PDA

View Full Version : Tie Break Question

ElevatorEscapee
21-04-2006, 09:12 PM
G'day all.

Serious question here:

What is current preferred "Tie Break" method for a)Swiss and b)Round Robin/All-Play-All tournaments?

Swiss Perfect offers several "tie break" methods, some of which achieve different results (ie Player 1 may be considered to be better off with one tie-break method, yet Player 2 is better off with a different tie-break method).

Would arbirter experts please be so kind as to advise as to what are the currently FIDE preferred tie-break methodologies for:

a) tournaments paired via Swiss systems,

and b) tournaments paired via the Round Robin/All Play All method?

My thanks in advance.

EE :D

Garvinator
21-04-2006, 11:57 PM
Hello EE,

Straight from swissperfect.

1. Buchholz

This is the sum of opponents' scores. The idea is that the same score is more valuable if achieved against players with better performances in a given tournament.

Looks like an ideal tie-breaking method and has been used since the Swiss system was invented.

However it has some weaknesses which are addressed by other methods (see Median-Buchholz, Progress, Berger below).

2. Median-Buchholz

Same as above but after leaving out the highest and the lowest scoring opponents. Swiss Perfect allows configurable number of highest and lowest scoring opponents to be left out (0,1 or 2)

Its idea is to eliminate distortions in Buchholz values caused by taking into account games against run-away winners and bottom placed players.

3. Progress

Calculated by adding points from a progress table eg if your scores were: Win, Loss, Win Draw then your progressive scores are 1, 1, 2, 2.5 and your Progress tie-break value is 6.5

This is an attempt to put a higher value on scores which were achieved by scoring better in the initial rounds than by finishing from behind. It is common knowledge that the latter is usually much easier to achieve.

The problem is that the order of the Progress tie-breaks is known before the last round (last round scores will change the actual value but not the order within a point group). This may encourage some undesirable tournament "tactics" in the last round.

4. Berger

This is calculated by adding scores of the opponents who were beaten by a given player and half the scores of the opponents who she drew with.

This has been adopted from round-robin tournaments and is usually used as a secondary method.

5. Number of Wins

Calculated by adding a point for a win and nothing for a loss or a draw.

Intended to discourage making quick draws. Popular in 70's and early 80's (particularly in round-robins). These days hardly justified.

6. Opponents' Rating Sum

Sum of the opponents' ratings. Uses the ratings ie presumed pre-tournament strength of the opponents rather than their performance in a given tournament. Also has the same problem with the last round as 'Progress'.

7. Minor Scores

The difference of minor scores FOR and AGAINST. An example of usage of minor scores is Othello (Reversi) where a minor score like 50-14 means that the winner finished with 50 pieces on the board and the loser with 14. Another example is a chess team tournament where the winning team receives 2 points, the losing team 0 as the main score and the sum of board scores eg 4.5 to 1.5 in a 6 board match is the minor score. Yet another example is goal difference in football / soccer.

Using minor scores as a tie break means that a player with a higher difference of the the FOR and AGAINST minor scores is ranked above the one with a lower difference. If a difference is identical then the one with greater value of minor score FOR is preferred. For example 120-100 is better than 90-70.

8. Brightwell

This tie-break is used in Othello (Reversi) and combines Buchholz with Minor Scores. It is the value of the Minor Scores FOR (in case of Othello this is the number of pieces the player scored, disregarding the number of pieces the oponents had) plus a configurable coefficient multiplied by the sum of opponents' scores (Buchholz).

For swiss events- I personally favour progressive and it is the tie break that is used in acf events for deciding non title ties.

For Round robins- Berger and then number of wins.

ElevatorEscapee
22-04-2006, 03:13 AM
Thanks Garv. :)

Garvinator
22-04-2006, 01:22 PM
EE, which have you preferred so far and why?

ElevatorEscapee
22-04-2006, 08:43 PM
Thanks for the question Garv. :)

I am not the current Tournament Director of my club, however I may have to assume the role next year if no one else will, and may be faced with such decisions.

SwissPerfect offers Progresseive & Bucholz as the default tiebreak methods, however, these methods do not necessarily work quite so well with a "Round Robin" event.

At the moment, my club is running a couple of Round Robin events as qualifiers for the 2006 Club Championship, whereby the first 3 in each event get to qualify for the Championship.

However, this can lead to a couple of scenarios where two people finish on the same number of points and some form of separation may be required as to deciding which one gets to compete in the Club Championship.

