18 or 19 tables

[lamford]

The North London club sadly had to turn away six people last night and send them home. It is clearly drawing people from far and wide to cross swords with SB or to watch the antics of ChCh. At a recent committee meeting the club decided to limit the number of tables to 17 and yesterday it would have had 18.5 tables. The club owns 16 Bridgemates (but can score the 17th or even the 18th table on the computer or by hand) and duplicates, with a hopper, one set of 36 boards per week not all of which are used every time. Physically the room could accommodate 18 tables at a squeeze. We do not want to duplicate two sets of boards per week or have two sections unless it is the only solution. We can possibly hire another small room at our venue. We want to play 24 boards, or 22 if there is a sit out. We could start at 7.15 instead of the current 7.30 and possibly play 26 boards. My questions are these:

 

a) I understand that there is an EBU requirement that all players play 70% of the boards. Is this recommended or cast in stone in order to have the event rated?

b) Is there a movement for 18 tables with only one set of boards and does it comply with a) above? Board sharing is fine.

c) Is it possible to have a NS and EW rover? Or even a third rover, alternating NS and EW, so that the 36 pairs sit out once each. This seems to get 19.5 tables with 34 boards in play, one table sharing, of which everyone could play 24 boards over 13 rounds?

d) Will this be a recipe for disaster when someone as incompetent as me roves to the wrong table?

 

Thanks in advance!

[chrism]

(b) Yes. There are 18 table webs for 26 boards in play or 24 boards in play, 2-board rounds. Normally there are two sets of boards, but one set can be shared, though different table-pairs share in different rounds so it's a lot of chaotic board movement and scope for error.

 

A rover can be added to either movement but the bump pattern is very irregular. I have the rover movement for 26 boards but I don't think I have it for 24. Of course the 26 board movement can be truncated to 12 rounds if necessary.

 

Easier would be 2 9-table Mitchells sharing between sections with a rover in one section, 8 or 9 three-board rounds, though that doesn't meet your single section requirement.

[chrism]

(d) Yes :-)

[chrism]

(d) ... but the greatest risk is sharing with the wrong table.

 

If it isn't feasible to preduplicate a second set of boards, you might get the players to table-duplicate boards the first time they have been played. Of course, that is also a recipe for disaster if players aren't reliable at this distinctly error-prone process.

[lamford]

(b) Yes. There are 18 table webs for 26 boards in play or 24 boards in play, 2-board rounds. Normally there are two sets of boards, but one set can be shared, though different table-pairs share in different rounds so it's a lot of chaotic board movement and scope for error.

 

A rover can be added to either movement but the bump pattern is very irregular. I have the rover movement for 26 boards but I don't think I have it for 24. Of course the 26 board movement can be truncated to 12 rounds if necessary.

 

Easier would be 2 9-table Mitchells sharing between sections with a rover in one section, 8 or 9 three-board rounds, though that doesn't meet your single section requirement.

Many thanks. It does seem better to have two sets of boards, I agree and any movement you can send me would be most welcome!

[Pig Trader]

At Sheffield BC, we like 9x3 rounds and for 18 tables, we use a Mitchell movement where EW move up two tables each round and Boards are shared between each pair of tables. This is effectively the same as having two 9 table Mitchells, but it doesn't seem so much like two sections. We use similar movements for 14.5 tables to 20 tables. There's more details in the Appendix at

http://www.sheffieldbridgeclub.co.uk/TinyFileManager/resources/files/129/admin/sbc%20-%20td%20guide%20-%20movements.pdf

[Vampyr]

At Sheffield BC, we like 9x3 rounds and for 18 tables, we use a Mitchell movement where EW move up two tables each round and Boards are shared between each pair of tables. This is effectively the same as having two 9 table Mitchells, but it doesn't seem so much like two sections. We use similar movements for 14.5 tables to 20 tables. There's more details in the Appendix at

http://www.sheffieldbridgeclub.co.uk/TinyFileManager/resources/files/129/admin/sbc%20-%20td%20guide%20-%20movements.pdf

 

Why don't you separate the two sections?

[Pig Trader]

Because it's easier to share boards with adjacent tables rather than having loads of players running up and down the stairs every five minutes with boards. We have been using this movement since decades before we got a duplimating machine 14 years ago! It's also quite nice to not appear to be obviously split into two sections! :rolleyes:

[Vampyr]

Because it's easier to share boards with adjacent tables rather than having loads of players running up and down the stairs every five minutes with boards. We have been using this movement since decades before we got a duplimating machine 14 years ago! It's also quite nice to not appear to be obviously split into two sections! :rolleyes:

 

I can certainly understand the stairs bit, but if the game is divided into two sectiona it doesn't seem sensible to try not to make it "obvious". It will be obvious enough when each section has its own winner anyway.

