Suitability of acceleration?

Discuss anything you like about chess related matters in this forum.
Wadih Khoury
Posts: 604
Joined: Sun Jul 12, 2020 8:14 pm

Suitability of acceleration?

Post by Wadih Khoury » Fri May 14, 2021 10:19 am

I would like some advice from experienced organisers/arbiters:

As some know, I am organising a standard tournament for juniors at the end of the month. I have not yet finalised the format of the event beyond rounds (5) and timing (60+30).
Keep in mind that my main goal is to have the children play enjoyable as many games as possible relevant to their levels.

My options are:
  • Multiple all-play all sections: rejected as this requires precise multiples of 6, and any no-show or bye is extremely disruptive
  • 2 grade based sections: rejected as grades are currently 18+ months out of date, online grade conversion is an art more than a science, and I'd rather not arbitrarily decide who belongs where (and possibly enter in arguments). Also I risk having 2 odd numbered sections.
  • Single Swiss section: given the field at hand, I expect the first round to be un-interesting games with 400+ estimated rating difference on many boards. Also, I wouldn't be surprised that some players start with 3 defeats in a row and would be disheartened/resign. In a normal tournament, that would be ok but I want to ensure they all play as many relevant games as possible.
  • Acceleration (Baku or otherwise): seems to be an ideal solution: it ensure initially that strong players and weaker player play each others (more relevant games), and eventually if a player does very well/poorly, they will be paired against stronger/weaker players. Also, the bottom players who get a 1 point bye will not automatically face the top half of the board. The main issue I can see is explaining it to the participants.
So the questions are:
  • If I use acceleration, does it have to be Baku (which I believe is the FIDE sanctioned method) in order to be graded by the ECF?
  • will Baku achieve my goals on a 5 rounds tournament, with players ranging from 1300 (est.) to 1900+(est.)? As I understand that will give 2 rounds with 1 point acceleration and 1 round with 0.5 point, leaving 2 un-accelerated rounds (1-1-1/2-0-0 ) ?
  • Assuming I can use a non-Baku method, would any other acceleration work better? (maybe 1-1-0-0-0 or 1-1/2-0-0-0 ? )
Again as a reminder, this is not a norm tournament, but a "return to over the board" standard junior tournament. The main goal is for the juniors to enjoy as many games as possible at their level.

Alex McFarlane
Posts: 1757
Joined: Sat Aug 02, 2008 8:52 pm

Re: Suitability of acceleration?

Post by Alex McFarlane » Fri May 14, 2021 10:45 am

Hi Wadhi,

I do not know what number of entries you are looking for. For a 5 round tournament I would suggest that if you are accelerating then you will need about 60 entrants minimum. The reason that I say this is that if you have a small number of entrants then you run the risk of the last round becoming meaningless as the top players will probably have already played each other and will therefore be playing weaker players from lower down the tournament. A last round where there is little to play for, other than the hope that a rival will lose to a weak player, is not the way to finish an event.

Another significant problem that you will have is trying to explain the system to parents and players. The parents may think it is some sort of witchcraft to stop their little Jonny/Jemima from winning. (I exaggerate only slightly here - but I have experience of dealing with parents/coaches/teachers who didn't understand a system and thought it was designed to favour one of the competitors.)

One thing I did do in the dim and distant past which was met with general approval was to run day one as effectively a rapidplay (5/6 rounds). At the end of day one I took the top 8 out of that tournament to play 3 rounds at the slower rate and the others continued in the Rapidplay. Of course you could divide the others up into sections too for 3 rounds of slower play, and this may well be the better option if your numbers are small. Trophies for the 'section' winners prove popular as even the worst players can win medals!

Matthew Turner
Posts: 3600
Joined: Fri May 16, 2008 11:54 am

Re: Suitability of acceleration?

Post by Matthew Turner » Fri May 14, 2021 10:50 am

Wadih,
My advice would be to keep it simple.
Is anyone really going to worried about playing someone 500 points different in round 1?
Kids understand what is going on, so consider the kid who has 0/3, how are they going to feel in they have had a rubbish draw and played lots of strong players, or if the pairings have been 'modified' so that the games are more equal matches?
So, I'd stick with a simple Swiss

You can have whatever pairings you want and have it ECF graded. We even had Crouch pairings once which went 1v2 4v3 5v6 8v7 etc.

