You are ignoring the author of this comment. Click to temporarily show the comment.

“Now it becomes clear that any claims against cheats cannot be based only on statistics but with additional elements of proof…”

What else is there but statistics? A single strange play will never be sufficient to convict a pair. What elements of proof do we need? Should we bug their hotel rooms in order to record cheating discussions?

Instead of requiring impossible elements of proof, perhaps cases should be heard by committees that (a) understand statistics, and (b) understand bridge.

You are ignoring the author of this comment. Click to temporarily show the comment.

Eric, the argument that other auctions also wrong-side notrump is not a good reason to wrong-side notrump here as well.

Of course, two things need to happen for 2NT to wrong-side notrump. First you should actually end up in notrump, which is certainly not a given after this start of the auction. Secondly, it needs to be important to play notrump from responder's side. Given that opener shows extras I think this will also not occur very often.

If a more artificial structure allows the partnership to show extra values below 3NT (while still showing their shape), then I think that is worth an occasional wrong-siding; more so in a standard setting than in precision.

In my partnership we currently play that 1S-2H-2S is a catch all, including most minimum hands without support, and that 1S-2H-2NT shows 3+ hearts, any strength. The idea is that this way we do not wrong-side notrump (as we will usually be playing hearts), we have a more room to investigate a heart slam, and that it allows us to use other bids (including 1S-2H-3H and 1S-2H-2S-2NT-3H) for different purposes.

I haven't made up my mind yet about whether I like what we do, and I think that 1S - 2H remains a very difficult auction.

You are ignoring the author of this comment. Click to temporarily show the comment.

I was referring to the proposal of Lessard, which I understood as 2S promising a minimum. I don't think that he proposes to keep the meaning of all other bids natural.

You are ignoring the author of this comment. Click to temporarily show the comment.

The fact that I also wrote that 5332 opposite 4333 with a 54 fit favors 3NT could have made it less surprising for you Michael.

As for 4333 opposite 5332, I did use more restrictions on my simulations, tying to take into account that second hand passed, etc. This could easily affect the outcome, given how close it is.

High card range also has an effect, though not the way Paul imagines it. I found that the weakest range often favors the 9-trick game, if I recall correctly.

Taking into account the lead also makes a quick auction to 3NT more desirable.

You are ignoring the author of this comment. Click to temporarily show the comment.

I am not sure I am convinced by your arguments Steve. It is not said exactly how strong 2H was. If the agreement is 6-10, the west hand is absolutely maximal. On the auction partner could have been considerably stronger.

Switch both red jacks with the corresponding kings, giving all hands the same number of hcp, and it also looks better for us. I think it is quite difficult to estimate the odds of 2H going down.

You are ignoring the author of this comment. Click to temporarily show the comment.

I spent some time investigating whether 3NT or 4M is better in these situation, and if I recall correctly found that 3NT is better at both pairs and imps. In fact, with 5332 opposite 4333 it is also better to play 3NT, even when you have a 5-4 fit.

The only form of scoring in which I was not sure is the variant of patton where a difference of 10 points doesn't count.

You are ignoring the author of this comment. Click to temporarily show the comment.

David, allow me to think out loud here, even though that usually ends up with me looking very stupid.

Suppose as you say that east drops a random card on the floor, and it happens to be a non-diamond. If east had 3 diamonds the chance of it being a non-diamond would have been 10/13, while if east had 0 diamonds the chance of it being a non-diamond would have been 13/13. Therefore (by Bayes Theorem if you like) the chance that east had 3 diamonds is 10/13 as small as the chance that east has 0 diamonds.

Now suppose that from the lead we can deduce with absolute certainty that east has the spade king. Also assume that east would never lead a diamond from Qxx (or from a void). Under these assumptions we can apply the principle of vacant spaces. The chance that east holds all three diamonds is 12/25 times 11/24 times 10/23, while the chance that west holds all three diamonds is 13/25 times 12/24 times 11/23. Luckily almost everything cancels, and the ratio of these two chances is also 10/13.

Therefore, if we assume that diamonds split 3-0 or 0-3, then the chance of east holding all three diamonds after a random card falls face-up on the ground is 10/23 (if the card is not a diamond), which is exactly the same as what the principle of vacant spaces would give.

You are ignoring the author of this comment. Click to temporarily show the comment.

Certainly the given hand seems to suit your methods very well. I would be happy to bid 2S showing constructive values with both minors. It will not only work well when partner has a balanced hand with 4-5 diamonds, but will be especially effective when partner is unbalanced.

In general I'd say that reserving two bids for hands with both minors should do well when you pick up a hand with both minors.

