PDA

View Full Version : Accelerating 1 or 2 rounds in a 7 round Swiss



bergil
07-06-2005, 10:27 PM
Please can anyone tell me the pro's and con's of accelating rounds in Swiss tournaments and your opinion of it.

Garvinator
08-06-2005, 12:19 AM
Please can anyone tell me the pro's and con's of accelating rounds in Swiss tournaments and your opinion of it.
Hello Bergil,

Acceleration is ok as it lessens the effect of the first round and some time second round mismatches, known on here as 'junk rounds'.

Even though the first two rounds are ok, the mismatches return in round 3. To reduce the mismatch effect in round 3, i recommend a 1 point acceleration if my memory serves me correctly.

That is if you want to use acceleration. Do you want any further information or more detail? I think most tournaments that dont use the 1 division monster swiss style use divisions instead of acceleration.

Are you thinking of acceleration for the Fairfield Summer and Winter cups?

Alan Shore
08-06-2005, 01:03 AM
I believe you want this thread: http://www.chesschat.org/showthread.php?t=229

Now you can sit down, get comfortable, perhaps grab a beer, and read through all 646 posts.

Alternatively, you can take my word, that accelerated swiss events take all the fun out of an open chess tournament - with no warm-up rounds and little chance to take a big scalp: something I'm sure we all dream of doing when we enter.

In short - acc. swiss bad.

Kevin Bonham
08-06-2005, 03:16 AM
Firstly there is no reason at all to accelerate for just one round because if you do this you will get the same matchups in round 2 you would have had in round 1. Always accelerate for at least 2 rounds in a 7-round swiss and do not accelerate for more than 5. (Typically you will not need to go that far because all the players from the bottom half will fall off the top scoregroup much earlier.)

Pros:

* Avoids unequal matchups in round 1, which some players may find boring or demoralising.
* Finds a winner faster. Good where the number of players you have is very large for the number of rounds.
* Ensures winner has been thoroughly tested against strong opponents.

Cons:

* Many computer programs do not implement it properly, eg the Swiss Perfect method is defective and can result in huge mismatches.
* Tends to lead to a more tightly packed field at the end. Poor for sorting places in the midfield and ratings prizes.
* Reduces the chances for weak players to improve their game by playing much stronger players.
* When you come off the acceleration you tend to get some mismatches. However they are not as bad as round 1 in the normal Swiss, no matter what some here may say.
* Works very badly if there are a lot of unrated players or a lot of ratings are unreliable.

I personally prefer to avoid acceleration and will use it only where the first round would otherwise be a completely uncompetitive joke or where there will not be enough time to sort the winners without it.

Ian Rout
08-06-2005, 09:41 AM
I'm sure everybody know my opinion but I'll repeat it anyway.

If you want to have a series of games of no great significance in combination, then accelerated pairings are fine, you will get more closely-matched games. The flip side as mentioned is that lower ranked players miss their one game against a big name.

But if it's something that matters, like a championship or a weekend tournament that people are taking seriously, then I don't believe it's appropriate. Giving a player a weaker opponent than they are due simply because they would lose against their correct opponent amounts to a handicap system.

In any case what is the purpose of throwing everybody into a single large group, which implies that they are fit to be matched together, and then trying to keep them apart, implying that they are not. If the disparity across the field is so great that it is a waste of time people pairing the top and bottom halves then it is better to play in two groups. No other sport wants to match seasoned internationals with rank beginners in the same group.

bergil
08-06-2005, 01:11 PM
Thank you all for your views so far, I was interested as a few people have mentioned it to me for weekend tournaments.
I



Are you thinking of acceleration for the Fairfield Summer and Winter cups?

