Colour allocation change in WCH round robins


Back to forum

Garvin Gray    (2010-10-06)
Colour allocation change in WCH round robins

Currently in the WCH Stage 1 groups (new players) thread, there has been a side discussion occurring about allocation of colours in round robins on this site.

Since Thib has said that he is open to discussing it in a new thread, this is a new thread.

I have made mention of the berger pairing tables, which is the accepted format in otb round robins for how to pair round robins fairly.

The Berger pairing table for seven and eight player round robin is now provided:

1 1:8 2:7 3:6 4:5
2 8:5 6:4 7:3 1:2
3 2:8 3:1 4:7 5:6
4 8:6 7:5 1:4 2:3
5 3:8 4:2 5:1 6:7
6 8:7 1:6 2:5 3:4
7 4:8 5:3 6:2 7:1

Thibault de Vassal    (2010-10-06 08:39:16)
Colour allocation change in WCH round robins

Hi again Garvin, I must say I didn't even know the Berger pairing tables, very interesting to pair 6,8,10,12... players, but maybe not so interesting for 5,7,9,11... as there is a bye.

Anyway, I may try to code it for these cases!

On the original topic, the discussion should continue on the question: Must we reverse the colors in WCH round-robin groups so that Seed #2 play White against Seed #1, as the tiebreak (TER: Tournament Entry Rating) is an advantage for Seed #1 already.

As I explained in the other discussions (you can find some statistics in the first one), the whole idea of the chess championship is to find the best player, and rating is an important element in the process IMO. By the way I'm not sure if such an update would change the results significally.

Reverse or not reverse the colors in WCH groups, we need your opinion on this point!

Simone Calvello    (2010-10-06 14:02:01)
Colour allocation change in WCH round robins

I found this interesting link with some explanations. I hope this will help! :)

Thibault de Vassal    (2010-10-06 14:20:56)
Colour allocation change in WCH round robins

Just noticed that Berger pairing tables for 8 players really look like the way 7 players tournaments are made at FICGS... Seed 1 plays White against Seeds 2, 4, 6, 8 while Seed 2 plays White against Seeds 3, 5, 7, 8 and Seed 3 plays White against Seeds 1, 4, 6, 8. Not really different and not a solution to solve the disadvantage of Seed 2.

Garvin Gray    (2010-10-06 14:43:08)
Colour allocation change in WCH round robins

Thib: Is there a table to view for how the pairings are made on here? From looking at the pairings from my current WCH group, I was struggling to find out how the pairings are different from the 7/8 berger pairing table.

I am seed 1 and I have white against 2, black against 3, white against 4 and so on.

Seed 2 has black against seed 1, white against seed 3, black against seed 4 and so forth.

I think this is right, but without the tables provided for this site, I can not confirm this.

I am not sure whether seed 1 should be white or black against seed 2. This is because the TER rules mean that if seed 1 ties with seed 3, then seed 1 progresses regardless of which colour.

Which colour seed 1 is allocated probably means more depending where in the serpentine pairings the group 1.

If it is group 1, then seed 1 will be playing seed 22(overall)(seed 2 in Group 1) and then seed 23(overall)(Seed 3 in Group 1).
But if we are talking about Group 11, then it is a battle for seeds 11 and 12. Seed 3 in Group 11 wil be around seed 33(overall).

I am not fussed either way, but this would depend on alot on what tables are used for pairings on here.

Thibault de Vassal    (2010-10-06 15:29:07)
Colour allocation change in WCH round robins

No table, only a simple algorithm (with eg. $p1 = 0,1,2,3,4,5,6...) :

for ($p1=0;$p1<$players_count;$p1++)
for ($p2=$p1+1;$p2<$players_count;$p2++)
for ($round=1;$round<=$rounds;$round++)

if (NOT $reverse AND ($round+$p2-$p1)/2 is EVEN) OR (YES $reverse AND ($round+$p2-$p1)/2 is ODD)
{ $player[$p1] plays White against $player[$p2] }

else { $player[$p1] plays Black against $player[$p2] }


Games are created the order they appear in the tournament. (may be hard to read, the idea may look like obvious to you though)

Garvin Gray    (2010-10-08 06:29:09)
Colour allocation change in WCH round robins

How does a person read that algorithm. I have tried repeatedly to work it out and I can not.

Thibault de Vassal    (2010-10-08 15:11:13)
Colour allocation change in WCH round robins

Okay, not exactly an algorithm, this is mostly simplified PHP. The result for a 7 players with 1 round (single round robin) tournament is:

White : Black

1 : 2
3 : 1
1 : 4
5 : 1
1 : 6
7 : 1

2 : 3
4 : 2
2 : 5
6 : 2
2 : 7

3 : 4
5 : 3
3 : 6
7 : 3

4 : 5
6 : 4
4 : 7

5 : 6
7 : 5

6 : 7

So here is the table for 7 players single round robin.

Garvin Gray    (2010-10-08 15:26:38)
Colour allocation change in WCH round robins

Hello Daniel Parmet,

In the other thread WCH Stage 1 groups (new players), in your first post-

Further i've noticed that the color pairings tends to be very very bad for seed 2 & 3 but great for seed 1 & 4.

As seed 2 I get black against 1&4 as seed 3 I get black against 1 4 but as seed 4 I get white against 1&3. Yet these color disadvantages are not at all included as a tiebreak in any way.

From looking at the pairings posted above by Thibault about how colours are determined on this site, I believe your post contains a factual error which does not help your argument at all about the unfairness of pairings on here in the context of colours.

On this site-
Seed 2 is black against 1&4.
Seed 3 is WHITE, repeat WHITE, against 1 and 4.
Seed 4 is BLACK, repeat BLACK against 1&3.

Garvin Gray    (2010-10-08 15:29:24)
Colour allocation change in WCH round robins

Now that I have seen the full list of pairings used on this site, I can not see any difference between the berger pairing tables that I have listed and the pairings used here.

The order of games is different, but this is because Berger Pairing Tables are normally used in otb tournaments, whereas on here all games are played at the same time, so no need for separate round game allocations.

Warren Porter    (2010-10-08 20:01:55)
Colour allocation change in WCH round robins

I have had a suite of webpages to produce and support Berger round robin tables for several years now: