One round in arrears

[Vampyr]

A poster recently mentioned that his NBO play Swiss events (at least pairs) one round in arrears. The EBU used to do this but changed the practice a few years ago. Whether the reason was that Bridgemate 2 made it easier or to shut me up on the subject is unclear. Anyway here is an account of an event under the old system. I don't know why I remember this one in particular, but I do, so I will give an accurate account instead of making one up. I will list the events of the first few matches, including the starting scores in VPs at the beginning of the match.

 

1. Played a (random draw) very good pair and had a big win, say 20-0.

 

2. Played another (random draw) very good pair and lost, say 20-0. Were matched with 20/20(us):could be anything (them). Obviously if we had played the teams the other way round the proceeding would be different in predictable ways.

 

3. Played another very good pair, based on our score from the first round. Say we beat them 20-0 (improbable because this was against a forum member that I never, or almost never, beat. Anyway starting scores 20/40: could well be 40/40.

 

4. So now we will have been matched with another pair who started round 3 with 20 and lost match 3 heavily. 40/60:20/60.

 

So four rounds have gone by and I still haven't had an appropriate pairing. I guess the scores are likely to even out in the next 10 rounds, but it is not assured. This event is going to end up under-Swissed, and the results will be derived from fairly random pairings.

 

Or am I missing something?

[gordontd]

I don't think anyone doubts that current round assignments are better, but they weren't practical before Bridgemates. Even with Bridgemates they slow things down a little, but not as much as I had thought they would. The extent to which it matters is affected by how many rounds there are. In a 24-round event it wouldn't make that much difference. In the 7-round events we commonly have here the impact is much greater, but there's always going to be a fairly large element of random-ness in the draws.

[pran]

A poster recently mentioned that his NBO play Swiss events (at least pairs) one round in arrears. The EBU used to do this but changed the practice a few years ago. Whether the reason was that Bridgemate 2 made it easier or to shut me up on the subject is unclear. Anyway here is an account of an event under the old system. I don't know why I remember this one in particular, but I do, so I will give an accurate account instead of making one up. I will list the events of the first few matches, including the starting scores in VPs at the beginning of the match.

 

1. Played a (random draw) very good pair and had a big win, say 20-0.

 

2. Played another (random draw) very good pair and lost, say 20-0. Were matched with 20/20(us):could be anything (them). Obviously if we had played the teams the other way round the proceeding would be different in predictable ways.

 

3. Played another very good pair, based on our score from the first round. Say we beat them 20-0 (improbable because this was against a forum member that I never, or almost never, beat. Anyway starting scores 20/40: could well be 40/40.

 

4. So now we will have been matched with another pair who started round 3 with 20 and lost match 3 heavily. 40/60:20/60.

 

So four rounds have gone by and I still haven't had an appropriate pairing. I guess the scores are likely to even out in the next 10 rounds, but it is not assured. This event is going to end up under-Swissed, and the results will be derived from fairly random pairings.

 

Or am I missing something?

I don't know this to be any problem here in Norway, our smallest Swiss events for pairs (at club Level) very seldom (if ever) have less than 9 rounds.

 

A "standard sized" event for 20+ tables has 15 or 16 rounds with 4 boards/round.

 

In your case your position after the first few rounds could be very favourable, you would be able to "ride" on what we name a "Monrad wave" and end up close to top. I was in a similar situation many years ago but still ended up 4th (among 46 pairs) after 15 rounds.

[Cascade]

I don't think anyone doubts that current round assignments are better, but they weren't practical before Bridgemates. Even with Bridgemates they slow things down a little, but not as much as I had thought they would. The extent to which it matters is affected by how many rounds there are. In a 24-round event it wouldn't make that much difference. In the 7-round events we commonly have here the impact is much greater, but there's always going to be a fairly large element of random-ness in the draws.

 

I don't think the comment about 'before Bridgemates' is accurate. With pickup slips the scores can be entered almost as soon as the last table is finished playing and a draw can using be made in a few seconds.

[pran]

I don't think the comment about 'before Bridgemates' is accurate. With pickup slips the scores can be entered almost as soon as the last table is finished playing and a draw can using be made in a few seconds.

What delays the event is the time needed for players to be told where to go for the next round. There is a significant difference when this must be done between rounds rather than during rounds.

[gordontd]

What delays the event is the time needed for players to be told where to go for the next round. There is a significant difference when this must be done between rounds rather than during rounds.

And it's much faster when done by Bridgemates than requiring everyone to consult a list.

[Vampyr]

And it's much faster when done by Bridgemates than requiring everyone to consult a list.

