number of sections at the Ottawa Spring Open

[Aris Marghetis]

Hi there, I have been receiving widely disparate feedback on the number of sections at my Ottawa weekend events. On the one hand, some people really like the many small sections, because they are guaranteed 5 great battles, and they feel they are completely in control of their quest for prize money (i.e. cannot have an unlucky upwards pairing compared to someone else going for the same money). On the other hand, some people say that they would like their pairings to be less predictable, and not with the same players thay have already often played against. So, please respond to this poll within a day or so, thanks!

This tight polling timeline is because I have to officially post the event details ASAP. We already have 23 players registered, just knowing the dates of March 4th-6th, but we need to officially decide on the number of sections for everyone else who might be interested. Note the the $30 triple-early discount period expires at the end of January, so that is yet another reason to give people all of the details by this Thursday evening at the latest.

Option 1: many sections: as is now, a section almost every 200 rating points

Option 2: just 2 sections: with manual accelerated pairings for the first 2 rounds
p.s. there would be class prizes almost every 200 rating points

Finally, I have been asked by 4 people now to consider adding CYCC-qualifier sections to this event, in a different part of the building. The only way that I could do this is if I had quite a stable of dedicated volunteers to use as arbiters, messengers, and parent crowd control. If you are interested in helping out that weekend, please email me separately.

Thanks and regards, Aris.

[Garland Best]

Re: number of sections at the Ottawa Spring Open

Personally Aris, I would simply merge a couple of the smaller sections. For instance in your last tournment merging the U1800 section with either the U2000 or U1600. From a pairing sense in the 1st round there is only a 200 point mismatch which is no sure win.

[Aris Marghetis]

Re: number of sections at the Ottawa Spring Open

Personally Aris, I would simply merge a couple of the smaller sections. For instance in your last tournment merging the U1800 section with either the U2000 or U1600. From a pairing sense in the 1st round there is only a 200 point mismatch which is no sure win.

Hi Garland, thanks for your feedback. Just a small clarification, is that when I have had less sections, some of the pairings in the first couple of rounds can be at least 300 rating points apart. Also, when a combined section gets quite big, sometimes by fluke, one person at the top of a section will get apparently tougher pairings than someone right beside them in initial ranking, but that is part of the excitement of a Swiss System based on initial ratings, especially when some ratings are bumped up by FIDE, FQE, etc.

As I write this, the score in the poll is 6-4 in favour of the many sections, which is what I expected, as the silent majority seem to put more value on having a close battle every single round. However, virtually all of the written feedback by email directly to me, is in favour of trying less sections, and the unpredictability that would inject into the suite of opponents one would face for the weekend. I also expected this input, which I would like to at least try again. However, whereas my old cutoff was 2000, I think that it is time to try FIDE rating down to 1800. When I look at how that would have worked out over my last few events, some of the rating gaps for the first pairings would have been over 400 rating points, which in my opinion, would be a waste of time for most players.

Therefore, unless someone can convince me otherwise in the next few hours (and I really am open to any further input on this), I am leaning heavily towards having 2 sections, with the cutoff at 1800, and with accelerated pairings in the first 2 rounds.

That should satisfy both the interests people who want a wider variety of opponents, and to some extent, people who want the closest games possible. In fact, it might help the latter, as on any given weekend, it seems to me that at least a quarter of the players perform nowhere near their actual rating (either much better or much worse).

I propose to have the same acceleration in both sections. In the 1st round, I will split each section by rating into 3 sub-groups, and add 1, 1/2, 0 points respectively. For the 2nd round, I will split again by rating, into 2 sub-groups, with the higher-rated half getting an extra half-point. I believe that this model should protect us at least from pairings where the players are say 500 points apart, but not overly segregate players.

[Aris Marghetis]

Re: number of sections at the Ottawa Spring Open

Some players have been emailing me, asking for the event details for the Ottawa Spring Open. I am just finalizing them now (really, only the sections stuff is changing), and will send them today to the EOCA webmaster to post. Until then, you could check out the event details from my last event, and just change the dates to March 4th-6th. The entry fees, Amateur option, timing of rounds, etc., all that regular stuff will be the same.