Just after info and opinion at the moment.

ursogr8
08-06-2005, 01:44 PM
Please can anyone tell me the pro's and con's of accelating rounds in Swiss tournaments and your opinion of it.

hi bergil

I am reading your post that seems to ask what is our opinion on a solution, but which does not seem to define the problem.
So, let me take a guess > you are concerned by personal evidence, or related anecdote, that a large standard SWISS has two main faults that are disliked by many players
1) the first two to three rounds are junk rounds where the average rating difference in the pairings is greater than 450 rating points
2) even in a nine round SWISS, the yo-yo effect, where players are successively paired against a much higher opponent followed by a much lower opponent, persists for the top quartile of players into the 7th and even the 8th round.
If my guess is correct, read on.
If I have the wrong problem, feel free to ask again.

Is acceleration a solution to these two problems?
I don't think Ian Rout sees acceleration as a solution to these two problems.
I don't think Belthasar regards these are problems. So, naturally does not support acceleration.
I think gg'' is remarking that a better solution is to create tourney divisions rather than to accelerate.
I think KB's general experience and choice is to avoid acceleration.

Locally, we don't accelerate.
We do frequently split the field into Divisions, and this mitigates greatly the two problems you may have been addressing.

regards
starter

Garvinator
09-10-2006, 10:12 PM
Since accelerated pairings have popped up again in the WTM thread, thought I would dig this thread out instead of us bombarding that thread with some dry stuff.

I am curious about Kevin's comments about acceleration and how it actually is supposed to work and sp's defects in this department.

I had thought you apply the acceleration for two or so rounds and then pair the field normally from there. Is this right Kevin?

Kevin Bonham
09-10-2006, 10:39 PM
I am curious about Kevin's comments about acceleration and how it actually is supposed to work and sp's defects in this department.

I had thought you apply the acceleration for two or so rounds and then pair the field normally from there. Is this right Kevin?

There is probably a larger and even older thread about this (and no, not the one linked to above :D ) somewhere.

What SP does is simply add points to the designated players for the purposes of pairing them for as many rounds as specified. Sometimes this leads to bottom-half players who have had an upset and a lucky draw playing on top board.

A correct accelerated system, however, involves pairing together top half players with perfect scores, and feeding bottom half players with perfect scores to top half players with less than perfect scores (but not too much less) until either no bottom half players are on perfect scores or the tournament is nearly finished.

Rather than decide in advance to accelerate for exactly two rounds, the controller should decide to accelerate for at least two rounds, taking the acceleration off once all the bottom half players have dropped points, or when there are only a couple of rounds left (by which stage any bottom half player still on 100% deserves to be playing on top board).

The correct system does a much better job of keeping mismatches off the very top boards, which is important if your aim is to ensure that the winner(s) are worthy.

Garvinator
10-10-2006, 12:24 AM
Assumes wins for all higher seeds, draws only for players within 100 points of each other (this is just a test/example).


