Last night I paired the first round in a club event that has 14 players and 9 rounds, playing one game per week. There was nothing special that I had to do in the first round, but looking ahead, if I’m not careful I could have trouble pairing the last few rounds.
In previous weekend tournaments when I’ve had, say, 6 players in a section and 4 rounds, I’ve used rule 29L and it hasn’t been too difficult. Pair the first round normally, assign round robin pairing numbers which match the first round pairings, and pair each of the last three rounds using one of the lines from the round robin table.
For this club event, though, I don’t think rule 29L will work very well. It’s unlikely that the same 14 players will play in all 9 rounds. Some of the players who played the first week will miss one or more rounds, and new players who didn’t play last night will join later in the tournament.
I could use Variation 29L1, 1 vs. 2 pairings, in later rounds but it’s not what the players are used to and I’d rather make pairings that are as similar as possible to the normal top half vs. bottom half pairings.
Then I remembered what Bill Smythe has said about setting the maximum rating difference for making color switches for alternation to 0. I don’t have any experience doing this, but I’m willing to give it a try. As the tournament gets closer to the end I’ll look ahead and make sure it will be possible to pair the last few rounds without rematches.
If this method works I think it would be worth adding as another option in the rulebook, say as Variation 29L2.
If you pair as a Swiss be careful before round 8. If 7 players play each of the other 7 during the first 7 rounds you’ll have to pair the groups of 7 against each other and you’ll have an odd player in each group playing each other a second time. SwissSys does not look ahead to avoid this situation so before you post round 7 manually see if eighth round pairings are possible avoiding playing twice.
Also realize that you may end up with strange looking pairings in round 8 or 9 due to players having already played the opponents close to them in score. In the past I’ve seen things like:
8-0 vs 3-5
7-1 vs 2-6
6-2 vs 4-4
6-2 vs 5-3
5-3 vs 2-6
4-4 vs 1-7
3-5 vs 0-8
It is pairings like this that make it palatable to have players play at the top and bottom each other again:
8-0 vs 7-1 (rematch)
6-2 vs 5-3
6-2 vs 4-4
5-3 vs 3-5
4-4 vs 2-6
3-5 vs 2-6
1-7 vs 0-8 (rematch)
Depending on the version of the software, it may do this anyway.
In the Monadnock Marathon, which was an annual 12 round tournament, the rule was that rematches were allowed in the last four rounds. In a rematch the colors had to be the opposite of what they were the first time the players met. Maybe we can do something similar in this tournament.
I don’t think you should use the round-robin tables. The high possibility of drop-outs and drop-ins would cause problems.
Don’t make the colors work too well. Transpose only to equalize, not to alternate. (You already mentioned this.)
Don’t bother with pairing software. A manual approach is better when the rounds-to-players ratio is high. It encourages the TD to keep a close eye on things.
When pairing each round, look ahead to future rounds, to the extent possible. But with four or five rounds still to go, looking ahead can be almost impossible.
Announce, early in the tournament, that some players may face the same opponent twice. Try to reverse the colors whenever this happens.
The drop-outs and drop-ins will actually make things easier, since there will be fewer potential repeat-pairing situations to stumble over.
Hmm, “this” method was really just a combination of miscellaneous ideas – or was that just my reply?
Why even use a Swiss System at all in this situation, or some kind of blend of RR and SS? Why not 1 versus 2 pairings, with repairings possible after X rounds? After Y rounds, the tournament goes into “King of the Hill” mode so that for the last few rounds, there are no restrictions on repairings. These pairing rules are very easy to do by hand. You just pair top-down, backing up and doing transpositions sometimes on the lowest boards to avoid premature repairings.
This system is called the “Australian Draw”, because it was first widely used in Scrabble tournaments in Australia. I understand that It is now quite common for Scrabble tournaments world-wide, though it is still not very common for chess tournaments.
Why not? If it’s useful information why not make it available to all TDs, instead of just the ones who follow this forum? If the concern is that the rulebook is too big then some chapters can be moved to a separate book or document if and when there is a new edition.
