I have a pairing question that just came up in a tournament where someone was using my TDX pairing software - in the last round there were 2 players left undefeated, but both had played black in the previous 2 rounds, and TDX did not pair them as a result. The TD adjusted the pairings so that they would play each other - would that be considered the “better” pairing?
Thanks!
The pairing software did not want to give one of the players 3 blacks in a role. The rules have change in the past few years granting the director the right to give 3 blacks. If I had to give someone a 3rd black, would talk with the player for the reason. Then hope he would not get upset! If you have to give a player a 3rd black, there was some very bad pairings in the past rounds. The TDX software could have be written, when the director could not give someone a 3rd black. Not sure how old the TDX program is.
As a player, I would be upset having a 3rd black in the event. Would walk up to the director asking how did this pairing came up. Would not challange the pairing, just want to understand why. On the face of it, would like the TDX pairings for the last round.
Pairings become nasty with a small field of player, with a large number of rounds. If you can tell us how many players in the tournament and how many rounds, that would give some help in understanding the directors problem.
In my final judgement, would have looked at the pairings the TDX came up with. Had the two players come to the directors table, and ask if they want to play with each other with one getting a 3rd black. If one did not like the idea, would have used the TDX pairings.
Rule 29E5f says that no player shall be assigned the same color three times in a row, unless there is no other reasonable way to pair the score group or unless necessary to equalize colors.
I think that it is pretty clear that one of the players has to be given the third black.
FWIW, I think that one of the greatest shortcomings in pairing software is too much dependence on making colors work. Sometimes pairings can get awfully strange.
Alex Relyea
It’s actually Rule 29E5f1 that is the deciding factor. Had it not been the last round, I would probably have given them both white. Although it is speculation because we don’t have all the info needed to make the determination.
variation 29E5f1. Last-round exception. Except for the last round, when it may be necessary to pair the tournament or class leaders, players shall not be assigned the same color in three successive rounds.
Since it is a variation, the software was written as a rule not as a variation. The director, did not follow this variation. If it was myself as the director, would have followed this variation.
Its’ a question of pairings, its also a question with the players in the prize money. Since the director paired the players, one to have black for a third successive time. Both White and Black could or will win a prize with the final score of the game. As it is the final round, any pairings will have a rational idea what the final scores should be. Change the players on board one and two, can change the chance of board 3 having a beter chance for a prize.
True very true. We do not know how many rounds; or the number of players; or the pairings looked like for the final round. Myself would not have over turned the judgement of the software.
I feel the pairings did change the out come of the prize money. When looking at the pairings and know it is the final round, have some sense to say who should win and who should lose and some its’ to close. The change in the pairings, the director has changed the sense of who should win and who should lose and some its’ to close.
It sounds as if the TD made the right call. I personally have never had a situation where I had to give someone 3 blacks in a row.
The tournament in question was a non-rated scholastic event, 5 rounds, 70 players. Both players had colors of BWBB for the first 4 rounds.
What I will end up doing, I think, is modifying TDX to add an option so that you can force score groups to not be broken simply based on color allocation. Should this only be applied to the last round, or all rounds, and to the top score group only or all score groups?
The more I look at it, the better I like the TDX pairings then the directors pairings. As the directors pairings did give one player BWBBB. Having 4 blacks and 1 white, that player will not have nice words.Since it was a non-rated scholastic event, that would trow off any idea how stronger the players are for the first two rounds.
If you have the tournament again with the same or equal number of players. Since they did play with each other, use the final score of this tournament to help with the roster in the next event.
The players in the same group, just place them as random within the ranking of the roster. Its’ not as good as having USCF ratings for the first round pairings, it will make having 11 fields with each field being random a little better then having the roster all random. If you can place each field with a USCF rating, it would help the software with the pairings for the rest of the tournament. Its’ not the best system but it can help the pairings for the next event.
The director had no choice once the final round came. I would love to see the wall chart because I can’t believe the software did a good job.
How many sections? How many were in the section in question?
TwinFeats do you have a wall chart you could post? Just curious.
The TD for this event is going to email me the crosstable tonight, so I’ll post that when available.
It would have to be a wall chart, the crosstable won’t show the colors.
I once directed an event with about 20 players in it where Black won every game in both rounds 1 and 2. It was a 4 round event.
