| |
BGonline.org Forums
Correlation between ER and opponent
Posted By: poseidon
Date: Friday, 10 December 2010, at 3:17 p.m.
(How) can we relate the expected E.R. of a player of a known strength, given the strength of his opponent? I do not refer to cases of adapted play (can a fish taste twice as good?). Suppose that both players play at their best. I mean what is the difference of a player's (A) E.R. against GNU compared to the same player's (A) E.R. against a mediocre tournament player (B) (E.R. around 6.0). I suspect that playing against a bot is more difficult, due to various reasons: 1) More often games take complicated forms (that also last longer due to more hits) due to unusual checker plays from the bot (strong) side. 2) More often cube decisions from the bot (strong) side are at "grey", close zones (if a player is not very strong (B) he usually doubles too early or too late, thus making the decision for the taker (A) easier/clearer (?)). On the other side, although a mistake at a grey zone would result more often, it would cost less than in the case you blunder against player B. 3)? Ps1) I understand that points 1 and 2 are not always true ( there are players of E.R. 5.0 that are risk-attracted and may produced complicated games more often than a bot(strong) player would (e.g. not run in cases that should run, hit more often than optimal play etc.)) but overall I think most agree that those points stand. Ps2) Let assume that other factors (pip count, pressure) do not apply (e.g. assume that player A plays live both against a very strong tournament player and player B)
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.