[Fluffy]

Swiss qualifiers have these things, final standings after 12 rounds:

1. 230
2. 224
3. 206
4. 201
5. 199
6. 198
7. 196
8. 196
9. 195 The Latin
10. 195
11. 195
12. 195
13. 194
14. 191

The tie breaker is IMP coeficent, given that we were on top of the 4 teams pack it is fair to think that if we had scored 1 more VP we would be between the 8 qualifiers.

Given that we won last match by 26 IMPs (21-9), and 1 more IMP would move us to the 22-8 scale, you can imagine, we spent 1 hour looking at every score, adding and counting in the search of a lost IMP, wich didn't happen.


I think we had the most IMPs by a far amount for teams in our range, winning 2 of the 12 matches by 25-0, the main problem was, that we didn't share them evenly .

[Gerben42]

I am sorry to hear that you missed qualification by such a small margin, but it's still a good result in such a field.

Swiss qualification is not a fair method, but at least it's unfair to all in the same way so next time it will be you who will be lucky!

[Walddk]

Gerben42, on Feb 12 2009, 06:55 PM, said:

Swiss qualification is not a fair method

Why is Swiss not fair when there is not enough time to play against all other 45 teams?

Roland

[jdonn]

Walddk, on Feb 12 2009, 12:08 PM, said:

Gerben42, on Feb 12 2009, 06:55 PM, said:

Swiss qualification is not a fair method

Why is Swiss not fair when there is not enough time to play against all other 45 teams?

Roland

It might be the fairest method available. But it's not fair precisely because there is not time to play against all other 45 teams.

[Fluffy]

IMO Last match's importance outweights the first blatantly. A bad break on first match its not the same as the last one. (Specially when you play only 4 boards a match)

7-8 made 16.33 VPs a match, 9-12 did 16.25, its very very low, and show how tight the skill level was around.

[Mbodell]

jdonn, on Feb 12 2009, 09:24 AM, said:

Walddk, on Feb 12 2009, 12:08 PM, said:

Gerben42, on Feb 12 2009, 06:55 PM, said:

Swiss qualification is not a fair method

Why is Swiss not fair when there is not enough time to play against all other 45 teams?

Roland

It might be the fairest method available. But it's not fair precisely because there is not time to play against all other 45 teams.

Well, you could always just play as many matches as you can, and then use math to calculate the expected results in the unplayed matches based on the results in the played matches. Then who ever has the highest scores (adding up the played match scores with the unplayed simulated scores) qualifies.

If you play a significant number of the field, I think that would be fairer than the Swiss method, as it comes closest to properly adjusts for strength of schedule.

[xcurt]

jdonn, on Feb 12 2009, 12:24 PM, said:

Walddk, on Feb 12 2009, 12:08 PM, said:

Gerben42, on Feb 12 2009, 06:55 PM, said:

Swiss qualification is not a fair method

Why is Swiss not fair when there is not enough time to play against all other 45 teams?

Roland

It might be the fairest method available. But it's not fair precisely because there is not time to play against all other 45 teams.

Swiss is designed to optimally rank teams at the top of the table. In the extreme, it is designed to find the winner. By symmetry, it also does a good job of ranking the teams at the bottom of the table.

Swiss is worst at ranking teams in the middle of the table. If your qualifying cut-off is somewhere in between 1/3 and 1/2 of the teams, you're going to be better off dividing the field into groups and running a complete RR within each group.

Curt

[jdonn]

Can you elaborate please on what you mean by "designed to optimally rank"? Of course something can be designed to optimally rank teams but do a bad job of it (unless you meant "optimally designed to rank").

[HeavyDluxe]

Congrats on the result, G... I know it's not what you were hoping for, but it's still impressive for us mere mortals.

[Fluffy]

Hehe, sure I hoped for more, last year we finished in second place (both in the qualifying and the final). But I am getting used of being the first out.

In 2005´s transnational finished 9th on the swiss with 300 teams, in Bejing 5th of the round robin of our group with 4 qualifying, also we ranked 69th in the pairs, wih 68 qualifying lol.

This is hurting less every time

[barmar]

There was a thread a year or two ago from Hrothgar that examined the mathematics of Swiss ranking.

