Exploiting your opponent's weaknesses
Posted By: Timothy Chow In Response To: Giants List - Candidates etc...(long) (David Wells)
Date: Tuesday, 24 January 2012, at 6:59 p.m.
In Response To: Giants List - Candidates etc...(long) (David Wells)
Whether or not backgammon is a "solved game" (and I would argue that it isn't) is irrelevant to whether studying someone's games can give you an edge over them. Checkers (8x8) is certainly solved in the strict mathematical sense, and the top players are very strong, but they're not perfect, and studying their weaknesses can give you an edge. Chess isn't solved, but the strength of the bots is comparable to that of backgammon, and yet again it's possible to study a specific opponent and exploit his or her weaknesses.
What I think makes it tough to do the same in backgammon is those stupid dice. They make it very hard for you to control the direction of the game. You might know that your opponent is extremely weak in prime-versus-prime games, but if the dice don't allow it, you can't steer the game in that direction. Of course, in any game, your opponent's moves will always interfere somewhat with your plans to steer the game, but in backgammon the dice add another huge layer of noise on top of that.
Nevertheless, I believe that it is possible to study your opponent's weaknesses and profit from that information, even in backgammon. First of all, it's important to note that if someone plays (for example) a 2.5 ER, that doesn't mean that every move that the player makes is within 0.0025 of the best move (according to the bot). There will be blunders. If you get a large enough sample, you'll be able to detect some patterns in those blunders. Secondly, cube decisions count for a potentially enormous amount of equity, and it's here that you can profit the most. If you know that your opponent makes take/pass errors in certain kinds of positions, then that can affect your doubling strategy. Thirdly, despite those stupid dice, it is still possible to steer the game to some extent through checker play. In most games, there will be a few times where there is more than one viable move, and where the choice of move can influence the shape of the rest of the game.
In short, one should not be fooled into thinking that someone who plays with (say) a 2.5 PR is "nearly perfect." At high levels, players are looking for any edge that they can get. If you take two players who both play 2.5 on average and you give one of them access to all the other player's games but not vice versa, you can bet that it will give that slight edge he's looking for.
Messages In This Thread
BGonline.org Forums is maintained by Stick with WebBBS 5.12.