[Manastorm]

I suggest that a sample of NT ish hands is given to a pair and see how they actually bid. During a competition as long as the pair stays within the range subject to
a statistical test, they are clean. If they deviate too much from the test sample, they have given misinformation and are breaking the rules.
I would really like that to see that all agreements would be pretested by a test sample and only then allowed to be used. If you want to use stayman, you better test how you actually use it and follow it. How to actually do all this, no idea. I have a feeling this can't be too difficult to achieve with all the computer techonolgy we have.

[sfi]

I'm not sure how practical this would be - there are rather a lot of pairs who play this game. But my more fundamental question is what problem are you trying to solve?

[hrothgar]

Here's a related thread

http://bridgewinners...an-opening-bid/

[Manastorm]

sfi, on 2019-September-03, 05:04, said:

I'm not sure how practical this would be - there are rather a lot of pairs who play this game. But my more fundamental question is what problem are you trying to solve?

It is about what the common 1NT-range consists. Some disclose 15-17bal, but actually upgrade many 17s to the higher NT-range, never downgrade 18s and upgrade 25% of 14balanced from the 1st position. When taken into account vulnerabilities, positions and sometimes match situation the range spreads even more. I would like to know what opponents actually play, when they claim to have an agreement. I don't think it is wrong to expect evidence for the claim, instead the other way around. How can it even be proved reasonably that the 1NT-range isn't what is claimed. Tactical bids and psyches are artificial names for bids which are wildly out of expected range, maybe they are in the range after all. If you can see the range from data, what more could you want, why would you want to give less.

[Vampyr]

Manastorm, on 2019-September-03, 07:10, said:

It is about what the common 1NT-range consists. Some disclose 15-17bal, but actually upgrade many 17s to the higher NT-range, never downgrade 18s and upgrade 25% of 14balanced from the 1st position. When taken into account vulnerabilities, positions and sometimes match situation the range spreads even more. I would like to know what opponents actually play, when they claim to have an agreement. I don't think it is wrong to expect evidence for the claim, instead the other way around. How can it even be proved reasonably that the 1NT-range isn't what is claimed. Tactical bids and psyches are artificial names for bids which are wildly out of expected range, maybe they are in the range after all. If you can see the range from data, what more could you want, why would you want to give less.

How would you collect this data?

[barmar]

So before a pair is allowed to enter any competition, they first have to spend a day or more bidding thousands of hands so we can collect statistics about all their bids? Remember, NT ranges are not the only things that need to be disclosed. There's also the criteria for opening 1 bids versus weak 2 bids, how often you open weak 2 with 5-card suits, the kinds of hands you open 2♣ on, what kinds of hands you invite game with, etc.

And players who have participated in competitions like Challenge the Champs admit that they don't always bid the same way when being tested as they would in real life. So perhaps we need to collect this information from actual play. But there's a chicken-and-egg Catch-22: they're not allowed to play an agreement if we don't know their criteria, but we can't determine their criteria until they've been playing the agreement long enough to calculate statistics.

[Manastorm]

Vampyr, on 2019-September-03, 07:20, said:

How would you collect this data?

I have no idea how to do this. I do not expect that the current bridge software could collect and save such data in a useful way. I do not expect anything like that to happen in near future to say the least. However if the data collection is as easy as it is to play in BBO, I would like to test my ranges.

[Manastorm]

barmar, on 2019-September-03, 07:53, said:

So before a pair is allowed to enter any competition, they first have to spend a day or more bidding thousands of hands so we can collect statistics about all their bids? Remember, NT ranges are not the only things that need to be disclosed. There's also the criteria for opening 1 bids versus weak 2 bids, how often you open weak 2 with 5-card suits, the kinds of hands you open 2♣ on, what kinds of hands you invite game with, etc.

And players who have participated in competitions like Challenge the Champs admit that they don't always bid the same way when being tested as they would in real life. So perhaps we need to collect this information from actual play. But there's a chicken-and-egg Catch-22: they're not allowed to play an agreement if we don't know their criteria, but we can't determine their criteria until they've been playing the agreement long enough to calculate statistics.

If the magic software existed: giving suitable samples, collecting it and giving as detailed summary as one wished, then it could be fun. I don't think giving the data itself would be a problem as long as it is playing bridge so to speak, any extra work is a major annoyance. I don't think this would have much relevance outside top level bridge, I could see long term partners using the magic software. I would certainly use it, since a longtime partner is not needed. What comes to bidding so randomly that someone would digress from the straight and narrow, I don't see why the current "rules" should allow it either.

Anyway I find it cumbersome to defend the magic software. I don't wish to answer how this and that could be done since I don't have a clue. This was just wishful thinking.

