|
BGonline.org Forums
Maybe it's time for a change?
Posted By: Florin Popa In Response To: Maybe it's time for a change? (Rick Janowski)
Date: Saturday, 20 February 2016, at 7:03 p.m.
"What is your practical, fully worked out, alternative to PR as a measure for performance? At BMAB we have considered and tested alternative VR techniques to measure winning performance, but unfortunately found this approach unreliable. It isn't difficult to find some small theoretical failings of the PR approach but that is and would be the case for any performance measurement system. In practice, PR works. Consider all the players that have reached Grandmaster level. It is abundantly clear that all these players deserve their status. There are other notable players who haven't yet engaged with the PR process of course."
I am fully convinced that BMAB, with what XG provides at the moment, picked the best solution. I am not expert like you but what I can say is that promoting backgammon as a leading international mind-sporta is a very very good idea, but because is aimed the computer to be the suprem judge, XG needs to improve, first of all the weight of the luck.
An example: I have a profile with a friend of mine who played money games against XG, the score was 100-87 for him after 94 games. XG estimated that based on PR (4,83) he overperformed by 39 p ! So you would expect a huge luck but no, it says only +0.003 per move, slight lucky if I am not wrong. So I think this is not acceptable. I agree about Grandmasters,they are for sure all strong players but I think there are other players who maybe are not able to score so low PR (Lars Trabolt could be one, I remember a his interview in PrimeTime) but could be stronger. The problem I can see regards the lower categories
"From my discussions with the top players, who have been top 16 Giants, there are very who do not agree that PR is the best practical measure of performance. When players say they played to win rather than for PR as an explanation for poor or average PRs, interrogation of the matches virtually always show that the blunders are related to obvious bad plays or oversights, and not through some strategic/tactical adjustment." Is true almost always for the moves, but not for the cube. Actually til now I haven't paid so much attention to the complessive PR of my live matches but I have always looked at the kind of my mistakes and to the MWC. One day I'll do a BMAB application but first I need to study more and also to try playing for PR only and see what happens with my game. But my idea that backgammon is not only about equity will never change. :-)
|
BGonline.org Forums is maintained by Stick with WebBBS 5.12.