Has anyone noticed any problems with WinTD’s calculation of Sonneborn-Berger tiebreaks for RR’s? I know that it doubles the points, but when I calculate by hand, they’re always off.
Does the S-B system adjust tiebreak point of “opponents” when those opponents have unplayed games? I know that, according to the rule book, nothing is added (FOR THE PLAYER) for (THE PLAYER’s) unplayed games, but what about the unplayed games of one’s opponents?
Thanks,
-Terry
I have a similar problem, but no solutions. When I ask WinTD to sort by tie-breaks it never does it correctly. Not sure how to fix it either…
Anybody with any thoughts on this?
MY REPLY TO ME: Sonneborn_Berger does not adjust the scores of opponents. I was erroneously thinking that there was an analagous situation with Modified Median and Solkoff, but this isn’t the case.
STILL: WinTD doesn’t calculate the tie-breaks correctly. If someone would care to ascertain this for me, I would appreciate it. Perhaps my version (which is 4.0) has a bug?
The help file states as follows: “As it is reported in WinTD, a player’s Sonneborn-Berger is the sum of 2x(scores of opponents defeated) + 1x(scores of opponents drawn). This is twice the usual S-B. We do this so that the it can be shown with just one decimal like the other tie breaks”.
My Round Robbin tie breaks were correct. The only thing is they multiply
the tie breaks times 2. I also did them by hand before the end of the tournament just to make sure.
Terry Dean Likens
Not sure if there’s a difference, and I will test this later, but my first thought is that there’s a problem when an unplayed game, i.e. after a forfeit loss/win, is involved.
Sd: did you have any unplayed games when you calculated the breaks? Also, which version of WinTD do you have?
I appreciate your reply.
-Terry