Board Orders

[Alex Holowczak]

I think a lot of leagues may have moved from tossing for colours to the away team having white on odd boards.

In the Birmingham League and Dudley League, the home team is white on odds. In the Wolverhampton League, the away team is white on odds. The only toss for colours are in end of season playoffs for promotion/relegation.

[Matt Mackenzie]

I think a lot of leagues may have moved from tossing for colours to the away team having white on odd boards. If you hear at short notice that your top board is not coming, and you are wrestling with the ethics, you may not fancy telling your team that they have all prepared the wrong colour openings.

And that (or the other way around) is and always been a vastly superior method to all normal rational people

Which explains why the certain type you find at all chess clubs (and, even more so, league administration) persisted with the toss because "its the traditional way of doing things" - many are the same people who defend(ed) adjudications to the last breath, of course

[John Hodgson]

And that (or the other way around) is and always been a vastly superior method to all normal rational people

Enlighten an irrational grasshopper then, why is it 'vastly superior'?

In a league match with even boards and no board-order tie break I don't really see the advantage. It may make opening preparation easier, but it's not obvious to me whether this is a good or bad thing.

[Brian Valentine]

John,

Usually playing on an even board in the Herts (tossing rules, odd number of boards) league, my captains have managed to gain us a significant number of extra white side games. I've had 2 whites and 7 blacks. An anecdote to support the idea that the coin is an inferior method.

[Matt Mackenzie]

And that (or the other way around) is and always been a vastly superior method to all normal rational people

Enlighten an irrational grasshopper then, why is it 'vastly superior'?

In a league match with even boards and no board-order tie break I don't really see the advantage. It may make opening preparation easier, but it's not obvious to me whether this is a good or bad thing.

Because playing on the same board for your team, you get a roughly equal number of whites and blacks (assuming teams play both home and away)?

In the "good old days" of coin tossing it was not unknown for people to get the same colour 8, 10, even 12 times in a row

[Mick Norris]

Manchester League matches have an odd number of boards (7 in the top 3 divisions, 5 in the other 2) and toss a coin for colour

Teams only play each other once (unless there are fewer teams, in which case they play home and away, currently in the 5 board divisions)

[Dragoljub Sudar]

In 'smaller' leagues like Nottinghamshire (compared to the SCCU for example) there are relatively few new players each year (my club's 1st team (5 boards) has had the same 4 players for about a decade), therefore if we had to play in order of grade or strength the same players would play each other year in year out and that becomes boring for the players concerned. So mixing up the board order can bring an element of variety for these players.

Allowing captains to select their board order how they like allows tactics (away team has white on odds) and it's exactly the same for both captains so the arguments about how it gives one team an advantage over another is not necessarily correct. When Keith Arkell played for Long Eaton in the 90s one or two teams chose to place their weakest player on board 1 but there was nothing to prevent Long Eaton playing Keith on a different board if they suspected such a tactic (we have to exchange team lists before the match so you don't know beforehand who'll you'll face).

Another advantage is that if a player has a grade much higher or lower than his real strength (as can happen if they've had a very good or very poor time last season), he can be placed on the 'correct' board rather than being forced to play too high up or too low down.

Also, you can change your team around if one player is having a poor season or another is having a very good one. I started my team in grade order last season but it wasn't working so I swapped boards 1 & 3 and boards 2 & 4 and my team went on to win 6 of our last 7 matches.

ps, well said Matt about adjudications. I'm pleased that I persuaded the Notts league to get rid of them.

[Alan Walton]

In the Manchester League the rule about board orders is that players within 20 points are allowed to swap (there is still nominated player for teams at the beginning of the year)

This was brought in this year after an incident which Dragoljub describe where a team played their weakest player against a GM, the League after discussions believe this wasn't in the spirit of the League and decided to avoid this happening again by bringing in the above rule

[MartinCarpenter]

That sounds sensible Anyone who's disadvantaged by a rule allowing 20 points leeway really is trying something rather extreme!