Wadih Khoury
Posts: 604
Joined: Sun Jul 12, 2020 8:14 pm

Re: Suitability of acceleration?

Post by Wadih Khoury » Fri May 14, 2021 11:07 am

Alex McFarlane wrote:
Fri May 14, 2021 10:45 am
Hi Wadhi,

I do not know what number of entries you are looking for. For a 5 round tournament I would suggest that if you are accelerating then you will need about 60 entrants minimum. The reason that I say this is that if you have a small number of entrants then you run the risk of the last round becoming meaningless as the top players will probably have already played each other and will therefore be playing weaker players from lower down the tournament. A last round where there is little to play for, other than the hope that a rival will lose to a weak player, is not the way to finish an event.

Another significant problem that you will have is trying to explain the system to parents and players. The parents may think it is some sort of witchcraft to stop their little Jonny/Jemima from winning. (I exaggerate only slightly here - but I have experience of dealing with parents/coaches/teachers who didn't understand a system and thought it was designed to favour one of the competitors.)

One thing I did do in the dim and distant past which was met with general approval was to run day one as effectively a rapidplay (5/6 rounds). At the end of day one I took the top 8 out of that tournament to play 3 rounds at the slower rate and the others continued in the Rapidplay. Of course you could divide the others up into sections too for 3 rounds of slower play, and this may well be the better option if your numbers are small. Trophies for the 'section' winners prove popular as even the worst players can win medals!
Thank you Alex for your advice.
I expect to have 17-32 players.

The players all are keen on standard, and I want to give them the opportunity, so that rules out the rapid-standard hybrid model.
The goal of the tournament is not simply to give out trophies (there will be trophies of course, and I'll probably have grading/age medals too to keep everyone happy and motivated) but to give the juniors the opportunity to play standard games after an 18+ months hiatus, ideally meaningful ones.

The challenge is that I can broadly estimate to have one third of strong players (let's say 1800-2000), one third of unknown (players with old ratings of 1400-1600 who are likely for some to be 1800+ while others remained at 1500) who should have the opportunity to play the "strong" pool if they do well and one third of unrated/weaker players who may or may not be dark horses.
The thing I want to avoid is the bottom third to be trashed in the first 2-3 rounds. Both players on the board will not enjoy the experience, and that only leaves 2 or 3 meaningful games over 5 rounds.
I feel that acceleration would solve the issue: a well performing low/middle pool player will have the opportunity to play the stronger players, weaker players will be protected from getting smashed, and there are enough middle/strong players to provide meaningful games at the top (I guess only if a strong player loses his 3 accelerated rounds does he have a chance of playing less meaningful games). As a bonus stronger players will face from round 1 hard games, which is the point.

My main goal is to give as much as possible 5 meaningful (in the sense of challenging) games to the juniors after 18 months without OtB.

I agree that it could be a pain to explain, I am a parent myself :lol:

Joseph Conlon
Posts: 339
Joined: Thu Jun 06, 2019 4:18 pm

Re: Suitability of acceleration?

Post by Joseph Conlon » Fri May 14, 2021 11:32 am

I have said this to you by email, but I think acceleration is a good idea. At the moment OTB chess is hard to set up and it is important not to waste the opportunities for OTB play through dud games from which neither side learns very much.

My view is very much that at this time, organisers of OTB events should be willing to be creative in terms of ensuring best chess playing bang-for-the-buck - its a good chance to think about whether there are more appropriate tournament models for the current climate than a traditional Swiss.

I would even consider double acceleration, with the top third starting on +2 points, the next third on +1 point, and the bottom on 0.

I think for ECF rating, it doesn't matter what pairing system you use, so you can define and use Khoury acceleration if you want ;)

Roger de Coverly
Posts: 21301
Joined: Tue Apr 15, 2008 2:51 pm

Re: Suitability of acceleration?

