[Oceanss]

For those of us that can´t see perfectly cold slam, unless vugraph commentator draws it for us in 3 languages and all colors, could it be possible to have a button, similar to GIB button, that would give Best contract or Par Score for that board?
(something Bridge Composer does http://bridgecomposer.com/Par.htm )

For example: NS 2N; EW 3♠; NS 3♥; NS 4♦; NS 2♣; Par +100; EW 4♠x-1

Thanks !

[uday]

A nice idea. Not sure if it will get much use. I'll let people like the idea here (or ignore it) and we'll see.

[Siegmund]

I like the par score idea. (It is 'expensive', software-wise... you certainly want to store it if you calculate it once, but not sure you want to calculate it for every deal unless it is requested.)

If you DO include a list of making contracts - PLEASE report the full 4x5 box of how many tricks, NOT just a list of which contracts can be made.

[hotShot]

Although it seems like a nice feature, i see a little problem with it.

Assume a hand that would require you to bid 3NT using your methods, but the DD sees that you can make only 8 tricks. Should the par Score be 3NT-1 or 2NT ?
Or the other way round
Using your methods you found that 2 keycards are missing so you did not bid slam. Fortunately the trump K will drop/the finesse is working so that DD the slam is always made. Should the par score be 6M or 5M+1 ?

Based on your methods neither you nor your partner was able to enter the auction, but your side could make 3♣-1 when opps played 2M. What should the par score be?

Perhaps one could take the par score from the result list e.g. the contract bid on most tables, but that is most likely biased too.

[barmar]

If we add par score display, we'll almost certainly do it the same way as is done on hand records produced by BridgeComposer and other similar applications. It will be the double dummy par, because anything else is too subjective.

It's not a perfect measure, but it's no less useful than the list of all the DD possible results that are commonly included on hand records.

[wynsten]

barmar, on 2014-May-15, 09:41, said:

If we add par score display, we'll almost certainly do it the same way as is done on hand records produced by BridgeComposer and other similar applications. It will be the double dummy par, because anything else is too subjective.

It's not a perfect measure, but it's no less useful than the list of all the DD possible results that are commonly included on hand records.

I'd like to see a scoring version, similar to cross IMPS, where the comparison is against par. i.e. if par is 100NS and NS score 420 then NS score +IMP(320) and EW score -IMP(320) where IMP(x) is the IMP's received for a differential of x points.

Not a perfect scoring system (since par sometimes requires bad play or bidding to achieve) but pretty good. When par is 600 (for 3NT making) and I score 630, it is no fun scoring negative IMPs because some idiot EW made a ridiculous sac.

[barmar]

wynsten, on 2014-May-22, 09:45, said:

Not a perfect scoring system (since par sometimes requires bad play or bidding to achieve) but pretty good. When par is 600 (for 3NT making) and I score 630, it is no fun scoring negative IMPs because some idiot EW made a ridiculous sac.

How would that happen? Suppose there are 31 tables. If everyone else scores 600, and you score 630, you get 1 IMP (1 IMP per pair, divided by 30 pairs). If one pair takes a sac and goes for 800, you get 0.83 IMPs (1 IMP x 29 - 5 IMPs, all divided by 30). You don't get below 0 IMPs unless there were at least 6 pairs taking that sac (how ridiculous can it be if 20% of the field took it?).


But it doesn't really matter. Yes, you might have gotten a negative score, but all the people who scored only 600 got an even lower negative score. Positive or negative isn't important by itself, it's just your score compared to other players that matters. All those pairs whose opponents took the sac will score 800 versus the 600 par, and they'll get 5 IMPs compared to your 1 IMP, so they'll still be 4 IMPs ahead of you.

[42krunner]

hotShot, on 2014-May-15, 07:35, said:

Although it seems like a nice feature, i see a little problem with it.

Assume a hand that would require you to bid 3NT using your methods, but the DD sees that you can make only 8 tricks. Should the par Score be 3NT-1 or 2NT ?
Or the other way round
Using your methods you found that 2 keycards are missing so you did not bid slam. Fortunately the trump K will drop/the finesse is working so that DD the slam is always made. Should the par score be 6M or 5M+1 ?

Based on your methods neither you nor your partner was able to enter the auction, but your side could make 3♣-1 when opps played 2M. What should the par score be?

Perhaps one could take the par score from the result list e.g. the contract bid on most tables, but that is most likely biased too.

How is this any different from regular hand results? Some of the hands, you just have to ignore the par. It is the 80% where it is right that helps us learn.

[wynsten]

barmar, on 2014-May-22, 10:52, said:

How would that happen? Suppose there are 31 tables. If everyone else scores 600, and you score 630, you get 1 IMP (1 IMP per pair, divided by 30 pairs). If one pair takes a sac and goes for 800, you get 0.83 IMPs (1 IMP x 29 - 5 IMPs, all divided by 30). You don't get below 0 IMPs unless there were at least 6 pairs taking that sac (how ridiculous can it be if 20% of the field took it?).


But it doesn't really matter. Yes, you might have gotten a negative score, but all the people who scored only 600 got an even lower negative score. Positive or negative isn't important by itself, it's just your score compared to other players that matters. All those pairs whose opponents took the sac will score 800 versus the 600 par, and they'll get 5 IMPs compared to your 1 IMP, so they'll still be 4 IMPs ahead of you.

Yes my score relative to the other players who played the same hand is unaffected (except for the non-linearity of IMPs) but it is affected on an absolute scale.


With a large number of tables (such as 31), cross IMPs (the way BBO scores IMPs now) works well. But as the number of tables gets smaller the oddball result at some other table gets undue emphasis and if the number of tables is one (yours), cross IMPs always results in ---. My suggestion is to score as if there were just two tables - yours and one other where the result was PAR. Essentially it takes the "duplicate" out of duplicate bridge and returns it to a 4 person game. For example you could play it for money at a penny per IMP.

[barmar]

I vaguely recall something like this on OKbridge. When you played in their equivalent of the MBC, it would report your cross-IMPs or matchpoints, and also report an IMP score based on the average expected score for your side's combined HCP. They didn't calculate the par for your actual hands, it was just something like with 25 HCP you could expect to make game. I don't remember what they call this.

[wynsten]

barmar, on 2014-May-26, 19:11, said:

I vaguely recall something like this on OKbridge. When you played in their equivalent of the MBC, it would report your cross-IMPs or matchpoints, and also report an IMP score based on the average expected score for your side's combined HCP. They didn't calculate the par for your actual hands, it was just something like with 25 HCP you could expect to make game. I don't remember what they call this.

And back in the day, bridge federations would compose a set of hands, each with some sort of playing point (like an obscure safety play, for example), and players received points for bidding to the correct contract and making it (or defeating it, if it was a defensive point, like underruffing to avoid being squeezed - always popular). They were called PAR contests.

[barmar]

I've seen reports on par contests in Bridge World. Those are more like puzzles than real play competitions.