[TylerE]

So, had a one session Swiss at the club yesterday...

The following happened...

Setup: 1 session, 11 teams, 4 rounds of 6 boards, STAC Team Game

4x H2H matches, 1 2-round 3-way

Team 6 and 7 were playing in the second round. The result turned in was Team 6 winning in a blitz. The wrong team number was written on the card, and it was actually team 7 that won, but this was not discovered until after round 3.

As a result of team 7 getting "blitzed", they were placed into the 3 way and promptly ate the Flight C fish, and scored 40 VPs in the last 2 rounds, for an event winning 73 VPs.

Team 2, the team that finished 2nd in the event, with a score of 53VPs was extremly unhappy, as they had played strongish (by local standards) teams in all 4 rounds, and more importantly, never gotten to play team 7.

If the proper results had been turned in for round 2, Teams 2 and 7 would have played in round 3.

So, what the heck should director do in this situation?

(Directors actual ruling, which IMO wasn't legal but might be the pragmatic least of all evils, was to award Team 2 and 7 as a mutual tie for 1st.)

[gordontd]

We have regulations in our White Book that mandate score adjustments for such Swiss mismatches, but in the absence of such regulations you do have quite a problem. Of course any Swiss with only four matches is likely to be quite dependant on luck in any case.

[lamford]

TylerE, on 2015-August-31, 13:21, said:

Team 6 and 7 were playing in the second round. The result turned in was Team 6 winning in a blitz. The wrong team number was written on the card, and it was actually team 7 that won, but this was not discovered until after round 3.

Was it director error, or an error by one team captain? It seems like you were at sixes and sevens. In chess this sort of thing can be resolved by the Sonneborn-Berger score of each team, normally used only to break ties, which is the sum of the scores obtained multiplied by the final scores of the teams played. That might be, for example, 13 x 40 + 20 x 40 + 20 x 23 + 20 x 23 for the team that scored 73, 40 of those against fish. For team 2 they might have had, say 13 x 40 + 13 x 45 + 14 x 40 + 13 x 45 for example, for the team that scored 53 against stronger opposition, which happens to be 2250 against 2240 for the team that benefited from the error. But it might have been less, depending on how the opposition scored in other matches.

I wonder if this might be a better method of resolving mismatches, as it takes into account both the actual score and the opposition? I suspect one just has to follow the White Book method now, but it might be worth considering in future?

[TylerE]

Hrrm, that's an interesting thought.

Unfortunately, I don't have the full results matrix, but for my team (Team 2, surprise, surprise) it would be...

For Team 2:

20x45 + 16x49 + 11x24 + 7x46 = 900+784+264+322 = 2270

For Team 7: (I think this is right, might be off by a couple)
13x48 + 20x24 + 20x11 + 20x31 = 624+480+220+620 = 1926

So, according to Sonneborn-Berger Team 2 should get 1st...

[Vampyr]

Wouldn't Swiss Points be a bit simpler than the above?

Were Bridgemates/pads somehow not in use?

[campboy]

lamford, on 2015-August-31, 14:10, said:

I wonder if this might be a better method of resolving mismatches, as it takes into account both the actual score and the opposition? I suspect one just has to follow the White Book method now, but it might be worth considering in future?

I see a couple of problems with using this to resolve mismatches (and suspect there are more I haven't seen yet). Since it works on the final score of opponents, it doesn't distinguish how much of the difference in opponents' scores was because of an actual mismatch, and how much is because of differences in their performance in subsequent rounds. And since you multiply the two together it works much better at reducing the score of a team that wins heavily when they are playing lower-ranked opposition than at increasing the score of a team that loses heavily against higher-ranked opposition (and the latter may be not at fault for the mismatch).

In this case the first isn't so much of a problem as the mismatch occurred so late, but here you have the opposite problem that team 7 is punished heavily by this formula for facing weaker opposition in round 2 -- which was before the mismatch occurred!

[lamford]

campboy, on 2015-September-01, 01:25, said:

In this case the first isn't so much of a problem as the mismatch occurred so late, but here you have the opposite problem that team 7 is punished heavily by this formula for facing weaker opposition in round 2 -- which was before the mismatch occurred!

The formula weights the results based on the opposition met. It is correct that if you play Futile Willie's team in round 1 and win 20-0, that will be worth far less than a 10-10 draw with the Bermuda Bowl winners. However that is not a problem; you were lucky in the first instance to be randomly drawn against Futile Willie, and the formula is designed to compensate for good and bad luck, whether that is through a random correct draw or an erroneous mismatch. There is an argument that the final order should not be VPs won, but Sonneborn-Berger score, which is just a more accurate way of taking into account your actual results.

But, no doubt, too avant garde for bridge authorities, who will stick to the tried and tested methods.

[barmar]

Vampyr, on 2015-August-31, 20:50, said:

Were Bridgemates/pads somehow not in use?

The OP says "The wrong team number was written on the card,". If there was a card, they presumably weren't using electronic scoring devices.

This also seems to answer the above question about whether it was captain or director error. Unless somehow the director filled in the card for them.

[Vampyr]

barmar, on 2015-September-01, 10:22, said:

The OP says "The wrong team number was written on the card,". If there was a card, they presumably weren't using electronic scoring devices.

This also seems to answer the above question about whether it was captain or director error. Unless somehow the director filled in the card for them.

When cards are used, if memory serves, the winning captain brings both teams' cards to the scorer, checking with the other captain that the result is the same. Were both cards mis-scored?

