Our club In Peoria meets every Monday night and last Monday we just finished a tournament that ran over three Mondays. We played two games a night with one game as white and one as black against the same opponent. We have done this before with local ladder games but this was the first time that we did a USCF rated event this way.
We played game 30 with 5 second delay. Got started by 7:30 and were usually done before 9:30.
When I went to submit the ratings I discovered SwissSys has a conversion tool but I didn’t like the fact that it converted a win / loss to a draw draw over two rounds so I manually recreated the double three rounds to six before submission.
Since players were playing both colors I didn’t make a big issue of who played white first etc, but I think in the future I would have the person that the computer assigns as white play white first. I also would have them report all the info including color with the results.
Have other TDs run double swiss’s like this? Any special tips etc? I did get 18 cautionary warnings when I submitted my USCF rating report online, but it was due to the players playing each other more than once.
We had several upsets. I wish that time had allowed us to schedule a fourth night to this.
I don’t see how SwissSys could do anything else. If you run it through SwissSys as a double-swiss, the results you enter each round correspond to the number of points scored (0, .5, 1, 1.5 or 2), not the results of the individual games.
Maybe you could tell Swis-Sys you are running a 6-round single Swiss instead of a 3-round double Swiss. Let the program make the pairings in rounds 1, 3, 5, then you force all the pairings in rounds 2, 4, 6 by just reversing the colors from the previous round. This would get around the problem of distinguishing between win-loss and draw-draw.
It would also mean that the program would try to alternate colors, between the second game of each double round and the first game of the next double round. This could be good or bad, but since you said you like to specify which player should play white first, you might also want to alternate all six colors as much as possible.
Another trick available in a double round (when done as two single rounds, as above) is the “trye”. If there is an odd number, instead of giving one player the bye, you give three players the trye. First A plays B (once), then B plays C (once), then C plays A (once). This way nobody gets cheated out of a game. Since byes (or tryes) are usually given to the lower-rated, lower-scoring players, they can usually get in three games while everybody else is playing two.
I’ve done several events like this. I think that they’ve all been quick events, but have had the same issues you say. I figure that the individual game results don’t matter very much, since for rating purposes it is only the total score and the opponent’s ratings that matter, plus, as a matter of form, the only thing I know is the total score (out of two) that each player scores. I do make sure to submit the actual crosstable (with the double rounds) to the Oklahoma Chess Quarterly for publication, as well as keep a copy for my own purposes. Since my wife is always available to be a house player, I never have to worry about a bye (or trye).
Basically that’s what I did when I manually created the rating file. A nice addition to the program might be to simply give you to scoring modules when doing a double swiss and then simply display them as consecutive rounds.
The try is a good idea. At club on Monday nights we usually have a player willing to play to even things out when necessary or not to play as the case may be.
The ratings algorithm doesn’t care if someone scores D-D or W-L in such a situation, the resulting ratings change will be the same.
However, the Junior Grand Prix idea that Bill Goichberg has been working on may care, I’ll have to check how he’s handling the awarding of points for a win vs a draw.
Yeah I was pretty sure that rating wise it didn’t matter. I just didn’t want some player looking up his data later saying “Hey I won one and lost one, I didn’t draw twice. What gives?”
Can you tell without a lot of work if the USCF receives a lot of this type of tournament? I like that for the most part it removes the color advantage.
No easy way to tell how many double-game events there are, but I know that about 10% of the events we validate have ‘multiple games against the same opponent’ alerts in them. (That includes matches.)
I’ve run double swisses any number of times for quick rated events, so I imagine this isn’t uncommon. I use WinTD which handles it with no difficulty. I’m able to input the results of each game, not just the score for each round, and it simply produces a crosstable with double the number of actual rounds to report the results of each game (correctly).