Post by Roger de Coverly » Fri May 14, 2021 11:37 am

Wadih Khoury wrote:
Fri May 14, 2021 11:07 am

The thing I want to avoid is the bottom third to be trashed in the first 2-3 rounds.
A Swiss system is effective at finding losers as well as winners. I doubt there would be more than one player on zero after three rounds. In any event, pairing systems rely on the players being ranked. If you have to rely on out of date or multiply sourced ratings to get a seeding order, there's going to be a degree of randomness about the pairings particularly in the first round. That's also true with Baku, who gets the dummy point and who doesn't could look random.

Alex McFarlane
Posts: 1757
Joined: Sat Aug 02, 2008 8:52 pm

Re: Suitability of acceleration?

Post by Alex McFarlane » Fri May 14, 2021 12:08 pm

My view is that with the numbers concerned acceleration will not work. You have the great risk of a 'bad' last round. Leaders being downfloated for relatively meaningless games and potentially weak players being upfloated to heavily lose their last round (the one you perhaps want to be memorable to encourage them back).
Acceleration is designed for two purposes (a) to obtain title norms and (b) to obtain only one 100% score when the numbers are greater than 2^n.
Neither of those apply.
The 4NCL 3N has used a 'decelerated' swiss in the past to keep the top games for the final weekend. A modification of this might work in your case.
For example, divide into 4 quarters.
Rd 1: Q1 v Q2, Q3 v Q4
Rd 2: Q1 v Q1, Q2 v Q2, Q3 v Q3, Q4 v Q4
Rd 3 as Rd 1
Rds 4 & 5 Normal swiss
Not ideal (Rd 2 in the 4NCL was Q1 v Q3 and Q2 v Q4) but I think it may be the closest you will get to what you want. The big downside is that I don't know of any software which will do this.

Roger Lancaster
Posts: 1910
Joined: Tue Mar 17, 2015 2:44 pm

Re: Suitability of acceleration?

Post by Roger Lancaster » Fri May 14, 2021 1:10 pm

While I take Alex's points, I think I'd disagree. In the absence of any draws, which tend to be few and far between in junior Swiss events where the competitors have a wide range of abilities, a normal pairing system would reduce a field of 32 or fewer to one outright winner with 5/5. In effect, the tournament becomes a knockout event because, once a junior loses a single game, all his/her remaining games become - to use Alex's phrase - "relatively meaningless". Even the occasional draw wouldn't affect this because even a winning score of 4.5/5 would be unattainable by anyone who had lost a game.

Paul McKeown
Posts: 3732
Joined: Thu Apr 12, 2007 3:01 pm
Location: Hayes (Middx)

Re: Suitability of acceleration?

Post by Paul McKeown » Fri May 14, 2021 10:23 pm

Wadhi, with the numbers that you intend to have participating, you absolutely should not consider accelerating Swiss pairings. As Alex M. has made clear, you will just end up with meaningless pairings for the top boards in the final round(s).

Swiss pairings will normally work well to determine a winner from 32 players. Unfortunately there is no pairing system which can balm the wounds of a player that loses all their games, nor prevent them from doing so, either. The best that a parent or coach can do is to carry out some analysis putting on some positive spin, showing a few tactics that the player missed and telling them that if they had spotted a few of these tactics they might well have won where they had lost. "You played well, you just had a bit of bad luck, you'll do better next time!"

Paul McKeown
Posts: 3732
Joined: Thu Apr 12, 2007 3:01 pm
Location: Hayes (Middx)

Re: Suitability of acceleration?

Post by Paul McKeown » Fri May 14, 2021 10:25 pm

Roger Lancaster wrote:
Fri May 14, 2021 1:10 pm
While I take Alex's points, I think I'd disagree. In the absence of any draws, which tend to be few and far between in junior Swiss events where the competitors have a wide range of abilities, a normal pairing system would reduce a field of 32 or fewer to one outright winner with 5/5. In effect, the tournament becomes a knockout event because, once a junior loses a single game, all his/her remaining games become - to use Alex's phrase - "relatively meaningless". Even the occasional draw wouldn't affect this because even a winning score of 4.5/5 would be unattainable by anyone who had lost a game.
Roger, I have to say I find your argument a little odd. It won't matter what the pairing system is, anyone who loses a game in a five rounder is going to struggle to catch up with the leaders.

