Recent Posts

81

Requests / Re: Round Robin - 8 Player Doubles

« Last post by tkam on January 22, 2024, 11:17:40 AM »

Perfect, thanks again, Ian.

If you could assist with one more query, I'm seeking a variant that mirrors the previous structure but includes a round where players on the same court are sequentially ranked.

To clarify, I'm looking for a round where the player ranks are on court as follows:

Court 1: 12, 11, 10, 9

Court 2: 8, 7, 6, 5

Court 3: 4, 3, 2, 1

The aim is to ensure that there's at least one round where two matches are played where all players are of comparable skill levels. To clarify, no requirement for 12 to partner with 11 vs 10 and 9, simply that the round consists of those sequential ranks, the game order can be any combo within the four.

I'm hoping that the the schedule as a whole can still give a decently low balance score, though i suspect 112 isn't possible, but hopefully something close?

Much appreciated!

82

Requests / badminton tournamount

« Last post by superwomble16 on January 21, 2024, 03:34:10 PM »

hi i want a rotating pairs badminton tournamount using 20 players . it will be played over 10 rounds using 4 courts per round 
and therefore 4 players will sit out per round . The individual player will sit out all exactly twice and hopefully spread out over
the 10 rounds .
i have found one such tournamount using plan2play but the variety of partners is limited .
i would like a player to only play once with any other player and not play against the same pair twice .
A variety of matches with each player playing with the most players as possible creating a good mixup would be ideal .
i have just come across this site and it looks perfect for mr needs,hopefully. 
any help would be much appreciated . 

83

Requests / Re: Round Robin - 8 Player Doubles

« Last post by Ian Wakeling on January 20, 2024, 09:01:59 AM »

I think I am seeing that the partner/opponent schedule that you start with matters here.  I can improve a little on your result, as the schedule below has a score of 112.

 8  6 10 12
 8 10  6 12
11  3  9  5
11  9  3  5
 7  2  4  1
 7  1  2  4
 6  2  1  5
 6  1  2  5
 4 10  9  3
 4  9 10  3
 7 12 11  8
 7 11  8 12
 2 11 12  3
 2 12 11  3
 8  1 10  9
 8  9  1 10
 7  4  6  5
 7  5  4  6
11  4  5 10
11  5  4 10
 6 12  9  1
 6  9 12  1
 7  8  2  3
 7  3  8  2
 4  8  3  1
 4  3  8  1
 2  5  9 12
 2  9  5 12
 7 10  6 11
 7  6 11 10

84

Requests / Re: Round Robin - 8 Player Doubles

« Last post by tkam on January 19, 2024, 04:09:50 AM »

Thanks again Ian.

Sorry to be pedantic, would it also be possible to get a skill balanced version of the 1st schedule you posted? It has a slightly more symmetric distribution than the latest one.

I had a crack at it myself and the lowest I could get was 116, grateful if you could check for anything better?

85

Requests / Re: Round Robin - 8 Player Doubles

« Last post by Ian Wakeling on January 18, 2024, 12:30:25 PM »

I believe the schedule below has the same partner/opponent properties as before but has a skill difference score of 104.

 6  8 12 10
 6 12  8 10
 9  3 11  1
 9 11  3  1
 5  2  4  7
 5  7  2  4
 8  2  7  1
 8  1  7  2
 4 12 11  3
 4 11 12  3
 5 10  9  6
 5  9  6 10
 2  9 10  3
 2 10  9  3
 6 12 11  7
 6 11  7 12
 5  1  4  8
 5  4  8  1
 9  1  4 12
 9  4  1 12
 8  7 11 10
 8 11 10  7
 5  3  6  2
 5  6  2  3
 4  3  6  7
 4  6  3  7
 2 11  1 10
 2  1 11 10
 5  8  9 12
 5 12  8  9

I am doing the same thing that you described, just swapping around the player numbers, but the starting point was a similar schedule to the one in my previous post (generated by the same algorithm).

86

Requests / Re: Round Robin - 8 Player Doubles

« Last post by tkam on January 17, 2024, 06:57:58 PM »

Ian, thanks a lot, that’s fantastic!

Could we tweak the schedule to ensure the games are closely matched in terms of skill? Ideally, we want each team to have a similar aggregate skill level, so that player 12, being the strongest through to player 1, being the least skilled, are distributed in a way that makes each game competitive and balanced.

I've approached this by assigning each player a numerical ranking, with the top player receiving 12 points and the lowest-ranked player receiving 1 point. I then added up the points for each team in a game and determined the absolute difference in points between the teams. This value indicates the level of skill balance for each game; a lower total signifies a more evenly matched setup. My best attempt is the below which maintains the structure you provided but with player numbers swapped around. This leads to an overall total of 132. Is it possible to arrange the players to achieve an even more balanced total?

6  1  2  5
6  2  1  5
7  4  8  3
7  8  4  3
9  11  10  12
9  12  11  10
1  11  12  3
1  12  11  3
10  2  8  4
10  8  2  4
9  5  7  6
9  7  6  5
11  7  5  4
11  5  7  4
6  12  2  8
6  8  12  2
9  10  1  3
9  3  10  1
7  10  3  2
7  3  10  2
1  5  8  12
1  8  5  12
9  6  11  4
9  4  6  11
10  6  4  12
10  4  6  12
11  3  8  5
11  8  3  5
9  2  1  7
9  1  7  2

87

Requests / Re: Round Robin - 8 Player Doubles

« Last post by Ian Wakeling on January 17, 2024, 12:34:57 PM »

Hi Tej,

Here is my schedule which has 11 pairs who never partner, and opposition pairs that vary from zero to 4.