There do not need to be printed variations for everything that one can do. Right now one can do about anything one wants if it is announced in advance. The “approved” variations exist as much to define what publicity one must provide about the variation and when.
Therefore, to add more to the rules for something that impacts very, very, very few - and that may be generous - is distracting from the important stuff that is in there.
I agree with Allen. In terms of tournament organization, the rules are already very permissive. You can already organize and pair tournaments almost any way you desire. The USCF has defined in Rules 27-29 what it means for a tournament to be a “Swiss System” tournament, and presumably if you call your tournament a SS, you have to follow those rules. But nothing stops you from announcing and running a “Messenger System” tournament, and doing what your please. You have to inform the players what the Messenger System is.
The effect of 29K and 29L is that when a tournament is advertised as a 4-round Swiss but on the day of the tournament it turns out there are four players or less, or a tournament is announced as a 5-round Swiss and there are six players or less, the USCF sanctions changing the tournament to a different tournament format. The TD is not obliged by USCF rules to go though with the tournament as a SS tournament.
We shouldn’t be adding options to 29K and 29L. What we should be doing is removing both of them entirely from the rules, and replacing them with the following paragraph.
This allows changing to RR, as in the current 29K. to the RR/SS blend of 29L, to the 1 vs 2 pairings of 29L1, or to a variety of other systems, such as the one Bob proposes.
I think this is a difference in philosophy. I like the fact that the USCF rule book doesn’t just contain rules but also includes advice for tournament directors and organizers. Your proposed rule change would throw away the advice on how to handle Swiss system tournaments with a small number of players compared to the number of rounds and leave just a bare rule saying that in this situation the TD can change to a different pairing system. Rather than throwing away the existing advice I think there should be more of it, if the new suggestions are better in some ways than the existing ones.
I do think there’s something to be said for breaking up the current rulebook into two or three smaller ones. Possibly: one book intended for players, one book with rules for organizing and directing USCF-rated tournaments, and one book with advice for organizers and directors. The FIDE Laws of Chess (Chapter 15) shouldn’t be in the USCF rulebook at all, since there are other books on that subject and the Laws are subject to change by FIDE.
It is a difference in philosophy, but my philosophy is right. I think that if the title of the book is “USCF Official Rules of Chess”, then it should contain rules (prescriptions). These are the things which are obligatory, required, etc. That is what you want to have the Rules Committee and the Delegates discussing and adopting. You can have opinions about them, but you still have to follow them. If you violate them or make a ruling that contradicts them, you are liable to have your ruling appealed and overturned.
Guidance, tutorial information, ideas, tips, hints, suggestions, rules of thumb, observations about general player preferences or expectations, etc, are different. They don’t need to be voted on and adopted by the Rules Committee and Delegates. You don’t need TD’s memorizing them. You don’t need TD’s to be tested on other people’s tips. People can have different opinions about these things, and different ways of doing things. The result might be suboptimal (in which case the TD will learn, hopefully), but nobody can appeal to the Rules Committee or the TDCC because a TD doesn’t follow a “tip”. It isn’t that this tutorial information, etc, is not useful or valuable, or that it shouldn’t be somewhere. On the contrary. It just belongs somewhere else than the rulebook.
In the USCF Official Rules of Chess, it is not always easy to tell what is tutorial and what is an obligatory, prescriptive rule, especially in the Tournament section. While a lot of the “tips” are identified as such, there is a lot of “guidance” type information which is effectively more tips, but not identified as such. I’ve commented before on how hand-wavy the pairing rules are, for example. Also, the tutorial or informative sections increase the length of the rulebook dramatically.
I agree that those sections increase the length. They are helpful to a new TD - and sometimes an experienced TD.
One does have to be careful to recognize that tips are not rules. Also, the numerous variations can be confusing especially when a rule book lawyer shows up with a point drawn from a variation (or a tip) which is not really applicable. Fortunately this happens vey rarely.
There has been a thread promoting merging USCF and FIDE rules. Pairing rules are one area that are significantly different.
That’s particularly true in this case, as the proposed “variation” seems to be nothing more than a collection of miscellaneous small tricks that can be applied one at a time or collectively.