I'm not at all sure about fixing colours for each board in advance. It certainly doesn't produce balance over a season if you've got a bit of leeway in terms of board order. The opposite really - you can easily fix it so that certian players get certain colours etc. The Yorkshire league switched to away team white on odd (8 boards) recently, and it feels OK for the big matches but through a season I'm less convinced.

[E Michael White]

I've been meaning to reply to this for a while :-

Michael, I'm not sure what you mean by this. Do you mean my statement is incorrect in the limited sense that Brian demonstrated? Or just that a prediction is just a guess (however, statistically a large number of guesses would converge on average to the actual results)? The ECF changed grades recently because the grades were not predicting results, of course.

This related to your comment :-

(To predict the score of the match you work out the average grade for each team and apply this to work out a predicted score for the match. This works irrespective of the board order chosen by each team - an advantage gained on one board is always exactly balanced by an advantage lost on another.)

I guess you believe that the expected score between two opponents is the difference between the two player's grades divided by 100 plus .5. Why do you believe that ?

[David Williams]

I guess you believe that the expected score between two opponents is the difference between the two player's grades divided by 100 plus .5. Why do you believe that ?

I, for one, do broadly believe that. Only broadly - if you have an expected 10% score with a 40 point grade difference, it's hard to believe you still have a 10% chance with a 100 point difference. But, broadly speaking, if I score 70% over 30 games my grade will be 20 points higher than the average of my opponents' grades. And my expected score is 70% against someone graded 20 points less than me, and presumably 50% against someone the same grade as me.

[MartinCarpenter]

Broadly surely everyone does It doesn't of course follow that swapping board orders over won't make a difference. Even amongst players of broadly similar strength, the distributions of different player results vs different strength bands of opponents can be quite different. Some don't 'ever' lose to weak players, some retain a persistent 'tail' of losses etc.

With apologies for being markedly obsessive, I've got a Monte Carlo simulation of a league season by grade band here and I can see differences like 1/2 a match point and 4 game points over an 11 match season (of 8 boards/match), purely from how you arrange equally graded players in the team.

Not a huge effect but definitely not a totally trivial one either. Obviously markedly unrealistic to pretend that the results on the 8 boards in a match are independent, but what can you do?

[Alex Holowczak]

In cricket, there's a toss to decide who bats first. I'm sure that over the course of a season, a team might bat first on a disproportionate number of occasions.

In baseball, the away side always bats first; there's no toss.

Is this an analogous situation to the issue discussed here? Therefore, is baseball's way of deciding who bats first superior to the way that cricket does?

[Stewart Reuben]

Surely any event where there are two games or matches against the same opponent, it is best to have alternate colours, or alternate first batting for the two sports mentioned.
Have no doubt. For a 4 board match where the players have identical ratings for the top, second, third and fourth board ratings, but very different ratings, you can get totally different results depending on the board order. From 1.75-2.25 to 2.4-1.6, see page 14 of my book.
Consider two teams A and B which are evenly matched. A1 and B1 are rated 2800, A2 and B2 2550, A3 and B3 2300, A4 and
B4 2050. The expected scores with various pairings are shown below for a ten game match.

Pairings Result Pairings Result Pairings Result Pairings Result
A1-B1 5-5 A1-B2 8-2 A1-B3 9.5-0.5 A1-B4 10-0
A2-B2 5-5 A2-B3 8-2 A2-B4 9.5-0.5 A2-B1 2-8
A3-B3 5-5 A3-B4 8-2 A3-B2 2. -8 A3-B3 5-5
A4-B4 5-5 A4-B1 0-10 A4-B1 0 -10 A4-B2 .5-9.5
TOTAL 20-20 24-16 21 -19 17.5-22.5

Of course if you allow the captain to select his team and only then toss for colour, so that he can choose black on odd if he wins the toss, the results can again be skewed. Even more so, if the captain knows what colour the team will have on board one, he can select the team order accordingly.
The only correct way is to toss for colour after exchange of the team lists and the 'winner' of the toss gets white on the odd boards. The team must be played in approximate order of playing strength. Of course the ratings I chose above were hugely different; that was to illustrate the principal.