[Kevin Pacey]

Re: number of sections at the Ottawa Spring Open

Hi Aris

If you decide to go to just 2 sections, would there be a firm rating cutoff of, say, a regular CFC rating of 2000 ? That could be best, as I would guess that you would keep the top section FIDE rated (though please indicate if that's so, too).

[Aris Marghetis]

Re: number of sections at the Ottawa Spring Open

Hi Aris

If you decide to go to just 2 sections, would there be a firm rating cutoff of, say, a regular CFC rating of 2000 ? That could be best, as I would guess that you would keep the top section FIDE rated (though please indicate if that's so, too).

Hi Kevin, yes, there would be a firm cutoff at 1800, with the 1st section rated CFC & FIDE, and the 2nd section (<1800) being rated only CFC. With accelerated pairings in the first 2 rounds, I figure this model should suit everyone somehow. Any concerns?

[John Upper]

Re: number of sections at the Ottawa Spring Open

Therefore, unless someone can convince me otherwise in the next few hours (and I really am open to any further input on this), I am leaning heavily towards having 2 sections, with the cutoff at 1800, and with accelerated pairings in the first 2 rounds.

One reason I usually play as an Amateur in weekend swisses is that it really pisses me off when I go into the last round leading my section by 1/2 a point and get paired against someone who is 350 points higher-rated than the opponents of the players who are only 1/2 a point behind me. And that's not the end of the problem: this rating mismatch is also being suffered by that much lower-rated opponent of the player who is trying to catch me; so there are at least two people who will think the difference between 1st and 2nd/3rd was determined by that last-round pairing.]

It seems to me that the larger the rating spread within a section the more probable that sort of crummy last-round pairing.

[I'm not sure, but my hunch is that accelerated pairings don't make mismatches go away, they just push them backwards toward the money rounds.]

If I were really keen on "winning" my section I would try to avoid that sort of last round pairing by playing to draw a winning position in round 3 or 4... but I think that sort of gaming the pairing system is contemptible (though totally within the rules).

Any ideas for ways to avoid this?

I know I could try think that these "money round mismatches" make the event more exciting. But if I want excitement from randomized outcomes that augment or thwart my skill, I believe I would be better off playing poker.

[Aris Marghetis]

Re: number of sections at the Ottawa Spring Open

One reason I usually play as an Amateur in weekend swisses is that it really pisses me off when I go into the last round leading my section by 1/2 a point and get paired against someone who is 350 points higher-rated than the opponents of the players who are only 1/2 a point behind me. And that's not the end of the problem: this rating mismatch is also being suffered by that much lower-rated opponent of the player who is trying to catch me; so there are at least two people who will think the difference between 1st and 2nd/3rd was determined by that last-round pairing.]

It seems to me that the larger the rating spread within a section the more probable that sort of crummy last-round pairing.

[I'm not sure, but my hunch is that accelerated pairings don't make mismatches go away, they just push them backwards toward the money rounds.]

If I were really keen on "winning" my section I would try to avoid that sort of last round pairing by playing to draw a winning position in round 3 or 4... but I think that sort of gaming the pairing system is contemptible (though totally within the rules).

Any ideas for ways to avoid this?

I know I could try think that these "money round mismatches" make the event more exciting. But if I want excitement from randomized outcomes that augment or thwart my skill, I believe I would be better off playing poker.

Yes, I noticed those things too, which is why I went to the more sections. I am quite sure that some players do a "Swiss Gambit" for their class, but I am open to ideas on how to discourage it, aside from small classes and no 2nd place prizes for any class.

The acceleration is intended to avoid really lopsided pairings in the first couple of rounds, but I agree, I do not think that they will help for the last round vulnerability.

I am on the fence on this, and need to decide within the next couple of days at most. My gut instinct at this moment, is to try 2 big sections in March, with accelerated pairings, which I have never tried before for a weekender, and then re-poll the players.

[Kevin Pacey]

Re: number of sections at the Ottawa Spring Open

Hi Kevin, yes, there would be a firm cutoff at 1800, with the 1st section rated CFC & FIDE, and the 2nd section (<1800) being rated only CFC. With accelerated pairings in the first 2 rounds, I figure this model should suit everyone somehow. Any concerns?

Hmmm

Well, I voted for multiple sections anyway. If you do proceed with having just 2 sections, I think a CFC regular rating of 2000 is a better cutoff for the top FIDE rated section (even though the top section likely will be smaller) since this at least means a better chance of playing more FIDE rated players with a decent FIDE rating.

