| |
BGonline.org Forums
Reaching DMP
Posted By: Tom Keith In Response To: Reaching DMP (scotty)
Date: Monday, 9 June 2014, at 12:56 p.m.
I recently did a little survey of my own on FIBS to try to determine the actual gammon rate for my own interest. It didn't take long, perhaps about 100 matches to see that the gammon rate was approximately half that used in most match equity tables.
Hi Scotty,
Maybe you already know this, but let me point out that it is hard to calculate gammon rate directly from match results.
Gammon rate is the fraction of all games that would end in a gammon if those games were played to the end. The problem is that many games are not played to the end, i.e., those that end with double/drop. We don't know if those games would have ended in a gammon if play had continued.
You could ignore the double/drop games when calculating your stats, but that might introduce bias. For example, I can imagine it might be easier to lose your market when your gammon rate is very high. If that is true, then throwing out those games (after you double and opponent drops) means the games you are keeping are no longer representative of the overall gammon rate.
Also, in games from matches, there are many situations where one or both players don't care whether they win a gammon because the points on the cube alone are enough to win the match. Gammon wins in such cases are not meaningful because the opponent was not trying to avoid a gammon. You could throw those games out, but it is hard to know which games to keep. Even at 0-0 in a match to 9, if the cube reaches 8, gammons have a much lower value (to both sides) than normal.
I do think there is a lot of value in creating a database of human-vs-human matches, and even human-vs-bot matches. Good luck with your effort!
Tom
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.