Canadian Open Pairings Round 3

[Bindi Cheng]

I'm looking at the pairings from the site and I seriously don't understand how these pairings get done. First of all, I thought Canadian Open was advertised as a norm event. With no offense of my opposition, tell me how playing a 2200, a 2100 and 2000 in the first 3 rounds is supposed to promote the idea of a norm-hunting tournament for norm-seekers. Second of all, from what I understand, this tournament uses ghost pairings. In the beginning of round 1 there's supposed to be 4 quartiles. 3 ghost points for the top whoever, I think around 2200 and above and then 2 ghost points and then 1 ghost points and 0 ghost points for ever next descending quartile. in every round following, participants with ghost points are deducted 1 point and their remaining ghost points added onto their overall score. So I won both my games so I have 2 points and I'm in the first quartile so I would have 1 remaining ghost point left. That means I have 3 points. In round 3, I notice that I'm playing a 2000 player and again no offense to my opponent but I'm only dealing in rating purposes but somehow he has 3 points. Now if I understand correctly, he's supposed to have started with 2 ghost points and by round 3 he should have none left. That means since he had won both his previous games he should have 2 points overall with 0 ghost points by this time. How come he's listed as 3 points overall? I mean if we're just going by overall score from the very beginning I'd understand but if we're doing that, what's the point of ghostpoints and wouldn't it just be an actual open section with no sort of accelerated pairings?

[Jamin Gluckie]

Re: Canadian Open Pairings Round 3

I agree with you. I posted a similar thread asking the same questions.

There needs to be some transparency with how the pairings are being done.

[Erik Malmsten]

Re: Canadian Open Pairings Round 3

You have a high rating and so keep getting posted in the top half of the top group. For this round you're number 8 of 17 so play one of the lowest player who also has 1 ghost point.

If only one of the seven players higher rated than you still maintained their position you would have been shifted to board 1.

[Bob Armstrong]

Strange Hyper-Accelerated Pairing System?

Hi Bindi:

I am a bit confused too. It seemed at Rd. 2 that they accelerated for one more round than you have explained. They did not deduct a ghost point in Rd. 2. If you noted in the Rd. 2 pairings, the top quarter wiinners from Rd. 1 were still shown as having 4 rating points ( = their 3 ghost points plus their win ). So in Rd. 3, if they started deducting a ghost point, one would have expected the top quarter winners to have 4 pairing points ( = 2 ghost points + 2 win points ). But what do we see in the pairings list? The top quarter double winners have only 3 pairing points ( = 1 ghost point + 2 win points ). They seem to have deducted 2 ghost points all of a sudden for Rd. 3 ! Is this how the hyper-accelerated pairings system was meant to work?

Bob

[Erik Malmsten]

Re: Strange Hyper-Accelerated Pairing System?

Hi Bindi:

I am a bit confused too. It seemed at Rd. 2 that they accelerated for one more round than you have explained. They did not deduct a ghost point in Rd. 2. If you noted in the Rd. 2 pairings, the top quarter wiinners from Rd. 1 were still shown as having 4 rating points ( = their 3 ghost points plus their win ). So in Rd. 3, if they started deducting a ghost point, one would have expected the top quarter winners to have 4 pairing points ( = 2 ghost points + 2 win points ). But what do we see in the pairings list? The top quarter double winners have only 3 pairing points ( = 1 ghost point + 2 win points ). They seem to have deducted 2 ghost points all of a sudden for Rd. 3 ! Is this how the hyper-accelerated pairings system was meant to work?

Bob

Only rounds 1 and 2 were hyper, rounds 3 and 4 are just regular accelerated pairings to soften the transition to one big section where high-scoring 1500s will meet the masters like Bindi.

[Frank Dixon]

Re: Canadian Open Pairings Round 3

It seems the pairing system is similar to that used in Kapuskasing 2004. Round five was the settle-up round there.

It seems organizers are fixed on the idea of a single section for the Canadian Open, but want to generate norm chances at the same time. So, they resort to 'witchcraft' pairings which few, if any, players and observers can understand. There are about 250 players for this year's event, nine rounds, at least seven GMs, and many more IMs, FMs, and NMs.

Conclusion: the easiest and fairest way to generate norm chances at the Canadian Open is to use multiple sections!! A 2000+ section for strong players would have done precisely that this year. Many people understood that 30 or more years ago. Perhaps in my lifetime the Canadian Open will be standardized in a multi-section format, but I am not holding my breath on that. :)

[Alexandru Florea]

Re: Canadian Open Pairings Round 3

I have a question as well. I'm rated 2076 and supposed to play a 1077 for round three. No offence to my opposition, but to play someone rated 999 points below you at a canadian championship is absurd. Can someone please explain this?

Alex

[Kerry Liles]

Re: Canadian Open Pairings Round 3

I have a question as well. I'm rated 2076 and supposed to play a 1077 for round three. No offence to my opposition, but to play someone rated 999 points below you at a canadian championship is absurd. Can someone please explain this?

Alex

Perhaps this isn't the Canadian Championship...? ;)

Seriously though: your opponent may be vastly under-rated (otherwise how did he/she get to play you?).

