Ideas for Improving Tournaments

[jimloy]

For many years, I have suggested using Performance Ratings for tie-break in chess. This is the rating that the computer calculates for you, just for the one tournament. It follows the standard idea that your tie-break should be based on the strength of your competition in the tournament. But, instead of being based upon their scores, it is based upon their pre-tournament ratings, which I think is a better measure of strength. Besides, it almost never fails to breaks a tie.

[John Acker]

For many years, I have suggested using Performance Ratings for tie-break in chess. This is the rating that the computer calculates for you, just for the one tournament. It follows the standard idea that your tie-break should be based on the strength of your competition in the tournament. But, instead of being based upon their scores, it is based upon their pre-tournament ratings, which I think is a better measure of strength. Besides, it almost never fails to breaks a tie.

Two questions:

1. Would using Performance Ratings require TDs to use particular software to track and calculate ratings?
2. Can this software assign provisional ratings for first-time players?

[chipschap]

In computer pairing if a chess tournament program is used, colors are indeed important but may present less of a problem than you might think. Unless it's an extreme case (the same color 3 times in a row, for instance) programs pair by score, and won't do something like pair someone with a high score against someone with a low score just to fix colors. If an edge case does arise, it's likely to be in a later round, and the TD can do manual adjustments, but this should be rare.

We use pairing programs for blitz chess all the time (since games are played in pairs there is no concept of color), and we just go with the flow, and it's hardly ever an issue.

Players should never, ever play each other twice in a Swiss tournament unless there are fewer players than rounds. For a TD to decide to break a tie by pairing players manually for a second time is contrary to the principles which make Swiss system tournaments work properly.

I'd suggest that even imperfect computer pairings are far better than any manual system. In the chess tournaments I direct, I would hate to go back to the old pairing cards. With the computer, at least there is no possibility of being accused of bias.