Minimum Number for Swiss Teams
I thought it would have been easy to find the answer to this, but google proved me wrong.
If I'm running a Swiss Teams event with 6 rounds (of say 8 boards), what's the minimum number of teams I need to make it worthwhile? 7 is pointless (might as well run multiple teams), and 8 seems only marginally less so. Even with 10 the principle of Swiss isn't really being seen.
Thoughts?
Comments
Strangely, I was just discussing this with my club committee this afternoon. We have an annual Swiss Teams but sometimes we struggle to get a reasonable number of teams.
There used to be a rule of thumb that is is best to have, for n rounds, the minimum number of teams should be 2 to the power of (n-1). So that means if you have six rounds, you want 2 to the power of (6-1) teams, ie 32 teams. I haven't heard anyone quote this for many years now and I think in practice you can have many fewer teams for the number of rounds without it being said to be "over-Swissed".
We would also have 6 rounds as you would, and I suggested that 10 teams is an absolute minimum and 11 wouldn't work because of having triples or Triellas. (I used to call them threesomes, but one day I was running the Young Bridge Challenge and used that word and all the school kids started giggling!).
I digress. 12 would be ok-ish and 13 just about tolerable if the, er, triples are all long triples. 14 and more are fine.
That's all just my opinion. Others might differ.
Barrie Partridge - CTD for Bridge Club Live
The purpose of Swiss is to be able to run events when there are too many teams to be able to have a complete multiple teams event, so I would run a multiple teams unless you have too many teams. That is likely to mean when you have 12 teams, or 14 teams or more, depending on how many boards you want to play.
I think Barrie's formula was for the maximum number of teams for the number of rounds, rather than the ideal number or minimum number. That would lead to the conclusion that for 6 rounds you ideally want 17-32 teams.
Thank you both - very helpful
I do think you will struggle to produce a fair event if the numbers are lower than the minimum Barrie's formula suggests i.e. 17 teams for six rounds. At the Kent Congress last year, because of a clash with the Premier League, we only had 15 teams (and 2 of them were cobbled together at the last minute). The event was advertised as 7x7-board rounds. We changed that to 6x8-board rounds. We were committed to the break after 3 rounds because of the provision of food so the triples were short, long, long, short. If teams were not allowed in the triple more than once then 12 of the 15 teams would have played in a triple. (I have heard that there may be an advantage in having 3 teams in the middle of the field in the triple but I don't think there is an easy way of making this happen in EBUScore.) So although we didn't allow teams from the first triple into the second we allowed teams back in in the second session and in fact one team was in the triple from Match 2 until the end of the event.
We also offered the players the option of a 'Danish' last round were it would be 1v2, 3v4, 5v6 etc. even where this meant rematches but they narrowly voted against this. In the event the last round was 1v3, 2v4, 5v7... so not too bad. However beforehand I did run some simulations one of which had 1v5 on the last round and another 3v8 which potentially might have upset some people. However the option of Danish having been offered and rejected would been our answer to such complaints.
Perhaps this is the best answer. We can work out what is theoretically the best format (an international mix of Australian format Swiss Pairs with a Danish last round?) but in the end the event is there for the players. Teams lower down the field like Swiss over multiple teams because they can still get some master points and play more boards against players at the same level. Teams at the top like Swiss because they can play more boards against their real opponents rather than the luck of whether they get 'swingy' boards to beat the weaker teams by more than the other top teams. So if there are not enough teams for the theoretically fair Swiss then let the players decide the precise format then at least it won't be your fault.
Because of triangles 17 and 19 are worse than 16.
For the event Jeremy may have in mind, the organisers and players have disliked playing multiple teams, even with small numbers of teams - although a proper barometer movement with scoring up after each match may be acceptable. What I have done in the past with < 16 teams is to play 5 rounds of Swiss and a Danish last round, with the provision that teams lying 1st/2nd after 5 round are guaranteed 1st or 2nd, and similarly for 3rd/4th.
"I think Barrie's formula was for the maximum number of teams for the number of rounds ...."
That makes sense. It was a long time ago that I last heard of it.
I agree that there tends to be disappointment if you bill an event as Swiss Teams and change it to Multiple Teams. Last year I entered a team for the Notts One Day Swiss Teams. Notts contacted me to say they hadn't enough entries for Swiss and would my team still come if it was Multiple. Though disappointed, my team still said "yes", but some other teams declined and the event was cancelled.
I find one significant reason for the popularity of Swiss Teams is that it is more sociable than Multiple, particularly in that you get to interact with team mates after every few boards rather than having no contact with them between the beginning and end of the session.
Barrie Partridge - CTD for Bridge Club Live
If the boards have already been dealt and you have an even number, you could play it barometer and still get to score up every round.
Manning, the EBU manual, states that the number of matches should not exceed one third of the number of competing teams
"play 5 rounds of Swiss and a Danish last round, with the provision that teams lying 1st/2nd after 5 round are guaranteed 1st or 2nd, and similarly for 3rd/4th."
Is this another way of saying you play 5 rounds of Swiss to qualify two finalists and two teams for the 3/4 play-off?
Do the rest of the teams also play Danish for every pair of places, or is it simply one more round of Swiss (including re-matches) to determine all the rest of the placings?
No. In the final round all matches are scored by VPs - so effectively the leading team has a carry forward to the play off. Prizes and ranking master points will not go below 4th, so 5-last are just playing for match-won master points (and NGS!)
I'm not convinced that this would be popular. One feature of a Swiss that people are very much used to is that teams can leapfrog those above them on the last round if those above take points off each other.
The county allowed me to repeat this with 11 teams one year and 13 teams the next.