Bd # Res White Player Name # Res Black Player Name
1 14 0 Andrew S Barker (0.0) 1 1 Stephen J Solomon (1.0)
2 2 1 Michael Cashman (1.0) 15 0 Shayne Hunter (0.0)
3 16 0 Michael Gearing (0.0) 3 1 Tony Weller (1.0)
4 4 1 Brian Thomas (1.0) 17 0 Gene Nakauchi (0.0)
5 18 0 John C Ager (0.0) 5 1 John Alkin (1.0)
6 6 1 Alexandra Jule (1.0) 19 0 Matthew Gilpin (0.0)
7 20 0 Brock Mcintyre (0.0) 7 1 Oleg Korenevski (1.0)
8 8 1 Mark C Stokes (1.0) 21 0 Sean S. Erickson (0.0)
9 22 0 Peter Booy (0.0) 9 1 Antolin Feria (1.0)
10 10 1 Keith Macleod (1.0) 23 0 Garvin Gray (0.0)
11 24 0 John Boni (0.0) 11 1 Gerard Akers (1.0)
12 12 1 Andrea Altoff (1.0) 25 0 Peter G Bender (0.0)
13 26 0 Yi Lui (0.0) 13 1 Craig A Stewart (1.0)
14 27 1 Norm Braybrooke (1.0) 40 0 Jim Rogers (0.0)
15 41 0 Alex Miles (0.0) 28 1 Michael D'Arcy (1.0)
16 29 1 Kevin Bourke (1.0) 42 0 Kenji Nakauchi (0.0)
17 44 0 Curtis Teed (0.0) 30 1 Finn Sorenson (1.0)
18 31 1 Sebastian Jule (1.0) 45 0 Rachael Solomon (0.0)
19 46 0 Jack Shanks (0.0) 32 1 Kieton Beilby (1.0)
20 33 1 Andrew Webster (1.0) 47 0 Peter Seeto (0.0)
21 48 0 Alexander Seeto (0.0) 34 1 Beng Goh (1.0)
22 36 1 Allan Menham (1.0) 49 0 Andrew Morrison (0.0)
23 50 0 Aidan Leask (0.0) 37 1 Jily Goh (1.0)
24 38 1 Matthew Lloyd (1.0) 51 0 Jean-Paul Guilbaud (0.0)
25 52 0 Tetias Goh (0.0) 39 1 Anthony Solomon (1.0)
26 35 Lachlan van den Bergh (0.5) BYE
43 1 Matthew L Hopkins (1.0) BYE

Garvinator
10-10-2006, 12:29 AM
Bd # Res White Player Name # Res Black Player Name
1 1 1 Stephen J Solomon (2.0) 8 0 Mark C Stokes (1.0)
2 7 0 Oleg Korenevski (1.0) 2 1 Michael Cashman (2.0)
3 3 1 Tony Weller (2.0) 10 0 Keith Macleod (1.0)
4 9 0 Antolin Feria (1.0) 4 1 Brian Thomas (2.0)
5 5 1 John Alkin (2.0) 12 0 Andrea Altoff (1.0)
6 11 0 Gerard Akers (1.0) 6 1 Alexandra Jule (2.0)
7 13 Craig A Stewart (1.5) 14 Andrew S Barker (0.5)
8 15 1 Shayne Hunter (1.0) 27 0 Norm Braybrooke (1.0)
9 28 0 Michael D'Arcy (1.0) 16 1 Michael Gearing (1.0)
10 17 1 Gene Nakauchi (1.0) 29 0 Kevin Bourke (1.0)
11 30 0 Finn Sorenson (1.0) 18 1 John C Ager (1.0)
12 19 1 Matthew Gilpin (1.0) 31 0 Sebastian Jule (1.0)
13 32 0 Kieton Beilby (1.0) 20 1 Brock Mcintyre (1.0)
14 21 1 Sean S. Erickson (1.0) 33 0 Andrew Webster (1.0)
15 34 0 Beng Goh (1.0) 22 1 Peter Booy (1.0)
16 23 1 Garvin Gray (1.0) 36 0 Allan Menham (1.0)
17 37 0 Jily Goh (1.0) 24 1 John Boni (1.0)
18 25 1 Peter G Bender (1.0) 38 0 Matthew Lloyd (1.0)
19 39 0 Anthony Solomon (1.0) 26 1 Yi Lui (1.0)
20 35 1 Lachlan van den Bergh(1.5) 43 0 Matthew L Hopkins (1.0)
21 40 1 Jim Rogers (1.0) 48 0 Alexander Seeto (0.0)
22 47 0 Peter Seeto (0.0) 41 1 Alex Miles (1.0)
23 42 1 Kenji Nakauchi (1.0) 50 0 Aidan Leask (0.0)
24 49 1 Andrew Morrison (1.0) 44 0 Curtis Teed (0.0)
25 45 0 Rachael Solomon (0.0) 52 1 Tetias Goh (1.0)
26 51 0 Jean-Paul Guilbaud (0.0) 46 1 Jack Shanks (1.0)