I personally prefer the Sonneberg-Berger type tiebreak for these events. This is: one adds the total of the scores of the players who the player in question has defeated, and half the scores of the players they have drawn with...(although sometimes even this method may not be conclusive).

What concerns me is that SwissPerfect may possibly be using an old tie-break rule (which I have uncovered from an auld hardback edition of the "Official Blue Book and Encyclopeadia of Chess" by Kenneth Harkness), whereby one's opponents' scores are calculated on the basis of 1 for a win over the board, and half for a draw and half for a win by forfeit.

Personally, I don't think any player (say Player A) should be penalised just because his or her opponent fails to show (hence he or she should receive a win on forfeit).

Taking this further, I don't believe that a completely unrelated player who happened to beat Player A should also be penalised in a tie-break situation, just because one of Player A's opponents failed to show, and Player A received a win on forfeit.

In any case, if I had to judge, for an all play all/Round Robin, I would go for a Sonneberg-Berger system every time, rather than any other tie-break method (given the basic premise of sum of beaten opponents' scores, and half of drawn with opponents' scores, where wins by forfeit are not counted as being different to wins over the board.).

If players are still tied afer that (and time permits) then I would put forward the motion to employ a "tie-breaker" match, held under the same conditions (ie time limits) as the tournament.

Cheers,

~EE :)

Rincewind
22-04-2006, 08:52 PM
I agree. I can't see that Buchholz is useful for Round Robin tourny at all since everyone played the same field (except for each themselves) and likewise progressive is not a good factor either (in round robins). You should use Sonneborn-Berger type of system instead.

Garvinator
22-04-2006, 09:18 PM
I agree. I can't see that Buchholz is useful for Round Robin tourny at all since everyone played the same field (except for each themselves) and likewise progressive is not a good factor either (in round robins). You should use Sonneborn-Berger type of system instead.
i would probably use number of wins as the second tie break in an effort to avoid short draws (or phone draws);)

Rincewind
22-04-2006, 09:23 PM
i would probably use number of wins as the second tie break in an effort to avoid short draws (or phone draws);)

Here is a question...

Why do arbiters care about the length of the games?

Garvinator
22-04-2006, 10:22 PM
Here is a question...

Why do arbiters care about the length of the games?
because most people, arbiters included, want to see players actually playing chess, rather than just turning up, making a few moves and then shaking hands.

Rincewind
22-04-2006, 11:01 PM
because most people, arbiters included, want to see players actually playing chess, rather than just turning up, making a few moves and then shaking hands.

Seems perverse to me.

Garvinator
22-04-2006, 11:21 PM
Seems perverse to me.
what seems perverse? Also are you actually being serious in your suggestion that it is ok for players to be able to turn up and not really play?

Bill Gletsos
22-04-2006, 11:35 PM
I personally prefer the Sonneberg-Berger type tiebreak for these events. This is: one adds the total of the scores of the players who the player in question has defeated, and half the scores of the players they have drawn with...(although sometimes even this method may not be conclusive).

What concerns me is that SwissPerfect may possibly be using an old tie-break rule (which I have uncovered from an auld hardback edition of the "Official Blue Book and Encyclopeadia of Chess" by Kenneth Harkness), whereby one's opponents' scores are calculated on the basis of 1 for a win over the board, and half for a draw and half for a win by forfeit.Swiss perfect uses S_B by default for Round Robins.
However as you suspect it has issues with wins on forfeit.
As far as I can determine instead of as you suggest it is using half for a win bye forfeit, it does not count wins on forfeit in the S-B scores at all.

ElevatorEscapee
23-04-2006, 01:18 AM
Thanks Bill. :)

Rincewind
23-04-2006, 08:52 AM
what seems perverse? Also are you actually being serious in your suggestion that it is ok for players to be able to turn up and not really play?

Define "really". They turn up, they play, the result is a draw. Why should anyone care if it took 15 moves or 115?

Denis_Jessop
23-04-2006, 11:36 AM
The obvious thing to do when considering this topic is to see, as a starting point what the FIDE Handbook says which is as follows:

Actual Handbook
Handbook | C. General Rules and Recommendations for Tournaments | 06. FIDE Tournament Rules | Annex to the FIDE Tournament Regulations

Annex to the FIDE Tournament Regulations

Approved by the 1998 General Assembly.
1.