[rhm]

I made some comments in the Bridgewinner discussion.
I recite Jeff Rubens (Bridgeworld August 2019, page 55), because it is well worth repeating:

"People who write laws that tell others how to evaluate their hands deserve to reside in that circle of Hell in which all the decks have 51 cards."

Full disclosure should be an ethical endeavor.
My impression is it has become for many an obsession and anyone, who does not share their obsession, is considered by them a cheat.
It reminds me of the role of Jakobinism in the French Revolution.

We need to understand that Milton Work Count is not the holy grail of hand evaluation, it is the holy grail of communication about hand strength and as such is fuzzy.
Since we have no common agreement how to evaluate a Bridge hand precisely we have also no good way of communicating this.
If you doubt his look at all those discussions about hand evaluation solely on this forum. One of these discussions was just revived in the Expert_Class section and it alone goes over 19 pages so far.

There was a Bridge book by Irish International David Jackson and Ron Klinger in 2010 called:

"Better Balanced Bidding The Banzai Method"

The book uses the so called Cochran Count where a ten is worth 1 point up to the ace, which is assigned 5 points
Accordingly a deck has not 40 points but 60 points and when you communicate your strength you would need to roughly specify 2/3 of Cochran points to get to standard HCP.

However, look a bit deeper and you will find that this method devalues aces and upgrades lower honors.
A hand like Axxx Axx Axx Axx is not worth a strong notrump (15-17) but comes only to 13.7 and is considered by the proponents of this method a weak notrump
A hand like KQx QJx QJx QJTx is worth a strong notrump (15-17) because it comes to 15.7

Mr. Jackson and Mr. Klinger show numerous deals taken from high level tournaments where the field missed 3NT and their method would have got there.

Now I would consider the first hand too good for a 15-17 notrump and I would not even consider KQx QJx QJx KQxx worth opening a 15-17 notrump. Don't ask me why because that is not important.

But who am I to tell Mr. Klinger or Mr Jackson how to evaluate a Bridge hand?

If you restrict hand evaluation you are shaking at the very foundation, on which this game is based.

Rainer Herrmann

[sfi]

Manastorm, on 2019-September-03, 07:10, said:

It is about what the common 1NT-range consists. Some disclose 15-17bal, but actually upgrade many 17s to the higher NT-range, never downgrade 18s and upgrade 25% of 14balanced from the 1st position. When taken into account vulnerabilities, positions and sometimes match situation the range spreads even more. I would like to know what opponents actually play, when they claim to have an agreement. I don't think it is wrong to expect evidence for the claim, instead the other way around. How can it even be proved reasonably that the 1NT-range isn't what is claimed. Tactical bids and psyches are artificial names for bids which are wildly out of expected range, maybe they are in the range after all. If you can see the range from data, what more could you want, why would you want to give less.

I don't see the same trend of players not properly disclosing their NT ranges that you seem to do so. But I agree with Rainer that it's not just about pure point count. When someone says "15-17", I think it's normal and reasonable to expect that there is some fuzziness around the edges where good 14s (and very occasionally bad 18s) make it in there as well.

The other issue is that it may be right to have a wider opening 1NT range for reasons that are only specific to that round (or even board). Maybe your opponents are looking rattled because you opened 1NT twice in a row and bid well to get a good score. Maybe partner is looking sleepy and you want to make them dummy. Maybe you're just running late and want a quicker auction. How would you expect a statistical analysis to include considerations like these, and do you penalise a player when they use judgement?

[barmar]

Manastorm, on 2019-September-03, 09:09, said:

I have no idea how to do this. I do not expect that the current bridge software could collect and save such data in a useful way. I do not expect anything like that to happen in near future to say the least. However if the data collection is as easy as it is to play in BBO, I would like to test my ranges.

But this means that you could only use this mechanism for pairs who play together long enough to accumulate meaningful statistics. What about new partnerships?

And even with regular partnerships, depending on how often they play it might require years of history to get useful data. But that comes with its own problem: people change style over time, so you'd have to weight recent data more heavily than old data, but the weighting factors would probably be a total guess. If you have enough pairs who play very frequently for many years, you might be able to detect trends and calculate a reasonable average weighting scale.

There are also feedback effects. If the players know that you're calculating statistics like this, they might try to game the system by deliberately varying their style.

