[Hanoi5]

Is there somewhere I can find what's the best tournament to have 1 winner (as opposed to one E-W winner and 1 N-S winner) around?

I think Howells are better than mitchells, of course (even scrambled mitchells) but what about from 7 and a half tables on? When will it be better or neccessary to have two sections?

What about issues like balancing the field? (I suppose it's better to have the same amount of good/bad players in distinct sections/axles). How do you deal with those?

[TylerE]

How many players and sessions?

[Hanoi5]

2, 4 or 6 sessions. From 8 to 16 tables. I'm asking in general not for a specific case. I'd just like to have a list available for directors.

[Siegmund]

One obvious 2-session solution with 8 to 14 tables is to run one session as a Mitchell, and the second session as two Howells, all the original NSs in one section and all the original EWs in the other. Everybody plays everybody else once during the day. For some reason it's never used in North America except by a few directors who have internet friends in Europe.

The only obstacle to using a simple Howell movement of e.g. 25 rounds of 2 boards each for a 13-table field, is the security-driven need to keep people from talking about hands their friends haven't played yet at the break -- with the help of some pre-duplicated extra boards there are ways of achieving that but I'd have to look them up. (Playing the same board at every table at the same time, internet barometer style, isn't practical for any club I've ever been in.)

A standard reference work, providing solutions to many of the situations you're asking about, is Hallen, Hanner, and Jannersten 1994, Movements: a Fair Approach. (Edited to add info.) Plan to search for a while before you find a used copy at a fair price.

[NickRW]

Siegmund, on Sep 24 2010, 06:20 PM, said:

One obvious 2-session solution with 8 to 14 tables is to run one session as a Mitchell, and the second session as two Howells, all the original NSs in one section and all the original EWs in the other...

One obvious solution for 14.5 or 15.5 tables - assuming there isn't time to complete a full movements is to run the Mitchell as a 1.5 table appendix Mitchell (or Bowman or Ewing - same thing by different names I think). Also works for 15 and 16 tables, but really needs two sets of duplicated boards due to the fact that the last table has to share with different tables around the room - which may not be practical. You won't quite get everyone playing everyone - but nearly.

Two sets of duplicated boards - assuming we are talking 2 board rounds - are ideal even for the case of 14.5 and 15.5 tables as you have sharing between tables 1 and N-1 anyway - which isn't ideal for 2 board rounds.

Nick

[jonottawa]

I find that Individuals are the best tournaments to provide 1 winner.

[gordontd]

For 3n+1 pairs (where n is odd) or for 3n pairs where n is even, you can run a Mitchell + Howell combination over three sessions with movement of lines between sessions. Do it twice and you have six sessions.

For 5n+1 pairs (where n is odd) or for 5n pairs where n is even, you can do the same thing over five sessions with movement of lines between sessions.

If you have an odd number of tables you can run a Mitchell (missing out the first round) with an interwoven Howell.

Or you might run a Barometer Howell.

More detail when I have a proper keyboard!

[croquetfan]

for some reason, there is always an outcry when people think there is the slightest bit of imbalance. they seem to think that they were certain to get a top against pair X that they missed.
Once you get more pairs than you can fit into Howells, the fairest is probably swiss, but needs extra boards.
That book mentioned is great but some of the complex movements are risky unless players are very careful to follow movement cards. (and also you get the grumblers: why are we dong this rigmarole, our old director always just used arrowswitch")

[gordontd]

Amusing to see my old post replied to after more than 8 years!