Handling of unplayed games

To avoid improper influence of unplayed games on the ranking these games shall be counted as follows :

Independently of the result of an unplayed game (win by Bye, win or loss by forfeit, no game because the player has withdrawn or was absent for some round(s)) for reasons of tiebreak, the result shall be counted as a Draw against the player himself.

This will have no influence on the Sum of Progressive Score or Koya System. In these Systems only the result counts.
2.

List of commonly used Tiebreak Rules

In all Systems the players shall be ranked in descending order of the respective Tiebreak System. Within the following list there is no intention expressing an order of priority.
2.1

Tiebreak Rules using the Player's own results
2.1.1

Sum of Progressive Scores

after each round a player has a certain tournament score. These scores are added to the total Sum of Progressive Scores.
2.1.1.1

Sum of Progressive Score Cuts

is the Sum of Progressive Scores reduced by the tournament score of one or more rounds, starting with the first round.
2.1.2

Matchpoints in Team Competitions

2 points for a won match (a team has scores more points than half the number of boards a team should have)

1 point for a drawn match (a team has scored the same number as the number of boards a team should have)

0 points for a lost match (a team has scored less points than half the number of boards a team should have).
2.1.3

The Koya System for Round Robin Tournaments

This is the number of points achieved against all opponents who have achieved 50 % or more.
2.1.3.1

The Koya System extended

The Koya System may be extended step by step to include score groups with less than 50 %
2.1.4

Direct Encounter

If all tied players have met each other, the sum of points from these encounters will decide.
2.1.5

Number of games won
2.2

Tiebreak Rules using the Results of Opponents
2.2.1

The Buchholz System
2.2.1.1

The Buchholz Score is the sum of the score of each of the opponents of a player
2.2.1.2

The Median Buchholz 1

is the Buchholz Score reduced by the highest and the lowest score of the opponents.
2.2.1.3

The Median Buchholz 2

is the Buchholz Score reduced by the two highest and the two lowest scores of the opponents
2.2.1.4

The Buchholz Cut 1

is the Buchholz Score reduced by the lowest score of the opponents
2.2.1.5

The Buchholz Cut 2

is the Buchholz Score reduced by the two lowest scores of the opponents
2.2.1.6

The Sum of Buchholz is the sum of the Buchholz Scores of the opponents
2.2.2

The Sonneborn-Berger System
2.2.2.1

Sonneborn-Berger for Individual Tournaments

is the sum of the opponents a player has defeated and half the scores of the players he has drawn with.
2.2.2.2

Sonneborn-Berger for Team Tournaments 1

is the sum of the scores of the opponents teams, each multiplied by the scores achieved against this opponent team
2.3

Tiebreak Rules using Ratings
2.3.1

The Average Rating of Opponents

is the sum of the ratings of the opponents of a player divided by the number of rounds.
2.3.1.1

The average Rating Cuts

is the Average Rating of Opponents reduced by one or more of the ratings of the opponents, started from the lowest rated opponent.
2.3.2

Tournament Performance Ratings (including the 350 points rule)
3.

Application of Tiebreak System to different Tournament systems

The choice of the Tiebreak System to be used in a tournament shall be decided in advance taking into account the type of the tournament (Swiss, Round Robin, Teams, etc.) and the special structure of players to be expected in the tournament. For instance the applicaton of Tiebreak rules using the ratings of players is dubious in tournaments where the ratings are not available or not consistent or not necessarely correct (e.g. Junior/Senior tournaments)

Only one type of the three categories described above should be used for a given event.

For example : a) Sum of Progressive Score + b) Buchholz would be incorrect.

For different types of tournaments the Tiebreak Rules are recommended as listed below :

Individual Round Robin Tournaments:

* direct encounter
* Koya System
* Sonneborn-Berger
* Number of won games

Team Round Robin System:

* Games points
* Match points
* Direct encounter
* Sonneborn-Berger

Individual Swiss Systems (all players have consistent ratings):

* Average rating of opponents
* Tournament Performance Rating

Individual Swiss Systems (most players are mostly rated, ratings are not consistent):

* Direct encounter
* Sum of progressive scores
* Buchholz
* Sonneborn-Berger
* Won games

Individual Swiss Systems (most players are not rated):

* Direct encounter
* Buchholz
* Sonneborn-Berger
* Won games

Team Swiss Systems :

* Games points
* Match points
* Direct encounter
* Buchholz
* Sonneborn-Berger

Then you have to choose which one you like best provided you tell everyone beforehand. Note the cheeky reference to Senior/Junior tournaments - you can't put much past these guys :) .

DJ