[gordontd]

I can certainly understand the stairs bit, but if the game is divided into two sectiona it doesn't seem sensible to try not to make it "obvious". It will be obvious enough when each section has its own winner anyway.

If they have two sections sharing boards, they won't have different winners in each section - it'll be scored overall.

[Vampyr]

If they have two sections sharing boards, they won't have different winners in each section - it'll be scored overall.

 

This is incorrect unless the sections are seeded.

 

EDIT: No, maybe not. Can you produce a one-winner game via one or more "crossover" rounds, analogous to arrow-switching?

[pran]

This is incorrect unless the sections are seeded.

 

EDIT: No, maybe not. Can you produce a one-winner game via one or more "crossover" rounds, analogous to arrow-switching?

One-winner games in such cases are just poor (unbalanced) substitutes for a fair game.

[gordontd]

This is incorrect unless the sections are seeded.

 

EDIT: No, maybe not. Can you produce a one-winner game via one or more "crossover" rounds, analogous to arrow-switching?

You used to play at the Young Chelsea in the days when every session was composed of two sections!

[Pig Trader]

If they have two sections sharing boards, they won't have different winners in each section - it'll be scored overall.

 

Quite right. In fact, even the scoring program hasn't been told that there are effectively two sections. We don't generally get this number of tables nowadays as we have far more sessions per week, but in the days when we used this movement more frequently, I doubt that more than two or three players ever actually realised that the field was split in two. Our 20 table movement is similar but the sections don't correspond with odd and even numbered tables because of the effect of the share and relay element. Our 15, 16, 17 and 19 tables movements are essentially two sections meshed together so they truly are one section.

[Vampyr]

You used to play at the Young Chelsea in the days when every session was composed of two sections!

 

Yes and I still play in the IMP pairs there. With an arrow switch. I am not going to boycott games when I don't consider them a fair contest. I have to play somewhere.

 

I have a bit of nostalgia for the downstairs smoking section!

[weejonnie]

One movement that comes to mind would be an appendixed Mitchell - 13 tables in the loop and 5 as the appendix, sharing. (EW up one, boards down one, NS down 2) NS sit at tables 1-5 and EW at tables 14-18. NS play at tables 14-18 instead of 1-5 i.e. 7 - 18 - 16 - 14 -12. The problem is that there is a lot of movement which in a tightly packed room would cause problems. (You play 12 rounds only). I haven't checked it, but you could also possibly run it with 12 tables in the main loop, with a share and relay (make sure the share isn't one of the pairs of tables already sharing)

 

The advantage of this is that the shared tables are close together and fixed - however there is nothing to stop you creating more than one appendix to suit the actual physical dimensions of the room. (NS sit at one of the tables and EW at the sharing one.)

 

I know scorebridge allows you to create bespoke movements (and upload them to Bridgemates) - not sure about EBUscore/ ACBLscore etc.

[Pig Trader]

Back in the days at Sheffield when we had up to 26 tables, the Appendix Mitchell was used whenever we had 20.5 to 26 tables. The players would call it "The duplicating movement" because at most tables, after the first board had been played, the players duplicated another board with the same number and took it to the table 13 higher or lower. (The tables in the middle had to duplicate two boards.) That was before duplimating machines, of course!

 

The Appendix Mitchells were popular at Sheffield despite seeming out of favour elsewhere for reasons I never really understood. More recently, Web Mitchells burst back into fashion and these are not dissimilar and have the advantages of greater flexibility and no upper limit of tables.

[lamford]

Back in the days at Sheffield when we had up to 26 tables, the Appendix Mitchell was used whenever we had 20.5 to 26 tables. The players would call it "The duplicating movement" because at most tables, after the first board had been played, the players duplicated another board with the same number and took it to the table 13 higher or lower. (The tables in the middle had to duplicate two boards.) That was before duplimating machines, of course!

 

The Appendix Mitchells were popular at Sheffield despite seeming out of favour elsewhere for reasons I never really understood. More recently, Web Mitchells burst back into fashion and these are not dissimilar and have the advantages of greater flexibility and no upper limit of tables.

Is it necessary to duplicate a second set of boards, rather than just a few extra ones? I am no expert on movements, but it seems to me that for 20 tables, a movement should exist without sharing with no more than 44 boards.

[weejonnie]

Is it necessary to duplicate a second set of boards, rather than just a few extra ones? I am no expert on movements, but it seems to me that for 20 tables, a movement should exist without sharing with no more than 44 boards.

I don't understand that - you can play a 20 table Mitchell with 40 boards (skip after round 10, 2 rounds arrowswitched if you want a single winner). Of course you would have to play 14 rounds to qualify under the 70% rule.

