Page 1 of 3
British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 10:30 am
by Brian Beattie
My son recently finished playing the British in the junior rapid section. He ended up playing 3 times as white and 6 times as black. With the extra three blacks not only played in a row but also in the final 3 rounds of the tournament.
https://chess-results.com/tnr974090.asp ... =30&snr=21
I have some ideas on why this occurred as well as some ideas on how it could have been prevented but since I lack any experience in this area it is all speculation. If anyone can enlighten me on the specific facts on how the pairing system works or point me in the right direction I would be grateful.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 10:36 am
by IM Jack Rudd
There wasn't an awful lot that could have been done about that one: the 5½/8 scoregroup had four players - Keerthi Kumar, Beattie, Kendall and Liu. Keerthi Kumar had already played Kendall and Liu, so had to be paired against Beattie. Both players had a colour sequence going into that game of BWBWBWBB, so the higher-rated one, in this case Keerthi Kumar, got their due colour.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 10:43 am
by Jon D'Souza-Eva
His opponent had two blacks in a row going into round 9. Since they were paired together (not sure if this should have happened!), one of them was going to have to have three blacks in the final three rounds. My understanding of pairing rules is a bit limited, but I think that if two players are seeking the same colour then the higher rated player gets the colour they ought to get. "Higher rated" means the player with the highest score, or if these are equal then the one with the highest FIDE rating (or whatever rating is being used in that tournament).
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 10:45 am
by Brian Beattie
Is the higher-rated one the one that receives the preference in general or is because there isn't a tiebreaker system in place so it defaults to rating which in this case is FIDE.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 10:57 am
by Roger de Coverly
IM Jack Rudd wrote: ↑Mon Jul 29, 2024 10:36 am
Both players had a colour sequence going into that game of BWBWBWBB, so the higher-rated one, in this case Keerthi Kumar, got their due colour.
Is there a rule which says that if it wasn't the last round that the players would be floated so that they didn't get a BBB sequence (or a WWW one)?
The rules for pairings are deterministic which enables them to be generated consistently by computer programs. That said, there are periodic minor changes to resolve perceived anomalies that have arisen. In this case you can break the BBB sequence by abandoning the usually fundamental rule about playing people of the same score.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 10:59 am
by Jon D'Souza-Eva
Brian Beattie wrote: ↑Mon Jul 29, 2024 10:45 am
Is the higher-rated one the one that receives the preference in general or is because there isn't a tiebreaker system in place so it defaults to rating which in this case is FIDE.
Jack is the expert but my understanding is that once the tournament has started it's the player with the most points who gets the colour they are due. If both players have the same score (quite likely since they have been paired together!) then their ratings are used. Nothing to do with whatever tiebreaking system is in place.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 11:07 am
by Ian Thompson
Brian Beattie wrote: ↑Mon Jul 29, 2024 10:45 am
Is the higher-rated one the one that receives the preference in general or is because there isn't a tiebreaker system in place so it defaults to rating which in this case is FIDE.
Higher-rated is not quite right, in general. it's the higher-ranked player, in the current tournament standings, that gets their colour preference. In this case, that is the higher-rated player, but it need not be. An obvious example where it's not rating is when one player is floating up or down. In that case, the player with more points would get their colour preference.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 11:14 am
by Brian Beattie
Ian Thompson wrote: ↑Mon Jul 29, 2024 11:07 am
Brian Beattie wrote: ↑Mon Jul 29, 2024 10:45 am
Is the higher-rated one the one that receives the preference in general or is because there isn't a tiebreaker system in place so it defaults to rating which in this case is FIDE.
Higher-rated is not quite right, in general. it's the higher-ranked player, in the current tournament standings, that gets their colour preference. In this case, that is the higher-rated player, but it need not be. An obvious example where it's not rating is when one player is floating up or down. In that case, the player with more points would get their colour preference.
So, in a tournament with a tiebreak system then if two players were on equal score the one with the higher tiebreaks would have the higher rank and their colour prefenece. In lieu of a tiebreak system the higher rated would have the higher rank.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 11:22 am
by Ian Thompson
Brian Beattie wrote: ↑Mon Jul 29, 2024 11:14 am
Ian Thompson wrote: ↑Mon Jul 29, 2024 11:07 am
Brian Beattie wrote: ↑Mon Jul 29, 2024 10:45 am
Is the higher-rated one the one that receives the preference in general or is because there isn't a tiebreaker system in place so it defaults to rating which in this case is FIDE.
Higher-rated is not quite right, in general. it's the higher-ranked player, in the current tournament standings, that gets their colour preference. In this case, that is the higher-rated player, but it need not be. An obvious example where it's not rating is when one player is floating up or down. In that case, the player with more points would get their colour preference.