[pescetom]

This whole thing about NT range disclosure seems a bit surreal to many of us outside the US. ACBL is apparently hung up on rigid HCP-based ranges and some players seem equally concerned. Is there really a massive phenomenon of people playing wholly different ranges to the ones they declare? Are opponents being damaged? I don't see it over here, and I think there are other areas that present much bigger and more evident problems of disclosure - conventional interference, conventional raises,conventional changes of suit (fit-showing or similar), low level doubles, u.s.w. Even the style of 1NT opening poses a bigger communication problem than the range, which can at least be announced with just 2 numbers.

[mikestar13]

pescetom, on 2019-September-04, 09:37, said:

This whole thing about NT range disclosure seems a bit surreal to many of us outside the US. ACBL is apparently hung up on rigid HCP-based ranges and some players seem equally concerned. Is there really a massive phenomenon of people playing wholly different ranges to the ones they declare? Are opponents being damaged? I don't see it over here, and I think there are other areas that present much bigger and more evident problems of disclosure - conventional interference, conventional raises,conventional changes of suit (fit-showing or similar), low level doubles, u.s.w. Even the style of 1NT opening poses a bigger communication problem than the range, which can at least be announced with just 2 numbers.

Exactly, ACBL is full of Walruses who only count HCP and want to force everyone to do the same.

[rhm]

pescetom, on 2019-September-04, 09:37, said:

This whole thing about NT range disclosure seems a bit surreal to many of us outside the US. ACBL is apparently hung up on rigid HCP-based ranges and some players seem equally concerned. Is there really a massive phenomenon of people playing wholly different ranges to the ones they declare? Are opponents being damaged? I don't see it over here, and I think there are other areas that present much bigger and more evident problems of disclosure - conventional interference, conventional raises,conventional changes of suit (fit-showing or similar), low level doubles, u.s.w. Even the style of 1NT opening poses a bigger communication problem than the range, which can at least be announced with just 2 numbers.

mikestar13, on 2019-September-04, 10:44, said:

Exactly, ACBL is full of Walruses who only count HCP and want to force everyone to do the same.

Sorry to say that, but if I had to summarize the discussion on Bridgewinner (https://bridgewinner...onvention-card/) to my surprise the most notorious Walruses for strict HCP ranges were not representatives from ACBL but

1) David Burn

and

2) Ian Grant

To remain fair to EBU land a voice, which in my mind made many sensible statements, was Richard Fleet.
But he was heavily attacked by David.

Rainer Herrmann

[pescetom]

rhm, on 2019-September-04, 11:03, said:

Sorry to say that, but if I had to summarize the discussion on Bridgewinner (https://bridgewinner...onvention-card/) to my surprise the most notorious Walruses for strict HCP ranges were not representatives from ACBL but

1) David Burn

and

2) Ian Grant

To remain fair to EBU land a voice, which in my mind made many sensible statements, was Richard Fleet.
But he was heavily attacked by David.

I can imagine that NT disclosure is probably more of an issue in EBU land than it is elsewhere, due to the persistence of weak NT and dual or variable ranges. I gave up early on the BW discussion but I can hardly imagine David Burn as a Walrus - probably he felt forced by his "tell them what you play" philosophy to take an extreme position, which can happen in internet discussions - I doubt he would really have any problems with Richard Fleet during an auction. Be that as it may, I think NT range is a minor issue compared to other disclosure problems and that a simple announcement of two numbers works fine. If the RA wants to enforce (say) maximum 1HCP deviation then I guess I can live with that too. I don't remember the last time I deviated by 2HCP although I agree that my right to do so is part of bridge.

[rhm]

pescetom, on 2019-September-05, 06:43, said:

I can imagine that NT disclosure is probably more of an issue in EBU land than it is elsewhere, due to the persistence of weak NT and dual or variable ranges. I gave up early on the BW discussion but I can hardly imagine David Burn as a Walrus - probably he felt forced by his "tell them what you play" philosophy to take an extreme position, which can happen in internet discussions - I doubt he would really have any problems with Richard Fleet during an auction. Be that as it may, I think NT range is a minor issue compared to other disclosure problems and that a simple announcement of two numbers works fine. If the RA wants to enforce (say) maximum 1HCP deviation then I guess I can live with that too. I don't remember the last time I deviated by 2HCP although I agree that my right to do so is part of bridge.

I am aware that David Burn is not a Walrus, because he knows too much about the game, but he wants to protect any Walrus playing the game.
If I play a 15-17 notrump and decide that Axx Axx Axxx Axx is a hand too strong to open 1NT, there would be no announcement and no alert when I open 1♦ instead.
Have I violated full disclosure?
If as explained above Mr. Klinger or Mr Jackson opens the same hand with 1♦ but for the different reason because for them the hand is not good enough for a 15-17 notrump, how do you disclose that?

