| |
BGonline.org Forums
What should the PR difference be?
Posted By: Florin Popa In Response To: What should the PR difference be? (Stick)
Date: Tuesday, 13 March 2018, at 12:04 a.m.
I would buy it if not too expensive. Many of the cube decisions XG says are right in reality can be wrong when a tangible difference exist between players or/end when you know your opponent well. In complex positions the favourite should double earlier, the opposite in simple positions and then remains the issue of the real MET. I think we will never have a real answer even if collecting big data on servers like Ebif or BgStudio could help. I have seen some data, 1000 is not very big, and looked like playing under 4.0 PR doesn't show an improving in W/L rate. I looked also at some W/L rate on the Usbgf site for a few very strong players with the cube (from BMAB) and I have the impression they don't win so much, it can be for other reasons like stronger opponents because they play more master tournaments or the data is not big enough (about 400 matches). Hard to say who is adjusting properly and better than others, in the case of Akiko she has won a lot for 2 years but in the next 2 not much so does she lost this skill or it was more a question of luck/unluck? I think she can play under 4.0 in a Bmab tournament but looks like she is not interested in doing it. I still think we should use a PR for cube and one for moves, that could help in doing comparisons.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.