[JLOGIC]

How is it easier to find swisses scored by win-loss? Those don't exist here lol.

A 1 session swiss is a non serious event played sometimes in clubs or as a side game if you get knocked out of something at a small tournament. I'm sorry we aren't as rich as every European country but it is costly and time consuming to pre dupe for a stupid small event like that.

Even a 2 session swiss at a regional which is a more serious event would not have predupes. In Gatlinburg there were 102 tables in the 2 session swiss (these are generally only held on the last day of the tournament). It seems like pre dupiing 8 matches for 102 tables would still be time consuming and costly, for not much gain, it is fine if the best teams are not always winning these events and there's some more randomness, otherwise it would not be much fun for the weaker teams to just have no shot at all. I'm sure in Europe these tournaments are of equal size and frequency, and every country would make sure they were pre duped, but America sucks!!! People just have to make due playing against Meckstroth, Rodwell, Hampson, Passell, Pepsi, Grue, etc etc at their local tournaments on a regular occurence without predupes. In Europe it is much better.

[gwnn]

Bridge is a statistical game. If you bid a 50% game red and they don't, there's a 50% chance you win 10 imps and 50% that you lose 6. If you win a match by 1 imp, there's a large probability (almost 50%) that you were lucky and your bridge decisions were in fact worse than your opponents. So it makes sense to give both teams almost the same number of victory points for a match that was decided by only 1 imp. It wouldn't make sense to give 1 point to the winner and 0 points for the loser.

In basketball, football, archery, chess, bowling, etc, you know that if you do the absolute correct thing, you win (in some of these examples it is easy to say what the absolute correct thing is, in some it's not so easy). In bridge you don't know that, except that there is a good chance of winning. It doesn't make sense to disregard this fundamental distinction.

[JLOGIC]

gwnn, on 2011-May-29, 04:23, said:

Bridge is a statistical game. If you bid a 50% game red and they don't, there's a 50% chance you win 10 imps and 50% that you lose 6. If you win a match by 1 imp, there's a large probability (almost 50%) that you were lucky and your bridge decisions were in fact worse than your opponents. So it makes sense to give both teams almost the same number of victory points for a match that was decided by only 1 imp. It wouldn't make sense to give 1 point to the winner and 0 points for the loser.

Very well said

[campboy]

JLOGIC, on 2011-May-29, 03:16, said:

How is it easier to find swisses scored by win-loss? Those don't exist here lol.

Eh? My point was that it's easier to find pre-duped swisses than win-loss, even if it's still not very easy.

As to being costly for little gain, I don't think many events would be duplimated if this was the main reason to do it. The reason almost all events in England are duplimated is that the players like getting hand records at the end of a session.

[han]

If you look at baseball's regular season for an example (or the NBA), you might notice that the same pairs of teams are playing each other multiple times. At the end of the season, you don't add things up and say, well the Yankees won ten out of eighteen games with the Red Sox so we give the Yankees a "WIN" for that series... and the Yankees won two out of eighteen games with the Rays so that's a "LOSS" and they are 1-1. No, you count every single game separately even if it's against the same team. If you look at a bridge event in this way, and say that every single board should count in the standings rather than somehow combining into a "total result against that team"... well, you get BAM scoring... but it's a lot closer.

[shintaro]

The main problem running Swiss Teams where you do not have Pre-Duplicated Boards is what I will term the 'Lottery Act'

If you always go to a Table and Deal the Boards and those boards are then only played between your 2 teams Then it totally depends on what hands are dealt as to what you get to play with.

So not a very level playing field when other tables are playing either a worst boring set of hands or indeed a fine set of possible meaty Slam hands.

[rduran1216]

rogerclee, on 2011-May-28, 13:07, said:

Aaron you have provided a lot of evidence that the 20VP scale is capable of selecting a winning team which did not play the best bridge. If this is your only point, then I think we can all agree with you and move on.

That was my only point. Added on is that 20pt. vp scale is prone to this kind of crap, since apparently everyone has a story about winning all matches and still losing.

[Bbradley62]

rduran1216, on 2011-May-29, 10:27, said:

That was my only point. Added on is that 20pt. vp scale is prone to this kind of crap, since apparently everyone has a story about winning all matches and still losing.

This makes it sound like you still think W/L is more likely than VP to reward the team that "plays the best" with a championship; no one here has agreed with you. Most do agree that neither approach is perfect; each is capable of failing to reward the team that plays the best.

[gnasher]

In the days when English Swiss teams events were hand-dealt, I used to try to get to the table before the opponents so that I could make sure all the packs were properly shuffled.