So, in a tournament with a tiebreak system then if two players were on equal score the one with the higher tiebreaks would have the higher rank and their colour prefenece. In lieu of a tiebreak system the higher rated would have the higher rank.
Yes. Effectively, rating is being used as a tie-break for pairing purposes. If that's equal, FIDE titles are often used as the next ranking criterion and if they're the same then alphabetical order is used.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 11:48 am
by Roger de Coverly
Ian Thompson wrote: ↑Mon Jul 29, 2024 11:22 am
Yes. Effectively, rating is being used as a tie-break for pairing purposes.
Before the days when almost all players had FIDE ratings, a tie break system known as Buchholz was used for pairings in European tournaments anyway. It's simple enough to use with pairing cards, being the sum of opponent's scores. I think earlier iterations of swiss-perfect and similar programs may have used it. In round 1, I think they used FIDE or converted national to get the top half v second half, but ignored ratings from then on. This was 1990s. British arbiters had a habit of pairing rated v unrated.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 11:52 am
by Jon D'Souza-Eva
Brian Beattie wrote: ↑Mon Jul 29, 2024 11:14 am
So, in a tournament with a tiebreak system then if two players were on equal score the one with the higher tiebreaks would have the higher rank and their colour prefenece. In lieu of a tiebreak system the higher rated would have the higher rank.
I don't think this is the case. I think "higher ranked" means:
1) the player with the most points
2) the player with the highest position in the initial ranking list
And nothing else - the tie break system being used for the tournament is completely irrelevant.
Hopefully Jack or someone else who actually knows what they're talking about can clarify this.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 12:53 pm
by Alan Walton
IM Jack Rudd wrote: ↑Mon Jul 29, 2024 10:36 am
There wasn't an awful lot that could have been done about that one: the 5½/8 scoregroup had four players - Keerthi Kumar, Beattie, Kendall and Liu. Keerthi Kumar had already played Kendall and Liu, so had to be paired against Beattie. Both players had a colour sequence going into that game of BWBWBWBB, so the higher-rated one, in this case Keerthi Kumar, got their due colour.
I would be questioning why Beattie in Round 8 got black and not Dean, both had a 3-4 W/B split; and Beattie was the higher rated player, so I would have thought he should get white
But without knowing the exact logic in the computer pairing system, only speculation
Agree looks like R9 was correct, but under the Keerthi colour situation in this round it makes my Rd 8 point more valid
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 12:59 pm
by Alex McFarlane
I won't claim to know what I'm talking about!
This seems to be the day for asking questions on the FIDE pairing system. I've just had an email asking about pairing rule E5.
Jon is absolutely correct, the higher ranked player is the one with the higher score. If the scores are level then it is the player with the higher rating. If the rating is the same then the players are rated by title and finally alphabetically.
This means that the order of seeding within each score group will remain the same throughout the event. If any of the tie-break methods were to be added then seeding order would change with each round and make it more difficult to check or predict pairings.
Alan, Beattie played Shankar in round 8 so I'm a bit puzzled by your comment.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 1:19 pm
by Brian Beattie
Alex McFarlane wrote: ↑Mon Jul 29, 2024 12:59 pm
I won't claim to know what I'm talking about!
This seems to be the day for asking questions on the FIDE pairing system. I've just had an email asking about pairing rule E5.
Jon is absolutely correct, the higher ranked player is the one with the higher score. If the scores are level then it is the player with the higher rating. If the rating is the same then the players are rated by title and finally alphabetically.
This means that the order of seeding within each score group will remain the same throughout the event. If any of the tie-break methods were to be added then seeding order would change with each round and make it more difficult to check or predict pairings.
Alan, Beattie played Shankar in round 8 so I'm a bit puzzled by your comment.
My son (Beattie) was the lower rated in both R8 (Shankar) and in R9 (Keerthi Kumar). So twice in a row was given black. Ultimately leading to 3 blacks in a row for R7, R8, and R9. A tie break system or even an ECF rated event (which is far more accurate for juniors) may have avoided this. And, in any case, having three blacks to finish a tournament would seem unreasonable on any circumstance and I would have anticipated some intervention.
Re: British Chess Championship Pairings Question
Posted: Mon Jul 29, 2024 1:33 pm
by NickFaulks
Brian Beattie wrote: ↑Mon Jul 29, 2024 1:19 pm
and I would have anticipated some intervention.
The problem is that, unless I have missed something, neither you nor anyone else has come up with an idea as to how the pairings might have been different.