Tournament issue: 3 byes for the same player


Today in a tournament with Swiss system, we had a player that was on bye 3 times out of 6. I believe this should not happen, but I don’t know if this is some sort of bug in the implementation.

The tournament:


I will post the same issue again, since this happens very often. Today we had a tournament with 4 rounds where the same player got bye three times! And he won! Ok, he would win the tournament anyway since he is indeed much stronger, but this is not how things were supposed to happen.

The tournament:


It might be just how the Swiss system works. It tries to find the best player, but also eliminates the possibility that two players play each other twice. Riemsdijk was a bye in the first round (because pairing is by strength, and Riemsdijk is the strongest, this will always happen), then won from Starked in the second round (they were paired, because Starked was second strongest and won the 1st round as well), and then Starked won the third round as well, establishing that Starked is indeed the second strongest player, and Riemsdijk the strongest player in the tournament. Riemsdijk then won, because he defeated all other players who could have been a contender for the 1st place in the tournament.

I think you can do a similar kind of analysis for the first tournament you linked.

The Swiss system is meant to find the best players as efficiently as possible. With an odd number of players, it has to choose a bye that makes it least difficult to establish the winner, or in other words, those players where their ranking is least certain should play.

I’m still not entirely sure how the McMahon system works, but it might be worthwhile to use that system instead, see if it improves matters?


Thank you for the answer!

Indeed, the system was efficient to determine the best players correctly. I will search about how the byes are determined, perhaps in McMahon there is something different.

I’m fairly certain from my chess days that this is not how Swiss is supposed to work. But we don’t even know what exact Swiss system OGS uses, that’s why we can’t say for sure. Maybe in the future OGS could open source pairing systems.


This topic was automatically closed 91 days after the last reply. New replies are no longer allowed.