McMahon System

Running Chess Tournaments
1

I had never run into this system in (at the time) around 35 years of tournament play. Then while visiting my son at college, I had some time and decided to support a local on campus chess tournament by playing a couple of rounds. I was surprised at the pairings they used - which I later learned is called the McMahon System although I don’t know that anyone there called it that.

The McMahon System is like the Added Score Method of accelerated pairings, except that the score IS ACTUALLY permanently added for purposes of tournament standings, not just temporarily added for pairing purposes. Also, the score is added based upon a rating cut-off. So, for example, maybe only the top quarter of a tournament would receive the added score.

Apparently this method evolved out of Go tournaments.

Has anyone ever used this approach? Thoughts?

2

Yeah, for some reason this became the standard Go tournament system at some point. People in the Go community seem to like it fine. It may seem kind of unfair, but in a giant open tournament everyone knows that the people at the top totally outclass the others so the folks at the bottom don’t seem to mind already being effectively disqualified from winning the whole tournament.

3

I’ve seen this done in one-day tournaments with about a $20 EF and prizes and even did it myself in a free tournament with no prizes. The $20 Event had no 1st, 2nd place places overall, but instead had prizes of 1st and 2nd in each class. The rating cut-off was each class. U1200 started with 0 points, 1200-1399 started with 1.0, 1400-1599 started with 2.0, etc.

Good alternative format if you have a wide range of strength but not enough players to justify multiple sections.

4

Yeah, but the difference is that go tournaments (at least those I’ve played in) are handicap tournaments, in which players of unequal ability are (roughly) equally matched. The upshot is that the winner of the tournament isn’t necessarily the strongest player but rather the player who plays the farthest above his usual level. The point of the McMahon pairings is to try to maintain the smallest distance between opponents’ strengths in order to produce as many even matches as possible, as opposed to Swiss pairings, which try to maximize the distance in order to produce as many clear winners as possible.

5

The Swiss really only does that in the early rounds. As time moves on, by function of elimination, the spread narrows. True, Swiss pairings pick the widest part of the available spread, but the spread “per round” is “theoretically even” and narrows round by round.

(By “theoretically even” I mean this:

Swiss: all rank differences in this example are an even “4”

1 v 5
2 v 6
3 v 7
4 v 8

NCAA example: Rank differences vary

1 v 8
2 v 7
3 v 6
4 v 5

6

It does seem that this McMahon system should be a good way to run a class tournament when there are too few players to run a class tournament.

Likely it will produce a high percentage of “in-class” pairings, which in this context means players who started with the same initial handicap score.

Bill Smythe