 

If the list were projected onto a wall it woildn't be too bad.i have seen this done in Europe.

[Vampyr]

In the 7-round events we commonly have here the impact is much greater, but there's always going to be a fairly large element of random-ness in the draws.

 

Well of course, but matching people on the same score is the main thing,

 

 

In your case your position after the first few rounds could be very favourable, you would be able to "ride" on what we name a "Monrad wave" and end up close to top. I was in a similar situation many years ago but still ended up 4th (among 46 pairs) after 15 rounds.

 

I don't want to win that way, well actually I do because I want to win, but if I can do that someone else can do it when I would have won the event "properly".

[pran]

And it's much faster when done by Bridgemates than requiring everyone to consult a list.

But even With Bridgemates it must be done, and takes up time between the rounds.

 

Not so when results and next round seating informastion is handed out to each pair during a round.

 

Bridgemates may reduce the time wasted, but they cannot eliminate it.

[Trinidad]

But even With Bridgemates it must be done, and takes up time between the rounds.

 

Not so when results and next round seating informastion is handed out to each pair during a round.

 

Bridgemates may reduce the time wasted, but they cannot eliminate it.

That is technically true. But if the scoring software reads the bridgemates continuously, then it will be able to score the round when all scores are in. It will take a second to come up with the new seatings and to send them to the bridgemates.

That means that the players will get:

• 
   
  
• the right assignments: the assignments are for them personally
  
• at the exact moment when they need them: at the start of the next round.
  
• at the exact place where they need them: at the table where they are sitting.
  

 

If you hand out paper assignments during the next round, the players will get the right assignments (if the TD doesn't make a mistake), get them where they want, but not when they want. In practice, they will have to find their assignment slip between convention cards, score cards, bidding boxes, coffee cups, etc.

 

If you project the assignments on a wall, inevitably not everybody in the room will be able to see that wall, so the players don't get the assignment where they want it. In addition, they don't get the right assignment: all the assignments will be listed and the players will have to figure out which ones are theirs.

 

I don't have a particular problem with a one round delay in pairings for any swiss event (as long as the rounds are short and there are many rounds). But the argument that it takes so much time to do the pairing between the rounds doesn't hold, IMO, as long as your software is capable of sending the new assignments to the bridgemates (which, to the best of my knowledge, is possible from Bridgemate II).

 

Rik

[TylerE]

I don't think it really matters that much. In my view, swiss pairs is just a hopeless format.

[Vampyr]

I don't think it really matters that much. In my view, swiss pairs is just a hopeless format.

 

Why do you think so? I like Swiss Pairs a lot. And it keeps interest in the event even if you are doing badly, because you can still have a moral victory (and matchpoints, if you care) when you win a match.

[gnasher]

Swiss pairs with short rounds, as played in (at least) Norway and Iceland, is excellent. It's trivially better than a multi-section Mitchell event, because:

- Weak pairs have less influence on who wins.

- It largely eliminates the effect of random differences in strength between lines, sections and directions.

- By the end, the winners will usually have played all the other pairs who were in contention.

 

The longest English Swiss pairs events, like the 14-round Brighton Pairs, have similar benefits. The more common 7-round one-day Swiss format is quite random, because a lucky or unlucky draw has too big an influence on the result, and because there aren't enough rounds to differentiate the top pairs. But these events are also popular, and getting people to play is quite rightly one of the primary objectives of most bridge organisers.

 

And, of course, all Swiss events have the advantage that everyone gets the quality of game they have earned. Play well and you get good opponents; play badly and you get what you deserve.

[Vampyr]

The more common 7-round one-day Swiss format is quite random, because a lucky or unlucky draw has too big an influence on the result, and because there aren't enough rounds to differentiate the top pairs. But these events are also popular, and getting people to play is quite rightly one of the primary objectives of most bridge organisers.

 

Holding Flight A, B and C events might improve the above by making the fields smaller. But i believe that on the few occasions the EBU has offered Flight B, there has been a very low entry. I guess no one including the weaker pairs wants to play in a game where they will meet only weaker pairs.

[Trinidad]

Holding Flight A, B and C events might improve the above by making the fields smaller. But i believe that on the few occasions the EBU has offered Flight B, there has been a very low entry. I guess no one including the weaker pairs wants to play in a game where they will meet only weaker pairs.

Flighting a Swiss events seems somewhat strange to me. After all, the idea of the Swiss method is that it flights itself.

 

To flight a Swiss event would have all the drawbacks of flighted events (e.g. pairs playing in the wrong flight, not getting compared to everybody) without the advantages.