[campboy]

lamford, on 2015-September-01, 07:13, said:

The formula weights the results based on the opposition met. It is correct that if you play Futile Willie's team in round 1 and win 20-0, that will be worth far less than a 10-10 draw with the Bermuda Bowl winners. However that is not a problem; you were lucky in the first instance to be randomly drawn against Futile Willie, and the formula is designed to compensate for good and bad luck, whether that is through a random correct draw or an erroneous mismatch. There is an argument that the final order should not be VPs won, but Sonneborn-Berger score, which is just a more accurate way of taking into account your actual results.

But, no doubt, too avant garde for bridge authorities, who will stick to the tried and tested methods.

I did start my post with "I see a couple of problems with using this to resolve mismatches". My post was not intended as a comment on the merits of using S-B to score the tournament in the first place.

FWIW I think it is a very good idea to compensate for the overall strength of your opponents, but this is a poor way to do it.

[ddrankin]

The ACBL has, in its conditions of contest for Swiss events, the following:

"CORRECTION AND APPEAL PERIODS

PLAYER ERRORS:

No increase in score needs be granted unless the TD's attention is called to the error prior
to the announced starting time of the next match or 30 minutes after the completion of the
match, whichever is earlier."

Since this was a STaC, I think this should be enforced, which would give both teams 0 VP's for the match in question. The losing team is adjusted to 0, and the winning team is not adjusted.

[barmar]

Vampyr, on 2015-September-01, 11:33, said:

When cards are used, if memory serves, the winning captain brings both teams' cards to the scorer, checking with the other captain that the result is the same. Were both cards mis-scored?

Almost no one (in ACBL, at least) turns in two cards these days -- on the rare occasions that I've seen it tried, the director just tosses the one from the losing team. The normal procedure is that the winning team turns in a card. They're supposed to get the captain of the losing team to initial it, but that's also often skipped, except in major events.

[Vampyr]

barmar, on 2015-September-02, 10:59, said:

Almost no one (in ACBL, at least) turns in two cards these days -- on the rare occasions that I've seen it tried, the director just tosses the one from the losing team. The normal procedure is that the winning team turns in a card. They're supposed to get the captain of the losing team to initial it, but that's also often skipped, except in major events.

With such a flawed procedure, the OP situation must happen all the time.

What happens to the losing team's card, anyway? Don't they get it back?

[blackshoe]

barmar, on 2015-September-02, 10:59, said:

Almost no one (in ACBL, at least) turns in two cards these days -- on the rare occasions that I've seen it tried, the director just tosses the one from the losing team. The normal procedure is that the winning team turns in a card. They're supposed to get the captain of the losing team to initial it, but that's also often skipped, except in major events.

Yes. I ran afoul if this the first time I played in a team event. As I was the only male on the team, I got "elected" captain, in spite of having no idea what that really meant. In one match, which we won, the captain of the opposing team came up to verify scores and said "I'll take the scores up". I let her. The match was scored as a win for her team. We discovered that on the way home, and by the time, a couple of months later, that I finally got hold of the director, he said "sorry, but I've thrown away all that, so nothing can be done". Actually, it was surely long past the correction period anyway, even if he'd had the records. I learned a lesson, though.

[blackshoe]

Vampyr, on 2015-September-02, 15:47, said:

With such a flawed procedure, the OP situation must happen all the time.

What happens to the losing team's card, anyway? Don't they get it back?

Nope.

[Vampyr]

blackshoe, on 2015-September-02, 17:50, said:

Nope.

So once you lose a match, you cannot keep track of your scores and you have nowhere to write your assignments? Weird.

[TylerE]

It's not usually a problem in a "real" event with multiple directors where the previous rounds VPs are already on the big board before assignments go up. One director running a small club game is very different.

[blackshoe]

Vampyr, on 2015-September-02, 18:02, said:

So once you lose a match, you cannot keep track of your scores and you have nowhere to write your assignments? Weird.

We would track scores on a private scorecard. Assignments I've never seen written down - well, not table assignments. Those the captain goes at looks at what Tyler calls "the big board" and goes back and verbally informs the pairs of his team where they are to go and, if he doesn't forget, which team they're playing. That team number is more often, in my experience, by getting to the table and asking the opposing team what their number is. Then it's written in the private score. Of course, some players don't keep a private score.

[mycroft]

Oddly enough, it was in the Roth Swiss that I saw people either not asking for an initial or not willing to provide one. Only time I can remember that happening in years (but I play in a specific area most of the time).

Vampyr - it's a match card, not a score card - and nobody gets it back, it's just the way to report the match to the directors. winning team #, losing team #, IMP difference, round, initial by losing team for agreement of score.

Yes, there are issues when people send in "loser" tickets instead of/as well as "winner" tickets, because they're now *so* uncommon that the TD doesn't always catch it. Yes, there are issues with the TD misreading/misentering the ticket (yes, I would like any of the systems where everything's electronically scored and matched, and we just build the leaderboard, why do you ask?) *Usually* it's not a problem, and we catch things in time to do silly things like rematch the gen'ed match (to put the right winning or losing team in) and fix it in the computer a bit later. This one seems particularly unfortunate, and I'm not sure what I would have done.

If the "losing" team were at all familiar with the way ACBL swissing works, I'd be asking why nobody wondered why a team at 33/40 after 2 would have "qualified" for the RR. I know I certainly got asked that question frequently in the 7-team games I used to run as 2x2 + RR (where due to the matching constraints, one high-scoring team will almost always be in the RR round 3) - one of the reasons I switched to Whist League 6x5 full round robin for 7-team events. If I thought there was anybody who *did* think that but went off to shoot fish in a barrel anyway, I'd be somewhat concerned.

[blackshoe]

mycroft, on 2015-September-03, 10:06, said:

Vampyr - it's a match card, not a score card - and nobody gets it back, it's just the way to report the match to the directors. winning team #, losing team #, IMP difference, round, initial by losing team for agreement of score.

Yes. I think of it as like a pickup slip for teams.