Uxbridge Chess Congress : 18th - 20th September 2009

[Alex Holowczak]

Blimey. I sense there's a lot of unhappiness about this. Were there a lot of complaints towards the organisers about this from people at the end of the day, or something?

The idea of accelerated pairings is to reduce the chance of that happening, or so I'm led to believe. It's daft at the British Championship where there are plenty of rounds (11). There were almost as many players in the Open in Uxbridge, yet only 5 rounds. When you have 66 players entering a tournament designed to find a winner for 32 or fewer, I can understand why accelerated pairings were used. Stewart has posted on here about how the Masters had a hyper-accelerated system because of having so many players and so few rounds.

Two players ending up on 5/5 is perhaps fairer, but I can understand why tournament organisers would want to try things to avoid it.

[Sean Hewitt]

Alex I don't understand your point at all. Accelerated pairings wouldn't stop two people getting to 5/5. Do the math.

I think that it is you who needs to "do the math" as you put it.

Accelerated pairings do not prevent two players from getting to 5/5 but they do make it far less likely.

[Sean Hewitt]

Blimey. I sense there's a lot of unhappiness about this. Were there a lot of complaints towards the organisers about this from people at the end of the day, or something?

No!

The idea of accelerated pairings is to reduce the chance of that happening, or so I'm led to believe.

Absolutely right.

[Ian Lamb]

if the tourny leader agrees a draw in the last round in a tourny with 73 in he must expect to be caught up that just life if he doesnt he has to win with 5/5 to be 1st alone . That weekend chess for you with big numbers and so few rounds.He must of known when he took the draw that there was a chance he not win it alone.

[IM Jack Rudd]

Blimey. I sense there's a lot of unhappiness about this. Were there a lot of complaints towards the organisers about this from people at the end of the day, or something?

There were not. Slavin himself seemed to take it in good humour, as I suppose you would if you'd just posted a TPR of 2721 to get your rating within touching distance of the IM title requirement.

[Alex Holowczak]

Blimey. I sense there's a lot of unhappiness about this. Were there a lot of complaints towards the organisers about this from people at the end of the day, or something?

No!

Oh good. The impression I got from the posts were that there were riots in the playing hall before the prize giving.

[Ian Lamb]

not one section was won with 5/5 which is very seldom you get that in a weekender because players try really hard to win as you have to to get first prize in a weekend tourny with big numbers.

[Sean Hewitt]

Oh good. The impression I got from the posts were that there were riots in the playing hall before the prize giving.

No - nothing about the acceleration.

There was some commotion about the final round pairings with one player arguing vociferously that we had got the pairings wrong (even at one point suggesting that we may have done so deliberately to favour another player). He was not prepared to listen to the fact that the pairings were correct, nor the reasons why.

He did get a letter explaining the swiss pairing rules, together with the reasons the pairings were as they were. These reasons were also pinned onto the wall of the playing hall.

After the player had read the explanation he conceded that the pairings were "probably right."

[Ian Thompson]

Accelerated pairings do not prevent two players from getting to 5/5 but they do make it far less likely.

If the accelerated pairing system makes it far less likely that 2 players will finish on 5/5, doesn't that imply that it is more likely that several players will finish equal first on a less than 100% score than would be the case with a non-accelerated pairing system? Is that preferable to the top two seeds finishing on 5/5 without having played each other?

[IM Jack Rudd]

If the accelerated pairing system makes it far less likely that 2 players will finish on 5/5, doesn't that imply that it is more likely that several players will finish equal first on a less than 100% score than would be the case with a non-accelerated pairing system?

Yes.

Is that preferable to the top two seeds finishing on 5/5 without having played each other?

Yes.

[isaac wallis]

By round 3 the damage had been done. Just as Peter Tart nearly won a previous time, the accelerated pairings again led to a lower rated player sneaking through without playing the strongest players in a way that wouldn't have happened with normal pairings. Things like this don't happen with traditional pairings.

The upfloat remains irrelevant -anyway it was to a weaker player

Also my maths is correct.

Jack - Knowledge increased. Correction. Two 5/5 scores happens rarely, probably even more rarely in an Open.

The fact is if a player can play who James Adair did and win such a strong tournament, the pairing system is faulty.

[Alex Holowczak]

By round 3 the damage had been done. Just as Peter Tart nearly won a previous time, the accelerated pairings again led to a lower rated player sneaking through without playing the strongest players in a way that wouldn't have happened with normal pairings. Things like this don't happen with traditional pairings.

No, other evils happen, which as Jack suggests, are less desirable.

The upfloat remains irrelevant -anyway it was to a weaker player

The upfloat was a fundamental part of the acceleration process, it would appear. In a traditional pairing process, it would be considered "unlucky", so I don't see why that would change here.

The fact is if a player can play who James Adair did and win such a strong tournament, the pairing system is faulty.

Well, great minds have tried to come up with a fault-proof pairing system for much of the last century, probably. The fact of the matter is that with selective scheduling of opponents in tournaments (i.e. when you don't play everyone else taking part), you're always going to get freak results now and again, whether the pairings have acceleration or not. Be it two players on 5/5, or a weak strength of schedule. The current thinking is that the former is less desirable.

[Peter Rhodes]

The Neustadtl scores are as follows :

15.00 Alexei Slavin (4/5)
12.25 James Adair (4/5)
13.25 John Cox (3.5/5)
12.00 Keith Ardell (3.5/5)
10.00 Alexander Chernaiev (3.5/5)


We can send a man to the moon, we can map the genome, we can send a message across the world in mere milliseconds, yet we cannot devise a system that ensures the 2nd place contestant plays stronger competition than the 3rd place one ?

Call me sceptical, but I am not convinced the system cannot be improved.

What is now proved was once only imagined. (Blake).



EDIT :
By the way - James Adair's elo performance was 2418 which is pretty damn fine so gratz to him.

[Ian Thompson]

Is that preferable to the top two seeds finishing on 5/5 without having played each other?

Yes.

Why?

[Ola Winfridsson]

The whole issue boils down to the number of rounds. As has already been pointed out, once you reach a certain number of participants these things are occasionally unavoidable, regardless of whether an accelerated pairing system is used or not (an outsider playing out of his or her skin).

The only surefire way of avoiding this is to either cap the number of participants or increase the number of rounds, which is more less impossible in any normal weekender without changing the time controls in the initial rounds (e.g. play seven rounds with the first three being rapidplay and the last four 90min + 30 sec.). Would that even be allowed in a Fide-rated event?

Simply put, it's all Sean's fault for organising such nice events attracting too many participants!