Garvinator
10-10-2006, 12:31 AM
Bd # Res White Player Name # Res Black Player Name
1 4 ___ Brian Thomas (2.0) 1 ___ Stephen J Solomon (2.0)
2 2 ___ Michael Cashman (2.0) 5 ___ John Alkin (2.0)
3 6 ___ Alexandra Jule (2.0) 3 ___ Tony Weller (2.0)
4 13 ___ Craig A Stewart (1.5) 35 ___ Lachlan van den Bergh (1.5)
5 27 ___ Norm Braybrooke (1.0) 7 ___ Oleg Korenevski (1.0)
6 8 ___ Mark C Stokes (1.0) 28 ___ Michael D'Arcy (1.0)
7 29 ___ Kevin Bourke (1.0) 9 ___ Antolin Feria (1.0)
8 10 ___ Keith Macleod (1.0) 30 ___ Finn Sorenson (1.0)
9 31 ___ Sebastian Jule (1.0) 11 ___ Gerard Akers (1.0)
10 12 ___ Andrea Altoff (1.0) 32 ___ Kieton Beilby (1.0)
11 33 ___ Andrew Webster (1.0) 15 ___ Shayne Hunter (1.0)
12 16 ___ Michael Gearing (1.0) 34 ___ Beng Goh (1.0)
13 36 ___ Allan Menham (1.0) 17 ___ Gene Nakauchi (1.0)
14 18 ___ John C Ager (1.0) 37 ___ Jily Goh (1.0)
15 38 ___ Matthew Lloyd (1.0) 19 ___ Matthew Gilpin (1.0)
16 20 ___ Brock Mcintyre (1.0) 39 ___ Anthony Solomon (1.0)
17 41 ___ Alex Miles (1.0) 21 ___ Sean S. Erickson (1.0)
18 22 ___ Peter Booy (1.0) 40 ___ Jim Rogers (1.0)
19 46 ___ Jack Shanks (1.0) 23 ___ Garvin Gray (1.0)
20 24 ___ John Boni (1.0) 42 ___ Kenji Nakauchi (1.0)
21 52 ___ Tetias Goh (1.0) 25 ___ Peter G Bender (1.0)
22 26 ___ Yi Lui (1.0) 49 ___ Andrew Morrison (1.0)
23 43 ___ Matthew L Hopkins (1.0) 14 ___ Andrew S Barker (0.5)
24 44 ___ Curtis Teed (0.0) 51 ___ Jean-Paul Guilbaud (0.0)
25 48 ___ Alexander Seeto (0.0) 45 ___ Rachael Solomon (0.0)
26 50 ___ Aidan Leask (0.0) 47 ___ Peter Seeto (0.0)

Garvinator
10-10-2006, 12:33 AM
Ok in round 3, is board four correct. Acceleration was applied to round 3

Kevin Bonham
10-10-2006, 10:12 PM
Ok in round 3, is board four correct. Acceleration was applied to round 3

The rules in Reuben's book (first edn) for round 3 are as follows:

Pair together all top half players having 2 points.

Done

Pair all bottom half players with 2 points against top half with either 1.5 or 1 point

Not applicable

Pair together all remaining top half players with 1.5 points

Only one to pair.

Pair bottom half players on 1.5 points with top half players on 1 point and, if necessary, 0.5 point

Here we have a problem because instead of doing this the bottom half player on 1.5 has been paired with a top-halfer on 1.5.

(Next comes Pair together all remaining players)

I am not sure if there's an established convention regarding what to do in this situation but it is definitely more in the spirit of the system to pair the bottom-half player on 1.5 with a top-halfer on 1, and the top-halfer on 1.5 with a top-halfer on 1 as well. Not only does this provide better weeding out of the top end of the field but it also eliminates a mismatch.

Therefore I think board 4 is probably not the ideal pairing. I'm reluctant to say it's "wrong" in case there's some accepted acceleration system that leads to it for some reason.