I have an upcoming tournament. I anticipate at least 3 sections of 10.
Someone asked me about not pairing folks who drove in together. I
wonder exactly how this would work with SwissSys–or should I even
consider it? Do I use “armchair TD?”
Another thing: I have a father and daughter in the same section. I am
much more inclined to not pair them.
I am afraid that too much tinkering with the pairings will potentially
result in unfair pairings to others. What do you guys think?
If people request it, and if it can be done without messing up the score groups too badly, then I won’t pair them. Generally, I find that they have good reasons.
In a small section, or when they are near the top or bottom, it’s not always possible, and most players understand this, but generally I don’t ask for reasons.
I think you should make EVERY effort to not pair people with each other when they’ve made a request. It’s no fun to travel, sometimes for hours, and pay to play in a tournament only to be paired with someone you play all the time. (Don’t assume that they don’t want to play each other unless they ask – many players won’t mind).
I realize the pairings can be very difficult with smaller sections/tournaments, and they’ll have to be understanding if you decide there’s no other reasonable pairing. There’s nothing wrong, in my opinion, with giving greater weight to requests from players related to each other than you’d give other player’s requests.
Rule 28U specifically allows players to request that they not be paired with each other (though this is left to the director to decide), so I don’t see how anyone can complain about this practice being unfair.
When you register the players assign them to the same club. (Give them the same club code.) Under rules for pairings, pairing restrictions check club. SwissSys will then make an effort not to pair them.
No problem with not pairing family members, etc in the first half of the tournament. Last round pairings, however, should, as a rule of thumb, not be changed for this reason.
In our club tournaments we have a couple of families with players of widely varying abilities. In WinTD I set up each family as its own team and set the program to avoid team pairings if possible (it can instead be set to absolutely prohibit team pairings, but I don’t use that option). That gets past the first couple of rounds without pairing a strong family member versus a weak family member before their scores put them in different groups. After that, for one large family with multiple strong players, I take the team code off of the strong family members because, since they are likely to end up playing each other before the end of the tournament anyway, I’d rather not have the strong family pairing automatically deferred until the final round that determines the trophy winners. That way the weak family members out of the running still have the program try to avoid pairing them, and thus they only end up getting paired with each other if there is no other reasonable way to pair those score groups.
Back in the 1980’s, was at a tournament were the husband and wife were paired against each other in the last round. It ended in a very quick draw, just only a few moves before they went home. The set up of the board and clock, then pack it all up did take longer then the game itself.
Since you are going to have the father and daughter in the same section. The change of the game ending in a draw would be high. With family members and friends, the games in my personal judgement has a better change to be a draw. Do not understand the rating difference betwen the father and daughter. The result of the game, will not make the father or daughter happy. If the game ends in a draw, there would always be the question of a pre-arranged draw.
I’m going to try the option of adding these to the same club, though
I think it would be unwise to force SwissSys to not pair two people.
I like the option better that it would avoid pairing them.
The father/daughter are less than 100 points apart in rating, so there
is a good chance they would need to be paired in the last round to
break ties.
Everyone wish me luck–this is my first big tournament, and while I’ve
done several dry runs with SwissSys, I’ve yet to use it in a tournament.
I usually advise family members who are in the same section that I will try to avoid pairing them for the first half of the tournament, but that I may not be able to do that in 2nd half of the tournament and almost definitely not in the last round, especially if there are prizes at stake.
When players who travelled together make a request not to be paired, I tell them the same thing.
I think most players understand that if two family members (or travel mates) are the only 4-0 players in a 5 round event, there is very little chance that they will not be paired against each other.
Do not be overly concerned by last-round draws, they occur quite frequently and are not a justification for paranoia.
There are several factors that lead to them, some inherent in the nature of tournament chess, some that are a consequence of the prize structure:
In the last round of a Swiss, players should be fairly evenly matched. That increases the liklihood of a draw. (It would be interesting to do an analysis on the USCF’s ratings database to see if last round draws are more frequent as the ratings of the players increases and the extent to which draws are more frequent than between similarly matched players in earlier rounds.)
Players who don’t need an extra half point to secure a prize have little incentive to fight for it. One tournament design strategy to add back some incentive is the ‘plus score’ payout format, which has been discussed in several other threads. (Brief description: The cash prizes are weighted so that two players drawing take home less combined money than if one of the player wins the game, eg. a 4.0 score pays $100, a 3.5 score pays $40 and a 3.0 score pays $25. Thus in a drawn game the players have combined winnings of $80 while in a decisive game they have combined winnings of $125.)
By the last round, players are getting tired (newcomers to tournament chess are often astonished at how much energy it takes to JUST SIT THERE!) Thus they may be less likely to take risks and more likely to use openings and strategies that are more prone to lead to draws.
Outside, non-chess, factors can also make last round draws more attractive. A player facing a 5 hour car ride back home might be more inclined to accept a draw so he can head home sooner.
Yes, that’s how I avoid pairing people in the tournaments that I direct, and no, it doesn’t have to be a real zip code. In last month’s Massachusetts Open, for example, I used 1 for the first set of people who shouldn’t be paired, 2 for the second, all the way up to 9! This was mostly to avoid pairing family members, but in a few cases people from the same club requested not be paired against each other.
If you use this feature make sure zip code pairings are enabled before pairing each round. The word “Zip” should be displayed in red letters. Sometimes Swiss-Sys turns the feature off when you exit the program and start it back up again, or at least it did in earlier versions.
jwiewel wrote: “In WinTD I set up each family as its own team and set the program to avoid team pairings if possible (it can instead be set to absolutely prohibit team pairings, but I don’t use that option).”
My work is largely with pairings at scholastic events where both teams and families are considerations. I use team codes for the schools and club codes for the families. But I do set the program to avoid pairings team members or club members unless necessary. When one school brings a large number of players, it is inevitable that team members are rated. I’ve even had it happen on the top board with brothers in the last round! Using the option in WinTD is, I think, a very good idea.