[Egidijus Zeromskis]

Re: Canadian Open Pairings Round 3

Perhaps this isn't the Canadian Championship...? ;)

It is an OPEN where everybody can play. Just looking into the first-board pairings I think it was a good thing to use those hyperwiper :D

1r. IM Lawrence Day - GM Harikrishna Pentala
2r. GM Harikrishna Pentala - IM Michael Mulyar
3r. GM Eduardas Rozentalis - FM Aman Hambleton

[Kerry Liles]

Re: Canadian Open Pairings Round 3

It is an OPEN where everybody can play. Just looking into the first-board pairings I think it was a good thing to use those hyperwiper :D

1r. IM Lawrence Day - GM Harikrishna Pentala
2r. GM Harikrishna Pentala - IM Michael Mulyar
3r. GM Eduardas Rozentalis - FM Aman Hambleton

I agree that the accelerated pairings produced pretty worthwhile pairings in the first round; I am not so sure about the 2nd and 3rd rounds. I know the organizers wanted to provide norm chances (at least Bindi Cheng was of that impression) but having unknown players in the mix can lead to a strong IM playing someone who is grossly under-rated but performing well... I presume that could lead to problems with the norm process since the opponent is so under rated?

[Bindi Cheng]

Re: Canadian Open Pairings Round 3

Erik, these aren't accelerated pairings. In round 2 if it was, I wouldn't have played someone in the 2nd quartile while I was in the first. Technically I would have had 2 ghost points and 1 real point and he would have had 1 ghost point and 1 real point so it wouldn't even match. Furthermore, if you look at round 1 for the pairings, you'll see that my round 3 opponent had only 2 ghost points while I had 3. So currently in order for us to play each other that means he would still have 1 ghost deducted while I would have 2 ghost points deducted in order for us to both have a score of 3 points. Essentially these aren't real accelerated pairings like the the Canadian Open in Ottawa and if it were so, there is no way in any big tournament would you have in round 3 people separated by couple hundreds of points away from each other playing in the top boards. Accelerated pairings are meant to have people in the same rating category play each other early on and later rounds play the higher rated players if the higher rated players are not doing well and then that's when the rating difference will appear.

[Vlad Drkulec]

Re: Canadian Open Pairings Round 3

I don't like accelerated pairings in general. It seems to me that they only postpone the inevitable disparities until a later round. I am perplexed on how the pairings were done but I am not complaining about the fact that I am playing the highest rated player in the tournament next round in GM Harikrishna Pentala (2678).

[Larry Bevand]

Re: Canadian Open Pairings Round 3

I have a question as well. I'm rated 2076 and supposed to play a 1077 for round three. No offence to my opposition, but to play someone rated 999 points below you at a canadian championship is absurd. Can someone please explain this?

Alex

maybe we should consider a multi sectioned event based on ratings...naww...that would be sooo silly :). I understand the tradition...but then one must be prepared for the consequences.

Personally I don't believe accelerated pairings work very well when the field ranges from 2600 plus to unrated...I agree with Vlad that the system, under these circumstances, only delays the inevitable...and sometimes worse....but we always hope for better :)

Phil Haley (who I respect totally) was one of the inventors of the accelerated pairing system...perhaps we can have his perspective.

Larry

[Jean Hébert]

Re : Re: Canadian Open Pairings Round 3

maybe we should consider a multi sectioned event based on ratings...naww...that would be sooo silly :). I understand the tradition...but then one must be prepared for the consequences.

Personally I don't believe accelerated pairings work very well when the field ranges from 2600 plus to unrated...I agree with Vlad that the system, under these circumstances, only delays the inevitable...and sometimes worse....but we always hope for better :)

Phil Haley (who I respect totally) was one of the inventors of the accelerated pairing system...perhaps we can have his perspective.

This whole business makes me feel even better about having missed this Canadian Open (which by the way did not offer even free entry to the Canadian champion...). Staying near the pool is probably better, even for my chess.
It reminds me of an old Canadian Open in Toronto, I think it was 1976 but that would have to te checked. Then accelerated were supposed to be used but the TD did it for the first round only, unaware that doing that had no effect at all, just in effect switching rd 1 and 2! If memory serves me right, round 2 pairings were redone after a long delay. Things seem to be more complicated in 2010, but not much better.

[Jonathan Berry]

Re: Re : Re: Canadian Open Pairings Round 3

It reminds me of an old Canadian Open in Toronto, I think it was 1976 but that would have to be checked. Then accelerated were supposed to be used but the TD did it for the first round only, unaware that doing that had no effect at all, just in effect switching rd 1 and 2! If memory serves me right, round 2 pairings were redone after a long delay. Things seem to be more complicated in 2010, but not much better.

As it happens, just a couple of days ago I posted the docs for the 1976 Canadian Open here:
http://www.chesstalk.info/forum/show...5408#post25408
and here:
http://www.chesstalk.info/forum/show...5409#post25409

I do remember that there was at least one tournament where the TD neglected to apply the necessary technique for a second round of Haley Accelerated Pairings, but don't remember which event(s) it (they) was (were). I don't think so, and the linked report does not talk about that, but who knows, maybe I sanitized my report in 1976? The news was dire enough without stressing what might have been a simple accident.

I do have a modest proposal, though. The proposal is that we no longer use the term Accelerated Pairings (capitalized). For the system that we know, it is Haley Accelerated Pairings. Phil Haley invented this, it's his system, that's what he called it, it's Canadian, it was years ahead of its time, it is a work of genius, so dammit let's give credit where credit is due. The general term might be an accelerated pairing system (no caps). The main other such system is Système Suisse Accéléré Degréssif which is used at the annual Cappelle La Grande tournament in France. It is also a work of genius. IMHO. But it does not cover every base. For example, aside from the leading scoregroups, it does nothing to reduce the yo-yo effect.

Here is an old article that gives some discussion about the issues of accelerated pairing systems and tradeoffs involved in single-section events.

I don't know what system is being used at the 2010 Canadian Open in Toronto. The tournament documents say to contact TD Hal Bond for details, but I have not done that. Elsewhere I put it under the category of hyper-accelerated pairing system, which in general terms is a system that divides in smaller than halves and longer than two rounds. Système Suisse Accéléré Degréssif is a hyper-accelerated pairing system, but Haley Accelerated Pairings are not.