| |
BGonline.org Forums
Walter and Jake chime in -- perspective
Posted By: Chuck Bower In Response To: Walter and Jake chime in (Chuck Bower)
Date: Friday, 27 July 2012, at 5:32 p.m.
To understand these numbers a little better, consider the following hypothetical situation:
Three players A("high open"), B("middle intermediate") and C("high open") play frequently at their local club. They also play enough ABT events to determine their verbal ranks (e.g. "high open"). On Tuesdays the play is standard singles matches and on Thursday it's doubles. C plays doubles with both players A and B but under the condition that he only makes (binding) cube decisions. A and B make all the checkerplay decisions without any influence from C. Throw in the obvious caveats that C is consistent, they've all played enough (doubles and singles) to have reliable ratings, etc.
Now call their respective singles ratings a,b,c and doubles ratings a&c and b&c. If A and B play each other on Tuesday night the probability of winning is simply based upon their singles ratings difference = a - b and line 2 of the earlier posted table applies. If they play doublets with C (and C', his clone :) then the match winning chances reflect the ratings difference of a&c - b&c and line 3 of the table applies.
Note that not only is it likely that (a - c) [not equal] (a&c - b&c) but it is likely that (a - c) [greater than] (a&c - b&c).
Still, as Bob mentioned, there is some offsetting compensation when a match is played without the cube -- there are more checkerplay decisions so more chances for the better player to gain edge (in the checkerplay space) than if the match were played with a cube. Note that this has two components: there are more games (because games are on average worth fewer numbers of points) and there are more checkerplays per game (because all games must be played until one player has borne off 15 checkers).
There is still the issue of which (no cube vs. more checkerplay decisions) has the bigger impact.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.