[rduran1216]

Bbradley62, on 2011-May-29, 10:57, said:

This makes it sound like you still think W/L is more likely than VP to reward the team that "plays the best" with a championship; no one here has agreed with you. Most do agree that neither approach is perfect; each is capable of failing to reward the team that plays the best.

none of my points about hand shuffled short matches where all teams play each other have been addressed. For people who are supposedly quite logical, you all seem to be overlooking that this is the only consideration here. Since all teams aren't playing the same boards, the nature of the boards will dictate the possible margins of victory with any amount of skill disparity. So its foolish to use margin's to ultimately decide the winner. Pre-duped, and larger matches where not all teams play against each other are a different story. But I can't believe people disagree with w/l in a hand shuffled, round robin style is inferior to VPs.

[gnasher]

rduran1216, on 2011-May-29, 12:07, said:

none of my points about hand shuffled short matches where all teams play each other have been addressed. For people who are supposedly quite logical, you all seem to be overlooking that this is the only consideration here.

If you wanted us to discuss only all-play-all events, perhaps you shouldn't have chosen the subject "Swiss Team Scoring"?

Quote

Since all teams aren't playing the same boards, the nature of the boards will dictate the possible margins of victory with any amount of skill disparity. So its foolish to use margin's to ultimately decide the winner. Pre-duped, and larger matches where not all teams play against each other are a different story. But I can't believe people disagree with w/l in a hand shuffled, round robin style is inferior to VPs.

You're assuming that the volatility of the boards is the only determinant of how big the winning margin is. Whilst that can sometimes be a big factor, the way that people played is usually more significant. A 2-IMP winning margin may be because there were no opportunities for swings, but it's more likely to be because the opportunities were there but they weren't taken. Do you really expect us to believe that in the entire event you had no opportunity to do better than you did?

FWIW, the English Bridge Union does hold one event which attempts to address the problem you're concerned about. It's explained on pages 3-4 of this document:
http://www.ebu.co.uk...Pachabo2010.pdf

[awm]

Here's a counter-argument about W/L.

There is a lot of randomness in a short IMP match. Regardless of whether we are playing identical boards at each table, there could easily be a hand where someone takes an anti-percentage bid or play that happens to work out and swings a lot of IMPs. It's pretty easy to imagine that the team which played better could end up losing by a few IMPs in such a random situation. However, if one team wins by a lot in a short match, it's a good bet that the winner played better than the loser. So perhaps we should only be counting "convincing wins" here -- anything that's less than an IMP a board could count as a tie, and winning (or losing) by more than this margin would be necessary to score a win (or a loss).

But of course, my "an IMP a board" was quite arbitrary. You could pick any amount here and argue for it. For this reason, it makes sense to use a VP-type scale, where you need to win by a lot to score the full "winning" amount, but there is still some reward for winning by a little.

[gwnn]

In what universe is it difficult to find a 5-3 fit in a constructive auction?

1♠-1N
2♥-2N
3♥-4♥

wow!

[shintaro]

rduran1216, on 2011-May-29, 12:07, said:

none of my points about hand shuffled short matches where all teams play each other have been addressed. For people who are supposedly quite logical, you all seem to be overlooking that this is the only consideration here. Since all teams aren't playing the same boards, the nature of the boards will dictate the possible margins of victory with any amount of skill disparity. So its foolish to use margin's to ultimately decide the winner. Pre-duped, and larger matches where not all teams play against each other are a different story. But I can't believe people disagree with w/l in a hand shuffled, round robin style is inferior to VPs.

As you have put here you are not talking about Swiss Teams anyway but a Multiple Teams event where ALL play ALL so in fact you could ALL play the same boards anyway

[Mbodell]

awm, on 2011-May-29, 13:33, said:

Here's a counter-argument about W/L.

There is a lot of randomness in a short IMP match. Regardless of whether we are playing identical boards at each table, there could easily be a hand where someone takes an anti-percentage bid or play that happens to work out and swings a lot of IMPs. It's pretty easy to imagine that the team which played better could end up losing by a few IMPs in such a random situation. However, if one team wins by a lot in a short match, it's a good bet that the winner played better than the loser. So perhaps we should only be counting "convincing wins" here -- anything that's less than an IMP a board could count as a tie, and winning (or losing) by more than this margin would be necessary to score a win (or a loss).

But of course, my "an IMP a board" was quite arbitrary. You could pick any amount here and argue for it. For this reason, it makes sense to use a VP-type scale, where you need to win by a lot to score the full "winning" amount, but there is still some reward for winning by a little.