I had 4 or 5 players in the 2 point group, all of whom had 2 blacks in a row.
I called the top players together, explained the situation, and told them I was going to do the natural pairings but with a random draw to determine which 2 players got 3 blacks in a row.
They were fine with that procedure, and as I recall one of the players won his 3rd consecutive game with Black in that round, won his 4th round game (with White) and was clear 1st in the event after some of the other top players drew.
This was in the days before computer pairing.
You must pair players with the same score if possible. “Possible” here means that they have not already played. (Well, there is a reservation about “equalizing colors,” but I don’t think I’ve ever seen that come up.) This takes priority over “three in a row.” This was extensively discussed, and was an explicit change between the third and fourth editions of the Rulebook. It remains in effect in the fifth. Having the two leaders play is the only legal pairing, unless you are using (and have posted) a variation.
That has to be a rarity. I would imagine the people were able to take it well due to the circumstance. I don’t think I would have a problem mixing score groups for the 3rd round in that situation. You would have 5 all black 2.0, 5 all white 0 & 10 mixed on 1.0. Which is less desirable? I guess it’s a no win situation.
variation 29E5f1. Last-round exception. Except for the last round, when it may be necessary to pair the tournament or class leaders, players shall not be assigned the same color in three successive rounds.
The reason I like the TDX pairings then the directors pairings for the last round, is variation 29E5f1. If you want to reject this variation that is fine. The pairing program did give BWBBW for both players. Its’ acceptable to have 3 blacks and 2 whites with 5 rounds. If you can find having (BWBBB) 4 blacks and 1 white acceptable, would find it strange.
We understand both players as being 4 - 0 at the start of the pairings. The organizer could have had one trophy for first place, as it was a scholastic event. The director could have looked at the pairings of the TDX, if the players did play someone in the 3.0 or 3.5 score group. Then having due color of white to get BWBBW for the color history. The director would feel both players would win the final game in the TDX pairings. If there is only one trophy for first place, the organizer would have had a problem. It looks to be the director had a bias to have the players face each other, and hope the players would not have a draw.
Do not have a problem with variation 29E5f1. As both players were 4 - 0 after 4 rounds, as the leaders, they should have not been paired up with each other. Since there was no advertisment to state variation 29E5f1 would not be used, would have drop the players down into the second score group. The more information I get from the event, the more I start to defend the pairing program then the directors.
Starting to feel the director had a bias, as I cannot prove how many trophies there were or how they would deal with a tie. Having the two 4 - 0 players face each other is ideal for the organizer, as the chance of a draw in the last game is small for a scholastic game then a adult game. Having the two 4 - 0 players play down the score group, then get white in the final game. It would not be ideal for the organizer, as both player have a good chance going 5 - 0 in the event. If the organizer only has one trophy for first place, the organizer would be in a rock and a hard place.
My question is not with crosstables, mine is with the trophies. If there was only one first place trophy, it would have been a bias of the director to pair the players to have one player have BWBBB. During the tournament, I work for the players not for the organizer. The more information I get, I would have paired the players in a lower score group. If the players went 5 - 0, would tell the organizer to get a second first place trophy
I’m sorry, but this sentence makes no sense to me. Presumably there was no announcement that every other variation in the rulebook would not be used. Does that mean any or all of them could or should be used without notice?
In any case, 29E5f1 is completely irrelevant here. The only thing that variation does is give “three in a row” priority over scoregroup except in the last round. Since this was a last-round pairing, it doesn’t apply.
You can certainly argue that dropping the two leaders to avoid “three in a row” is a “better” pairing in the abstract. That seems to have been Tim Redman’s position when he wrote the 3rd-edition rules. However, the Rules Committee and the Delegates diagreed, and the rule now is clear: pairing within the scoregroup is more important than assigning the same color three times in a row.
Is this a typo? We’re going into the final round with only 2 players having a perfect score and they should NOT be paired against each other???
This also looks odd to me. We should pair one way if trophies are involved but pair differently for cash awards?
The tournament was a scholastic event. Now its’ common for scholastic events to have trophies. If it was for a cash award or for a trophy, would be looking into the color history to give a better pairing.
Soooo… We pair scholastic events one way but open events (for example) get paired differently?