On Mon, May 9, 2011 at 6:09 AM, Nick Wedd <[email protected]> wrote:
> In message <[email protected]>, Ingo Althöfer < > [email protected]> writes > > One comment added: >> >> In swiss tournaments, proper pairing for the first >> round is the more important the larger the number >> of players relative to the number of rounds. >> > > Suppose there are ten players, numbered from 1 to 10 in order of strength. > Is the "proper" round-one pairing > 1 v 6, 2 v 7, 3 v 8, 4 v 9, 5 v 10 > or > 1 v 10, 2 v 9, 3 v 8, 4 v 7, 5 v 6? > The proper pairing is top half vs bottom half starting with best player in top half vs best player in bottom half. That means your first example is correct. When pairing later rounds, you follow the same pattern. If there are 8 players with 4 points after 4 round, you rank them in order of strength and pair the best against the one in the middle, the second best against the second best in the bottom half, etc. Where there are an odd number of players in some "scoring group" you are supposed to promote the top rated player in the next scoring group to get an even number. For example if there are 3 players with 4 points you include the top rated player with 2 1/2 points. (Or 2 points if there are no draws.) There are rules that deal with not playing the same color too many time and you are not supposed to play the same player twice in a tournament. In general you are can make small adjustments in the ordering if otherwise some player is playing the same player twice or is playing the same color too often. If you are one of the top rated players you should easily win your first few games but it's supposed to get more and more difficult as the tournament progresses. The idea is that after a few rounds every player has found his level - the stronger players will generally have perfect scores but now it gets tough for them. The weaker players have lost a lot of games but now they are playing other losers. With improper pairings the strong players have knocked each other out, weak players may have excellent scores because they never had to play the good players and you get serious mismatches. The mis-matches should occur in early rounds, not late rounds. In theory, the largest mismatches should occur in the first round and even that is limited to half the field. The top player should never have to play a player weaker than the middle players. Of course that can change but only if there are upsets. But even if the weakest player keeps winning, it's not likely he will play the strongest player unless of course he KEEPS winning to the end (or close to it) or the strongest players is losing. Don > I ask only out of curiosity, and in case I some day do the draw for a > non-KGS Swiss tournament. For KGS tournaments, it is irrelevant, as I have > no control over the KGS scheduler, which I think does the first-round draw > at random. > > Nick > > > >> (Typical case: number of rounds = c * log(number of players), >> where c is a constant larger than 1, and log is meant for base 2.) >> >> In the KGS tournament, there were 12 rounds for >> 12 participants, so pairing for round 1 was not >> really crucial. >> >> Ingo. >> >> >> >> -------- Original-Nachricht -------- >> >>> Datum: Sun, 8 May 2011 15:02:28 -0400 >>> Von: Don Dailey <[email protected]> >>> An: [email protected] >>> Betreff: Re: [Computer-go] May KGS bot tournament: 19x19, fast >>> >> >> In swiss tournaments or knockout tournaments it's important to rank the >>> players to the best of your ability before doing the pairings. ... >>> >> > -- > Nick Wedd [email protected] > _______________________________________________ > Computer-go mailing list > [email protected] > http://dvandva.org/cgi-bin/mailman/listinfo/computer-go >
_______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