Given the “standard” methods that Arend was playing I still think that 2C is better than double or pass or any other alternative.

You are ignoring the author of this comment. Click to temporarily show the comment.

Although your picture keeps smiling, your tone seems a little agitated. Maybe I argue a little too fanatically, but that is not because I think that your suggested treatment is poor, not at all. I merely thought that you stated your case too strongly.

I didn't compute the odds just to prove you right or wrong, although such behavior would not be beneath me. The unbalanced 1D opening is quite common in the Netherlands, and I am genuinely interested in how to best deal with competitive auctions such as this one.

You are of course correct that by eliminating the minimal unbalanced hands, as Phil and you do, you will have a balanced hand considerably often.

And I can't help but repeating that after 1C - (1S) opener will hold a balanced hand only about 60% of the time, not 80%.

Han Peters

What else is there but statistics? A single strange play will never be sufficient to convict a pair. What elements of proof do we need? Should we bug their hotel rooms in order to record cheating discussions?

Instead of requiring impossible elements of proof, perhaps cases should be heard by committees that (a) understand statistics, and (b) understand bridge.

Han Peters

Han Peters

Han Peters

Han Peters

Han Peters

Of course, two things need to happen for 2NT to wrong-side notrump. First you should actually end up in notrump, which is certainly not a given after this start of the auction. Secondly, it needs to be important to play notrump from responder's side. Given that opener shows extras I think this will also not occur very often.

If a more artificial structure allows the partnership to show extra values below 3NT (while still showing their shape), then I think that is worth an occasional wrong-siding; more so in a standard setting than in precision.

In my partnership we currently play that 1S-2H-2S is a catch all, including most minimum hands without support, and that 1S-2H-2NT shows 3+ hearts, any strength. The idea is that this way we do not wrong-side notrump (as we will usually be playing hearts), we have a more room to investigate a heart slam, and that it allows us to use other bids (including 1S-2H-3H and 1S-2H-2S-2NT-3H) for different purposes.

I haven't made up my mind yet about whether I like what we do, and I think that 1S - 2H remains a very difficult auction.

Han Peters

Han Peters

Han Peters

Han Peters

As for 4333 opposite 5332, I did use more restrictions on my simulations, tying to take into account that second hand passed, etc. This could easily affect the outcome, given how close it is.

High card range also has an effect, though not the way Paul imagines it. I found that the weakest range often favors the 9-trick game, if I recall correctly.

Taking into account the lead also makes a quick auction to 3NT more desirable.

Han Peters

Han Peters

Switch both red jacks with the corresponding kings, giving all hands the same number of hcp, and it also looks better for us. I think it is quite difficult to estimate the odds of 2H going down.

Han Peters

The only form of scoring in which I was not sure is the variant of patton where a difference of 10 points doesn't count.

Han Peters

Han Peters

Suppose as you say that east drops a random card on the floor, and it happens to be a non-diamond. If east had 3 diamonds the chance of it being a non-diamond would have been 10/13, while if east had 0 diamonds the chance of it being a non-diamond would have been 13/13. Therefore (by Bayes Theorem if you like) the chance that east had 3 diamonds is 10/13 as small as the chance that east has 0 diamonds.

Now suppose that from the lead we can deduce with absolute certainty that east has the spade king. Also assume that east would never lead a diamond from Qxx (or from a void). Under these assumptions we can apply the principle of vacant spaces. The chance that east holds all three diamonds is 12/25 times 11/24 times 10/23, while the chance that west holds all three diamonds is 13/25 times 12/24 times 11/23. Luckily almost everything cancels, and the ratio of these two chances is also 10/13.

Therefore, if we assume that diamonds split 3-0 or 0-3, then the chance of east holding all three diamonds after a random card falls face-up on the ground is 10/23 (if the card is not a diamond), which is exactly the same as what the principle of vacant spaces would give.

Am I correct or did I made a blooper?

Han Peters

In general I'd say that reserving two bids for hands with both minors should do well when you pick up a hand with both minors.

Given the “standard” methods that Arend was playing I still think that 2C is better than double or pass or any other alternative.

Han Peters

I didn't compute the odds just to prove you right or wrong, although such behavior would not be beneath me. The unbalanced 1D opening is quite common in the Netherlands, and I am genuinely interested in how to best deal with competitive auctions such as this one.

You are of course correct that by eliminating the minimal unbalanced hands, as Phil and you do, you will have a balanced hand considerably often.

And I can't help but repeating that after 1C - (1S) opener will hold a balanced hand only about 60% of the time, not 80%.

Han Peters

Han Peters

Han Peters