| |
BGonline.org Forums
perfomance rating
Posted By: Phil Simborg In Response To: perfomance rating (Jake Jacobs)
Date: Thursday, 10 September 2009, at 2:47 a.m.
Here is one of the true problems with the Giant list, and why it's not valid to compare backgammon to horse handicapping.
Let's assume, for just a second, that we could have a pretty accurate list of the best players in the world. What odds would you give for the No. 1 player over the No. 32 player in say, a 13 point match? Would you say he was 7 to 5 or 7 to 6? My guess is that it would be something closer to 13/11.
What odds would you give between No. 1 and No. 10? 13/12?
What do you think the ER rating difference is likely to be in a typical 13 point match between those players? I'd love to have the underdog plus 1.0.
The point is, the difference of skill between all of the players is so very small that it is virtually impossible for ANYONE to objectively select one great player over another given our lack of empirical data. How can anyone know if Falafel is better than Sander, or where Nack and Francois and Neil and Mochy fit in? And as you go down the line, how can we know the difference between Wachtel and O'Hagan and Sax and Stick and MCG?
In a horse race we have definitive evidence...we know exactly what their records are and exactly what their times are on each kind of track condition. We simply don't have that evidence in backgammon.
THEREFORE, the Giant list, in my opinion, is useful only to see which 32 players are voted best at this time, but the order of the people on the list, for the most part, cannot possibly be useful in determining their relative skills or likelihood of winning a given tournament or match. MAYBE you can bunch them and say that the top 1/3 are probably better than the next 1/3 and that group is probably better than the bottom 1/3, but again, it would be very close and without clear evidence.
Going to the next step...it's probably reasonable to say that there are a good 50 players in the world that are not very far from the skill level of the players at the bottom of the list.
What would be really interesting is to see how accurate the voting was in past Giant votes. It would be fascinating to see what the match W/L records for all the players on the last Giant list have been, or how they have fared in major tournaments. I would bet there is little correlation to their order, and I would bet there will be many people with far better performance that were not on the list at all.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.