It'll be impossible for you to please everyone regardless. Plus most likely are putting their own interest as a participatant in a cash prize event first. My advice would have been to stick with whatever format that your events in the past drew the most entrants to. People get used to event formats, and tinkering can backfire (if it ain't broke...). I would have been happy with one of the older formats that had an U2400 prize. I may be wrong, but I suspect your event attendence has slipped slightly for events held since this format was abandoned. The final choice of format is of course yours as organizer (though perhaps a discussion at EOCA meeting(s) might make for a helpful consultation process).

[Aris Marghetis]

Re: number of sections at the Ottawa Spring Open

Hmmm

Well, I voted for multiple sections anyway. If you do proceed with having just 2 sections, I think a CFC regular rating of 2000 is a better cutoff for the top FIDE rated section (even though the top section likely will be smaller) since this at least means a better chance of playing more FIDE rated players with a decent FIDE rating.

It'll be impossible for you to please everyone regardless. Plus most likely are putting their own interest as a participatant in a cash prize event first. My advice would have been to stick with whatever format that your events in the past drew the most entrants to. People get used to event formats, and tinkering can backfire (if it ain't broke...). I would have been happy with one of the older formats that had an U2400 prize. I may be wrong, but I suspect your event attendence has slipped slightly for events held since this format was abandoned. The final choice of format is of course yours as organizer (though perhaps a discussion at EOCA meeting(s) might make for a helpful consultation process).

Hi Kevin, thank you very much for your feedback. You are right about not being able to please everyone, lol! Between events, I receive a remarkable amount of feedback from people who do not attend, but would like to pass on to me what it would take to get them to attend. Over the last couple of events, which averaged under 60 players, and thus were small even by Ottawa standards, a common theme was too many sections. I notice though on my poll from yesterday, that the vote is now running 8-4 for many sections, but there is a difference between what people prefer, and what people will focus on as the deciding factor whether to attend or not. I have decided to switch things up this event, and really test the feedback from some players who say they would attend only if they had a chance to play much better players, assuming of course that they are having a good event to match their score group, and so on. In addition, many players U2000 would like to develop their FIDE ratings, and I agree it is time to try that for them as well. The result is 2 big sections that will be much easier for me to just run, with regards to pairing score groups, both Amateur and section floaters, and so on.

The acceleration in the first couple of rounds should also dull some of the big-section effects. After the tournament, I would like to sit down with a couple of key regular players like yourself, to review any effects. I would like to go with something that we are very confident with for the 2011 Eastern Ontario Open, a 6-rounder around the Canada Day long weekend. I suspect it would be back to more than 2 sections?! ;)

[Kevin Pacey]

Re: number of sections at the Ottawa Spring Open

Hi Kevin, thank you very much for your feedback.

I can give you more feedback right now, though it's likely stuff you've already noticed or considered.

I seem to recall that the first Ottawa EOCA event that a U2400 prize was no longer offered for the top section (then O2000, I believe), there was a severe shortage of players who entered who were rated between 2200 and 2300, and event attendence as a whole was down as a result. Don't recall if this continued until the next change of format to Ottawa EOCA events, when the top section was made completely O2200. From that point the trend of mostly 2400+ (usually titled!?) players taking the top section prizes continued, I think, with instances of (apparently) underrated 2200 players like Humphreys exceptionally striking pay dirt.

For me, making the top section O2200 meant that I couldn't afford to be out of form or I would finish at the bottom. When the top section was O2000, I could almost count on having some 'easier' games. I was glad they weren't too 'easy', such as having opponents rated 18xx, since I could gain few CFC rating points (nowadays none, with the removal of participation rating points). Same consideration for any FIDE ratings involved - though with FIDE ratings being allowed to be U2200 nowadays, I value my FIDE rating rather less (but still hope for FM title, which is harder to get sooner with the top section to be O1800).

To finish with, if you happen to have a goal of any time soon getting your Ottawa weekend events to an attendence of 100+ players (occasionally reached in the 1990s in Ottawa, when overall CFC membership was higher), you may have to attempt more than trying out various event formats, such as trying to advertise to the general public (i.e. proactively recruit non-CFC members), or trying to get sponsorship. Individuals involved with EOCA governance could be involved with such attempts, besides yourself, assuming such attempts have not been made with a full court press effort for some time.