If I can construct a hand where I would upgrade by 2 points, say ATx ATx AJT9x xx to 1NT or downgrade by 2 points say QJx KQx KQx KQJx to 1NT, do I really write on my convention card 13(+)-19(-) and announce it as such, even though I am unlikely ever to get such a hand in my remaining lifetime?
What if I claim to play this range (13-19) and decide not to open Axx Axx Axxx Axx with 1NT, would there not be some Walrus, who would call the director?
Am I supposed to defend my motives then in front of an ethics commission?

To me this borders on the absurd. That's why I have the impression that full disclosure is for some not an endeavor any more, but become an obsession.

Rainer Herrmann

[barmar]

pescetom, on 2019-September-04, 09:37, said:

This whole thing about NT range disclosure seems a bit surreal to many of us outside the US. ACBL is apparently hung up on rigid HCP-based ranges and some players seem equally concerned. Is there really a massive phenomenon of people playing wholly different ranges to the ones they declare?

The original reason for the NT range announcement had little to do with getting the boundaries precisely right. It was just to let the opponents know whether you use strong or weak NT, since many pairs play different defenses. They could have made the announcements "strong", "weak", and "mini", but everyone has different ideas about where the dividing lines should be. So they went with numbers, and the opponents can use their own criteria to decide which defense applies.

It's only Secretary Birds who then got into the act and decided that these ranges need to be accurate, and call the TD whenever you deviate in the slightest. And that then led to threads like this.

However, if you announce 15-17 and reguarly open 13 counts, you're not being honest in your disclosure.

[pescetom]

rhm, on 2019-September-05, 07:33, said:

I am aware that David Burn is not a Walrus, because he knows too much about the game, but he wants to protect any Walrus playing the game.
If I play a 15-17 notrump and decide that Axx Axx Axxx Axx is a hand too strong to open 1NT, there would be no announcement and no alert when I open 1♦ instead.
Have I violated full disclosure?
If as explained above Mr. Klinger or Mr Jackson opens the same hand with 1♦ but for the different reason because for them the hand is not good enough for a 15-17 notrump, how do you disclose that?

If I can construct a hand where I would upgrade by 2 points, say ATx ATx AJT9x xx to 1NT or downgrade by 2 points say QJx KQx KQx KQJx to 1NT, do I really write on my convention card 13(+)-19(-) and announce it as such, even though I am unlikely ever to get such a hand in my remaining lifetime?
What if I claim to play this range (13-19) and decide not to open Axx Axx Axxx Axx with 1NT, would there not be some Walrus, who would call the director?
Am I supposed to defend my motives then in front of an ethics commission?

To me this borders on the absurd. That's why I have the impression that full disclosure is for some not an endeavor any more, but become an obsession.

I agree with most of what you say here. It is just good bridge to upgrade ATx ATx AJT9x xx or downgrade QJx KQx KQx KQJx, I'm more skeptical about upgrading or downgrading Axx Axx Axxx Axx which just look like 16+ to me (and to Kaplan & Rubens). But yes you have a right to reevaluate hands in your own way and it doesn't make much sense to try to announce your reevaluation methods, even if partner is aware of them.
I suggest that NT reevaluation style is a similar issue to NT shape style, only less important. Both need to be communicated in some way, I agree with David Burn on this. Burying them in the convention card is perhaps not the ideal solution, but certainly better than requiring contorted and UI prone announcements. In a perfect world, the 1NT card could say things like "occasionally upgrades a very good 13 HCP" and "may contain a 6-card minor, any 5-card, a singleton which is an Ace". Once play becomes electronic such information will be communicated automatically with the bid and a good part of the disclosure problem disappears.

[barmar]

pescetom, on 2019-September-05, 09:50, said:

Once play becomes electronic such information will be communicated automatically with the bid and a good part of the disclosure problem disappears.

Why do you think electronic play would solve it? The players still have to decide what details to put into their convention card. You really think they'll get that much more precise just because it's electronic?

Maybe if there were checkboxes for things like "could have 6-card minor", they might tick them. But good luck encoding upgrade/downgrade style.

[pescetom]

barmar, on 2019-September-06, 06:02, said:

Why do you think electronic play would solve it? The players still have to decide what details to put into their convention card. You really think they'll get that much more precise just because it's electronic?

Maybe if there were checkboxes for things like "could have 6-card minor", they might tick them. But good luck encoding upgrade/downgrade style.

I'm not a fan of a checkbox approach which can never be flexible enough. Styles can be described with text, following examples. I really think players would have no issue doing that, so long as they know it is really useful because it will be displayed contextually, not hidden in the small print of a System Card (and of course, opponents would be quick to point out any reticence or discrepancies).