[Gerben42]

The problem with Swiss is that later matches are more important than early matches. This is based on the assumption that you will later meet teams that are closer to your strength. This is incorrect because in Bridge, unlike chess, you have some element of luck, like if you bid a 50% contract that was not bid in the other room, it's basically a coin flip.

Since later matches are more important than early matches, those teams that get their good boards later will end up quit a lot in front of those who get their good boards early.

I already noticed Gonzalo's bad luck - for one he was on a team that deserved to be in the QF in Estoril. The strength-adjusted results for that tourney were given here:

http://www.geocities...elo/estoril.txt

(our hero is on team Lantaron, ending 9th in the Swiss but 7th in the adjusted rankings, significantly ahead of Spector and Sofia Pessoa)

[Fluffy]

Thx Gerben, its nice to know. I can't explain how I felt at that time in 2005, it was my very first tournament at "top" level, and last match it was all my fault (or so I felt).

Worst of all, it was on Sunday, after traveling 600 km back home that night I could find no friend to get drunk with .

Last year we played badly at the start, then improved, and got the most amount of luck in last match, first we incredibly avoided the teams a the top because or a miracoulous combination. And then we made almost twice the difference needed for 25-0 opponets missbidding and missplaying 5 slams (At both tables)


This year we made homework, and played against top20 teams 10/12 matches I think, this would had let us get a decent draw for last match in case we were in good posiion, but sadly we losed the underlast match 23-7 lol.

[Fluffy]

We gopt some of our luclk back on the consolation swiss.

With matches of 7 boards (8 rounds), where winning by 1 IMP is 18-12 already, and you can score 30-0 total, it is very important that you do not lose matches and avoid disasters.

Brenner couldn't play, and with 5 minutes to start we asked the organization to find us a substitute, they got us a junior Caddy from the staff, I offered to play with her because I am used of playing with pickup partners. She introduced herself as Ruri Ohte and we arranged a system in 5 minutes.

We kept things as simple as possible, no thin games, no competitive beyond level, and from the total of 56 boards, she played all of them almost perfect, she made every contract she played, nothing wrong on defence, nothing spectacular either, but that is not what we aimed at. You could be named Zia, Meckstroth or Bochi, and you would not win more IMPs than she did .

After 7 rounds, we were first with 186 VPs, second place where the Dutch with 156, so for the last round all we had to avoid 30-0 (less than 28 IMPs), very exciting because they really pushed in, board 4 they played a "decent" vulnerable slam that went off (other table played part score!) and it was over.


No matter how objective you are, if you win with a tournament, you are gonna like the scoring method . But I think there is a point for making shorter matches in swiss tournaments. with 48 Teams, playing a 20 ropund swiss you are gonna play against all the top teams sooner or later, and tings will even more.

[Codo]

well done Gonzalo and a real nice story.

[Fluffy]

Fluffy, on 2009-February-16, 11:59, said:

We gopt some of our luclk back on the consolation swiss.

With matches of 7 boards (8 rounds), where winning by 1 IMP is 18-12 already, and you can score 30-0 total, it is very important that you do not lose matches and avoid disasters.

Brenner couldn't play, and with 5 minutes to start we asked the organization to find us a substitute, they got us a junior Caddy from the staff, I offered to play with her because I am used of playing with pickup partners. She introduced herself as Ruri Ohte and we arranged a system in 5 minutes.

We kept things as simple as possible, no thin games, no competitive beyond level, and from the total of 56 boards, she played all of them almost perfect, she made every contract she played, nothing wrong on defence, nothing spectacular either, but that is not what we aimed at. You could be named Zia, Meckstroth or Bochi, and you would not win more IMPs than she did .

After 7 rounds, we were first with 186 VPs, second place where the Dutch with 156, so for the last round all we had to avoid 30-0 (less than 28 IMPs), very exciting because they really pushed in, board 4 they played a "decent" vulnerable slam that went off (other table played part score!) and it was over.


No matter how objective you are, if you win with a tournament, you are gonna like the scoring method . But I think there is a point for making shorter matches in swiss tournaments. with 48 Teams, playing a 20 ropund swiss you are gonna play against all the top teams sooner or later, and tings will even more.

Resurrecting this to celebrate that Ruri Ote (now Ruri Kato) just brillianty qualified for the play offs of the venice cup. Now they face the winners of the round robin: China. Hopefully more to come!