[Evan Frangakis]

Re: number of sections at the Ottawa Spring Open

Hi Aris,

I haven't yet had the pleasure to play at one of your events, however, as this is a general question, I thought I'd add my two cents.

Arguably, the most popular weekend tournament in Canada is the Mauricie Open in Trois Rivieres, QC. It regularly draws close to 300 participants and has section cut-offs every 200 points. A sure reasons for its popularity is the fact that every player has a realistic shot at a prize. Additionally, one can also expect to play closely matched opponents - which makes the experience more pleasureable. These are two very positive reasons for multiple sections.

It seems the main objection to multiple sections is the potential for playing the same opponents on a recurring basis. I have played at Trois Rivieres for the past 5 years and I can't remember playing the same opponent twice. Furthermore, in the lower sections, it really does not seem to be a problem, because as players improve (or otherwise) they move up and down in ratings and this would add variety to the pool of opponents in a given section. Also, it is inevitable that if one keeps improving (say beyond 2000), he would then end up meeting the same opponents in that section, so this is a fact of life like any competitive endeavour (ie. the best end up playing the best).

Best regards and keep up the good work!

[Aris Marghetis]

Re: number of sections at the Ottawa Spring Open

Hi Aris,

I haven't yet had the pleasure to play at one of your events, however, as this is a general question, I thought I'd add my two cents.

Arguably, the most popular weekend tournament in Canada is the Mauricie Open in Trois Rivieres, QC. It regularly draws close to 300 participants and has section cut-offs every 200 points. A sure reasons for its popularity is the fact that every player has a realistic shot at a prize. Additionally, one can also expect to play closely matched opponents - which makes the experience more pleasureable. These are two very positive reasons for multiple sections.

It seems the main objection to multiple sections is the potential for playing the same opponents on a recurring basis. I have played at Trois Rivieres for the past 5 years and I can't remember playing the same opponent twice. Furthermore, in the lower sections, it really does not seem to be a problem, because as players improve (or otherwise) they move up and down in ratings and this would add variety to the pool of opponents in a given section. Also, it is inevitable that if one keeps improving (say beyond 2000), he would then end up meeting the same opponents in that section, so this is a fact of life like any competitive endeavour (ie. the best end up playing the best).

Best regards and keep up the good work!

Hi Evan, thanks for your 2c worth! I completely agree with you, and I have been inspired by that tournament. The hitch though in Ottawa has been, especially between about 1600 and 1800, and then 2000 and 2200, that we do not get enough numbers to ensure many different opponents, especially across 5 events a year. I believe that one reason that 3-Rivieres gets so many more participants is a much broader and deeper prize fund, that has been built on a foundation of sponsorship across many years. They sure seem to have their whole act together. For my events, I feel that I am a good TD, a decent Organizer, and a useless sponsorship-guy, lol! In another post in this thread, Kevin was mentioning sponsorship, and doing it at the EOCA level, etc. For years now, and this offer still stands, I would welcome anyone, or any team of people, with wide open arms, who would like to team up with me to get us sponsorship. Someone make me an offer! :)

[Egidijus Zeromskis]

Re: number of sections at the Ottawa Spring Open

who say they would attend only if they had a chance to play much better players

In a case those players come only to play and not pursuit prizes, you may try offer 6-players round-robins sorted in groups by rating :)

As an out-of-city player, I mostly enjoyed a late start of Saturday's first round (1 pm!?) It gave a possibility to tour Ottawa.

good luck w/ your experiments.

[Aris Marghetis]

Re: number of sections at the Ottawa Spring Open

In a case those players come only to play and not pursuit prizes, you may try offer 6-players round-robins sorted in groups by rating :)

As an out-of-city player, I mostly enjoyed a late start of Saturday's first round (1 pm!?) It gave a possibility to tour Ottawa.

good luck w/ your experiments.

Yes, it's funny what works sometimes! For example, way more people (like yourself) have appreciated the late start on Saturday than have complained about it (some do).

Perhaps the best reason of all though is that I always take my wife out to brunch on Saturday morning. If I did not ensure that, then seriously, I would not be able to TD! ;)