| |
BGonline.org Forums
Suggested Idea for Performance Grading
Posted By: David Rennie In Response To: Suggested Idea for Performance Grading (Timothy Chow)
Date: Sunday, 5 October 2014, at 6:48 p.m.
To be fair that was Ray's estimate. I was just saying he may well be right. I can tell you from my own experience that 7-8 PR is apparently NOT the optimal range for winning regularly. But perhaps my sample size is too small. (That's what she said).
As to whether XG plays close to the Platonic ideal of Perfect backgammon I just don't believe we are qualified to know that. You may say it's obvious that it does but it was obvious that the sun revolved around the earth (I mean just look at it!) until Copernicus came along and messed that all up. Now I doubt very much that xg is exactly Ptolemaic but it may well be off the mark enough. At the very least its analyses are somewhat rough in enough positions to vary PRs significantly at the GM level. As Mary points out rollouts of a few key decisions can cause enough of a PR swing to make a 2.7 more of a 3.3.
But I maintain that even if it was an accurate metric for Perfect play so what? Backgammon is a game with an element of chance in it. That's the fun and frustration of it. Maybe the world champion one year was lucky or maybe he or she played a 0.00. It doesn't matter. He or she is the world champion in a game where luck plays a role! That's the game!
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.