 2  3  8 10
 2  8  3 10
 5  7 12  6
 5 12  7  6
11  4  1  9
11  9  4  1
 3  4  9  6
 3  9  4  6
 1  8 12  7
 1 12  8  7
11 10  5  2
11  5  2 10
 4  5 10  7
 4 10  5  7
 2  9  8 12
 2 12  9  8
11  1  3  6
11  6  1  3
 5  1  6  8
 5  6  1  8
 3 10 12  9
 3 12 10  9
11  2  4  7
11  7  2  4
 1  2  7  9
 1  7  2  9
 4  6 12 10
 4 12  6 10
11  8  3  5
11  3  5  8


My version of your two symmetric matrices are immediately below (partners and opponents counts), followed by the sum of the two matrices.  I have put zeroes down the main diagonals.

 0  1  1  1  1  0  1  2  1  0  1  1
 1  0  1  1  1  0  0  1  2  1  1  1
 1  1  0  1  1  1  0  0  1  2  1  1
 1  1  1  0  1  2  1  0  0  1  1  1
 1  1  1  1  0  1  2  1  0  0  1  1
 0  0  1  2  1  0  1  1  1  1  1  1
 1  0  0  1  2  1  0  1  1  1  1  1
 2  1  0  0  1  1  1  0  1  1  1  1
 1  2  1  0  0  1  1  1  0  1  1  1
 0  1  2  1  0  1  1  1  1  0  1  1
 1  1  1  1  1  1  1  1  1  1  0  0
 1  1  1  1  1  1  1  1  1  1  0  0

 0  1  1  1  1  4  3  2  3  0  3  1
 1  0  1  1  1  0  4  3  2  3  3  1
 1  1  0  1  1  3  0  4  3  2  3  1
 1  1  1  0  1  2  3  0  4  3  3  1
 1  1  1  1  0  3  2  3  0  4  3  1
 4  0  3  2  3  0  1  1  1  1  1  3
 3  4  0  3  2  1  0  1  1  1  1  3
 2  3  4  0  3  1  1  0  1  1  1  3
 3  2  3  4  0  1  1  1  0  1  1  3
 0  3  2  3  4  1  1  1  1  0  1  3
 3  3  3  3  3  1  1  1  1  1  0  0
 1  1  1  1  1  3  3  3  3  3  0  0

 0  2  2  2  2  4  4  4  4  0  4  2
 2  0  2  2  2  0  4  4  4  4  4  2
 2  2  0  2  2  4  0  4  4  4  4  2
 2  2  2  0  2  4  4  0  4  4  4  2
 2  2  2  2  0  4  4  4  0  4  4  2
 4  0  4  4  4  0  2  2  2  2  2  4
 4  4  0  4  4  2  0  2  2  2  2  4
 4  4  4  0  4  2  2  0  2  2  2  4
 4  4  4  4  0  2  2  2  0  2  2  4
 0  4  4  4  4  2  2  2  2  0  2  4
 4  4  4  4  4  2  2  2  2  2  0  0
 2  2  2  2  2  4  4  4  4  4  0  0

Note that the 12 players fall into two distinct groups with different patterns of partners and opponents (Group A = 1-5 & 12, Group B = 6-11).  Finally note that there are 6 pairs of players who never meet either as partners or as opponents.  Hope that makes some sense.

88

Requests / Re: Round Robin - 8 Player Doubles

« Last post by tkam on January 17, 2024, 04:52:16 AM »

Thanks for having a look Ian.

Actually your 10-11 pair gap and 0 to 4 against is much better than what i've discovered already. Would you mind sharing that?

Otherwise, what is the most evenly distributed possible schedule if the partner with everyone requirement is relaxed?

Currently, the best i've been able to get is the below, where partnering with is max 2, min 0 and playing against is more extreme with max 10, min 0 (it seems you have already got one a lot better than this).

Ultimately i'm happy with any schedule which ends up smoothing out the extremes and results in the same ratios for everyone for both WITH and AGAINST, albeit with instances of zero partnering occurrences...

89

Requests / Re: Round Robin - 8 Player Doubles

« Last post by Ian Wakeling on January 16, 2024, 05:49:38 PM »

I believe the partnering-with-everyone-once-with-the-exception-of-one-person part is not possible - at least I can't find a schedule like this - as I end up with 10 or 11 pairs who don't partner.   Then the opponent balance is terrible, it varies from zero times per pair to 4 times per pair.  You would be far better advised to pick a different format.

90

Requests / Re: Tournament 5 players , 2 games

« Last post by Ian Wakeling on January 16, 2024, 05:42:08 PM »

I think you can use the standard round robin twice and just swap around the games associated with each column.  For example:

Rounds 1-5
 Fifa  Darts
(1 4) (2 3)
(3 5) (1 2)
(2 4) (1 5)
(1 3) (4 5)
(2 5) (3 4)

Rounds 6-10
Darts  Fifa
(1 4) (2 3)
(3 5) (1 2)
(2 4) (1 5)
(1 3) (4 5)
(2 5) (3 4)