Exactly right. In fact you can think of the VP score as giving you the chance that the better team won (rounded to the nearest 5%). So if you win 14-6 it is saying there is a 70% chance, based on the outcomes of the boards, that you are the better team. And as a result you get 0.7 Wins, and the loser gets 0.3 Wins. When you win by 45 IMPS and get 20-0 VP then you get the full 1 win. When you win by 1 IMP then you are thought of as a 55% chance of being better and get 0.55 Wins. It is more accurate than W-L because the granularity is better and more data is used.

It, of course, would be better to use preduplicated boards. It would also be better to play longer matches. And to be able to cut the field and play even longer matches. But isn't always practical. It would be worse, however, to use the 30 VP scale and worse still to use the W/L scale.

[sfi]

I almost had a similar experience a couple of years ago. In a 12 round team event from a 14 team field (so almost a round robin), scored via international-style VPs, the final was:

Winners: 12-0, 226 VPs
Us: 7-5, 223 VPs

Had we lost our last match 20-10 instead of 24-6, we would have won the event. Scoring via win-loss, we would have finished in a tie for 5th rather than well clear of 3rd.

All this despite duplicated boards.

So is it better to keep getting small wins or to be able to convincingly beat lots of teams but also get small losses? It seems the real answer is that it depends on the scoring, since tactics will change depending on it.

[Vampyr]

awm, on 2011-May-29, 13:33, said:

there could easily be a hand where someone takes an anti-percentage bid or play that happens to work out and swings a lot of IMPs. It's pretty easy to imagine that the team which played better could end up losing by a few IMPs in such a random situation.

So players shouldn't be rewarded for "anti-percentage" actions which work? Perhaps all of the bids and plays on every hand should be recorded and the scores could be decided by a panel judging for merit rather than by the results of the boards.

[rogerclee]

Vampyr, on 2011-May-30, 17:26, said:

So players shouldn't be rewarded for "anti-percentage" actions which work? Perhaps all of the bids and plays on every hand should be recorded and the scores could be decided by a panel judging for merit rather than by the results of the boards.

Where did awm or anyone say or even imply this at any point in the thread?

[Vampyr]

rogerclee, on 2011-May-30, 18:28, said:

Where did awm or anyone say or even imply this at any point in the thread?

Six posts above yours.

[mrdct]

JLOGIC, on 2011-May-29, 03:16, said:

I'm sorry we aren't as rich as every European country but it is costly and time consuming to pre dupe for a stupid small event like that.

That's an interesting attitude which certainly seems to be the prevailing point-of-view in ACBL-land and no doubt saves the tournament convenors quite a bit of money in dealing costs and volume of boards required to run an event.

I haven't played bridge in Europe for several years, but I can certainly say that in Australia there would be virtually no congresses, national events or state events that don't use pre-dealt boards and I'd say the vast majority of bridge clubs these days either have a dealing machine or get their boards dealt by a third-party such that the overwhelming majority of club duplicates played in this country are with pre-dealt boards. I'd suggest that if someone put on a tournament here and asked the players to shuffle and deal there would be people asking for their money back and a lot less people turning up for the next event run by the same convenor. Whilst you might consider "stupid small events" unworthy of pre-dealt boards, if anything it's those events which provide good opportunities for lower-flight players to improve through going through the hand records and discussing them with their partners and more experienced players. Quite a lot of clubs in Australia, particularly the privately-run clubs, do formal talks after the duplicate to quickly whiz through a few interesting hands which is a bit of a selling point. I think it costs about $15-$20 to get a set of boards dealt, so you only need one extra table showing up because they like hand-records to make it profitable to do it. If I was running a private bridge club, I'd hand-duplicate the boards if necessary until I could afford to buy a dealing machine.

As it happens, I was playing in a little country event a couple of weeks ago with my daughter and because of an odd number of teams they needed to have a triangle which brought boards 33-40 into play which came to the table flushed* so we had to shuffle and deal. This was the first time since 1993 that I'd played bridge with hand-dealt cards.

*As an aside this was quite funny situation as the cards in the flushed boards were face down so we initially all took our hands out and sorted them (I had the ♣ suit) and we all had poker faces except for the clown sitting east who decided to call the director and spoil everyone's fun (especially his partner's who held the ♠ suit). I went a little bit Secretary Bird on him suggesting that it was inappropriate for him to call the director and give UI to his partner that he is holding an extraordinary hand and that what he should've done is shut-up and bid and play the hand and then when we get to the next board call the director and say "I think I recognise this hand".