[mink]

The problem cannot be solved without board duplication. However, if there is a machine to do this, web-mitchell movements are the best solution. At my club we use it with 15 or more tables. Each pair plays all deals 1-26, unless a sitout is necessary due to an odd number of pairs.

 

Too few bridgemates are no problem as all tables without bridgemates can use a sheet of paper to record the scores, which are hacked into the computer at the end of the session.

[Pig Trader]

The Appendix Mitchells also have all pairs playing all 26 boards.

 

If we ever had 21, 23 or 25 tables again, we'd still stick with the Appendix Mitchell as we only need two sets of duplimated boards. I think that's the only disadvantage of the Web Mitchells over Appendix Mitchells.

[lamford]

I don't understand that - you can play a 20 table Mitchell with 40 boards (skip after round 10, 2 rounds arrowswitched if you want a single winner). Of course you would have to play 14 rounds to qualify under the 70% rule.

I might be talking rubbish here, but I was thinking of having 32 boards pre-duplicated, plus another set of virgin 24 boards. Say there are 20 tables. Round 1 starts with 1-20 and boards are shared so you have everyone having played 2 boards from 1-20. 1-2, 1-2, 3-4, 3-4 etc. Tables 1-8, the first four to finish, now duplicate another set of boards 1-8, the ones they have after round 2, so there are 40 boards in play, two copies of 1-8 and one copy of 9-32. The 20 NS players play 24 of the 32 boards and do not need to board share thereafter. The EW players will also play 24 of the 32 boards. It seems to me

 

a) that you only need to bring one set of 32 boards, plus a spare set of boards

b) that you only need to duplicate 8 more boards for 20 tables after round 1

c) that you will not have board sharing after round 1

d) that the earlier to finish of each pair of the lower numbered tables can duplicate the boards. Except only SB will be able to duplicate the board he played ten minutes earlier from memory.

 

I seem to have done a spreadsheet for 20 tables with 40 boards of which there are only 32 different. You need 12 rounds for 70% of the boards to be played. For example, NS 1 plays boards 1-24 in order, NS 5 plays 5-28 in order, NS 9 plays 9-32 in order, NS 13 plays 13-32 and then 1-4 and NS 17 plays 17-32 and then 1-8. So boards 1-8 are played more times. EW 1 plays 1-2, 5-6, 9-10, 13-14, 17-18, 21-22, 25-26, 29-30, 3-4, 7-8, 11-12, 15-16. EW 11 plays 11-12, 15-16, 19-20, 23-24, 27-28, 31-32, 3-4, 7-8, 13-14, 17-18, 21-22, 25-26. I think the skip comes after 8 rounds.

 

Feel free to demolish the above!

 

It just occurred to me that the above can be improved significantly by having the duplimate machine make board 33 the same as board 1, board 34 the same as board 2 etc. Then there is no need for any duplication at the table, and no need for board sharing. You just bring 40 pre-duplicated boards, of which 8 are the same. Or you can have the hopper ready to double duplicate as many extra boards as you need as soon as the number of players are known.

[weejonnie]

Not sure how this works, obviously easier on a spreadsheet

 

The first round - you have 10 pairs of tables sharing

Is that 1 shares with 2, 3 with 4 etc or 1 shares with 17, 2 with 18 etc

 

1: 1-2

2: 1-2

3: 3-4

4: 3-4

....

19: 19-20

20: 19-20

 

 

And where do boards 21 - 32 feed in?

[lamford]

Not sure how this works, obviously easier on a spreadsheet

 

The first round - you have 10 pairs of tables sharing

Is that 1 shares with 2, 3 with 4 etc or 1 shares with 17, 2 with 18 etc

 

1: 1-2

2: 1-2

3: 3-4

4: 3-4

....

19: 19-20

20: 19-20

 

 

And where do boards 21 - 32 feed in?

If boards 1-8 are pre-duplicated twice for 20 tables, then one can have 1-2 on table 1 and also on table 17 in round 1, and then it doesn't require sharing. My original version had 1-20 in play in round 1, and then 21-32 feeding in in round 2.

 

However, there is a simpler solution still, and the attached spreadsheet has the following features:

 

a) If 32 boards are duplicated and 1-8 are doubly duplicated, we can have 20 tables without board sharing in rounds 1-10. For rounds 11 and 12 board sharing is needed, or the additional boards can be duplicated during the evening by the TD.

 

b) The skip is after 5 rounds but there is no need for a bystand or pivot.

 

c) One can add a 13th or 14th round very easily if time permits. Boards should be checked in rounds 11 and 12 as the board movement is slightly irregular.

 

And thanks to ChrisM and Barrie for some help!

 

https://www.dropbox.com/s/d5b4hcj09lpkow1/20%20table%20min%20boards.xlsx?dl=0