| |
BGonline.org Forums
Cube error rates
Posted By: Gregg Cattanach In Response To: Cube error rates (Fabrice Liardet)
Date: Thursday, 23 July 2009, at 7:09 p.m.
I guess the problem is that you want 'Favorite' to have some global and all-encompasing meaning when that was never the intent. It measures exactly what it measures: the difference in the MWC% errors for each player in this match + 50%.
From the Snowie documention:
Snowie also displays an estimate of how likely one player is to win a match against his opponent or how many points per game he should average in a money session. This estimate is computed in the following way. Snowie will cumulate the the real cost of all the errors which occurred during the match or the money session for each player. In a match the errors are cumulated in match winning chances while in a money sesssion they are cumulated in points per game. Based on these cumulated errors of each player one can compute how likely one player is to win the match. For example if player A cumulated errors of 10% and player B cumulated errors of 20%, player A will win 60% of the matches (50 + 20 - 10) if both players will average the same amount of errors for each match. The computation for a money session is done in an equivalent manner. Indeed, if player A cumulated errors of 0.15 pts/game and the other one 0.45 pts/game player, A will average 0.30 pts/game against player B.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.