 

I can see only one advantage of flighting: You get two separate fields. If you are at the top in a Swiss IMP pair, and your opponents bid the obvious slam, you would not lose any IMPs if only A flight players are in your field. But in a big field, those palookas at table 57 will have enough trouble bidding game and cost you IMPs.

 

I think I could easily solve that drawback of Cross-IMPs in a Swiss event (by weighting the results before averaging, in such a away that the results from neighboring tables weight heavier than those from tables far away), but I think that most players would see it as a drawback that the Cross-IMP result for +1430 depends on where you stand in the field.

 

Rik

[Cascade]

If the list were projected onto a wall it woildn't be too bad.i have seen this done in Europe.

 

Multiple times with networked computers. I thought this would be standard.

[gordontd]

Holding Flight A, B and C events might improve the above by making the fields smaller. But i believe that on the few occasions the EBU has offered Flight B, there has been a very low entry. I guess no one including the weaker pairs wants to play in a game where they will meet only weaker pairs.

No, we used to have more flighted events but now the only one that remains is at the Easter Congress in London. The reason for their decline is not that many players want to play up in a higher flight, but that there simply aren't enough new players starting to play in congresses and those who have played for a long time are no longer eligible. You can see that by how few there are in the B & C groups of the stratified events that have largely replaced the flighted ones.

[gordontd]

Multiple times with networked computers. I thought this would be standard.

No, standard is to send them directly back to the Bridgemate, with one or two large electronic displays for those who can't remember where they were sitting or simply can't be bothered to go back there.

[gnasher]

No, standard is to send them directly back to the Bridgemate, with one or two large electronic displays for those who can't remember where they were sitting or simply can't be bothered to go back there.

Or didn't manage to get to the Bridgemate before someone else had pressed OK twice, after which it's too late to see the assignments.

[Vampyr]

No, we used to have more flighted events but now the only one that remains is at the Easter Congress in London. The reason for their decline is not that many players want to play up in a higher flight, but that there simply aren't enough new players starting to play in congresses and those who have played for a long time are no longer eligible. You can see that by how few there are in the B & C groups of the stratified events that have largely replaced the flighted ones.

 

I don't think that I have ever played in a stratified event I the EBU, but in any case doesn't it make sense to replace whatever the current qualification is with NGS ratings?

[Vampyr]

Flighting a Swiss events seems somewhat strange to me. After all, the idea of the Swiss method is that it flights itself.

 

 

Not if there are too few rounds for the number of teams.

 

I think I could easily solve that drawback of Cross-IMPs in a Swiss event (by weighting the results before averaging, in such a away that the results from neighboring tables weight heavier than those from tables far away), but I think that most players would see it as a drawback that the Cross-IMP result for +1430 depends on where you stand in the field.

 

Cross-IMPs not relevant, since mainly this is about Swiss Pairs.

[blackshoe]

Cross-IMPs not relevant, since mainly this is about Swiss Pairs.

Swiss Pairs is a format of play. Cross-IMPs is a method of scoring. Surely the two are not mutually incompatible — although I grant that Cross-IMPs may not be the form of scoring you're used to in Swiss Pairs. It may not even be a very good method. IAC I thought what this was about is how quickly one should (or can) determine the pairings in such an event. In which discussion perhaps all scoring methods are irrelevant.

[pran]

Swiss Pairs is a format of play. Cross-IMPs is a method of scoring. Surely the two are not mutually incompatible although I grant that Cross-IMPs may not be the form of scoring you're used to in Swiss Pairs. It may not even be a very good method. IAC I thought what this was about is how quickly one should (or can) determine the pairings in such an event. In which discussion perhaps all scoring methods are irrelevant.

Quite correct.

 

And the Bridgemate system includes a feature to make faster one round arrears possible: They name it 95% scoring, and it works by calculating the next round seatings as soon as 95% of all results in the current round are available. Seating informations for the next round are then distributed to all players while the last 5% of the boards in the round are still played.

 

To my knowledge this feature has never received any popularity in Norway.

[gordontd]

I don't think that I have ever played in a stratified event I the EBU,

Of course you have. Most Swiss events in the EBU are stratified, but you wouldn't notice because you aren't eligible for the B or C strat master points or prizes.

[blackshoe]

And the Bridgemate system includes a feature to make faster one round arrears possible: They name it 95% scoring, and it works by calculating the next round seatings as soon as 95% of all results in the current round are available. Seating informations for the next round are then distributed to all players while the last 5% of the boards in the round are still played.

 

To my knowledge this feature has never received any popularity in Norway.

Has it ever actually been used? B-)