmers in the fields, who get points if they have nearby cities to sell corn to.
It's all pretty simple, but it rather relies on the idea that each player has the same chance (more or less; the tiles don't always quite divide equally) of getting each tile. Specifically, cloisters are often the most sought after items, because they can be quite valuable, and you can sort of leave them to themselves whilst you build other structures around them.
In my last three River games against four AI players, I have received precisely zero cloisters of the seven per game available. The chances of this are less than 2%. In my last game I came second, losing to an AI by 19 points. 26 of his points came from the three cloisters he acquired, of the seven available to five players. The chances of this are less than 15%. Yes, I confess that falls short of significant, but I figure if I whine about this now, I'll either collect more evidence in later games, or the universe will try to spite me by proving me wrong, and actually allow me to win a gorram game.
Those of you who suspect I'm simply applying basic statistics to a case of sore losing may not be entirely devoid of a point. This doesn't mean you can't all piss off, though.