Paul McKeown
Posts: 3732
Joined: Thu Apr 12, 2007 3:01 pm
Location: Hayes (Middx)

Re: Suitability of acceleration?

Post by Paul McKeown » Fri May 14, 2021 10:32 pm

Wadih Khoury wrote:
Fri May 14, 2021 10:19 am
If I use acceleration, does it have to be Baku (which I believe is the FIDE sanctioned method) in order to be graded by the ECF?
ECF rating doesn't care about your pairing method at all. Do what you want, just ensure that everyone enjoys themselves, and the Laws are upheld.

Angus French
Posts: 2151
Joined: Thu May 15, 2008 1:37 am
Contact:

Re: Suitability of acceleration?

Post by Angus French » Fri May 14, 2021 10:59 pm

Paul McKeown wrote:
Fri May 14, 2021 10:32 pm
Wadih Khoury wrote:
Fri May 14, 2021 10:19 am
If I use acceleration, does it have to be Baku (which I believe is the FIDE sanctioned method) in order to be graded by the ECF?
ECF rating doesn't care about your pairing method at all. Do what you want, just ensure that everyone enjoys themselves, and the Laws are upheld.
... and everyone is safe.

User avatar
Adam Raoof
Posts: 2720
Joined: Sat Oct 04, 2008 4:16 pm
Location: NW4 4UY
Contact:

Re: Suitability of acceleration?

Post by Adam Raoof » Fri May 14, 2021 11:05 pm

Use the Raoof Pairing System.

Rank the players in order of strength as usual for round 1. Pair seed 1 with 2, 3 with 4, 5 with 6 etc alternating colours.

In Round 2 pair those with 1 point the same way. Float up the top of the score groups with odd numbers and float down the bottom of the score group, not the median.

This way all players will get 5 good games against players of a similar strength if not rating, those who are doing well will play stronger players and those who are not doing so well will play weaker players.

This looks odd, but it is less artificial than acceleration and works with any size of swiss.
Adam Raoof IA, IO
Chess England Events - https://chessengland.com/
The Chess Circuit - https://chesscircuit.substack.com/
Don’t stop playing chess!

Roger Lancaster
Posts: 1910
Joined: Tue Mar 17, 2015 2:44 pm

Re: Suitability of acceleration?

Post by Roger Lancaster » Sat May 15, 2021 5:52 am

Paul McKeown wrote:
Fri May 14, 2021 10:25 pm
Roger Lancaster wrote:
Fri May 14, 2021 1:10 pm
While I take Alex's points, I think I'd disagree. In the absence of any draws, which tend to be few and far between in junior Swiss events where the competitors have a wide range of abilities, a normal pairing system would reduce a field of 32 or fewer to one outright winner with 5/5. In effect, the tournament becomes a knockout event because, once a junior loses a single game, all his/her remaining games become - to use Alex's phrase - "relatively meaningless". Even the occasional draw wouldn't affect this because even a winning score of 4.5/5 would be unattainable by anyone who had lost a game.
Roger, I have to say I find your argument a little odd. It won't matter what the pairing system is, anyone who loses a game in a five rounder is going to struggle to catch up with the leaders.
Paul, surely the arithmetic is quite simple. A typical accelerated pairing system, by bringing forward the games where the leading players meet, increases the chance that a score of 4 points will be sufficient to finish first or, more likely, joint first. As to the wider issue, whilst I respect the opinions of those who argue otherwise, I'm persuaded by watching what I consider the futility of first rounds which are simply rabbit-bashing.

Mike Gunn
Posts: 1022
Joined: Wed Apr 11, 2007 4:45 pm

Re: Suitability of acceleration?

Post by Mike Gunn » Sat May 15, 2021 10:54 am

I have run club tournaments with a similar number of players to the ones you are expecting. Pairing Q1 v Q2 and Q3 v Q4 in round 1 removes the worst mis-matches then just carry on as in a normal Swiss. I don't think anything else is justified (or works in practical terms) with such low numbers.

Post Reply