Join Bridge Winners
All comments by Wayne Burrows
You are ignoring the author of this comment. Click to temporarily show the comment.
There are alternatives to Muppet that circumvent this problem. I was talking with my partner about the merits of changing to Muppet after seeing several pairs playing those methods in the recent world championships.

What we play is over 3:

3 promises three or four spades or four hearts.
3/ promises a five card major.
3NT shows at most two spades and at most three hearts.

Over the ambiguous 3 rebid we then play:

3 asks for four spades (but does not promise four spades).
3 promises four hearts
3NT promises both majors.

So 3 is used both to find a spade fit and to play 3NT. Opener bids 3 with spades and 3NT without. Responder then either raises spades if appropriate or corrects to 3NT (or passes 3NT).

… 3
3 3
3NT = I want to play 3NT.

This means that at each juncture responder can depart from the major enquiry and make a minor suit slam try:

… 3
4m = slam try in the minor (and I was interested in a five-card major)

… 3M
4m = slam try in the minor (I was interested in the other major)

… 3
3 3NT
4m = slam try in the minor (with four spades since otherwise you would not ask)

… 3
3 3
4m = cue for spades otherwise you would not have asked.

… 3
3 3NT
4m = slam try in the minor.

If in addition you play non accepts as transfers to show five cards in the other major you can find a 4=4 or 5=3 major fit whenever at least one exists with these methods at the cost of some information leakage and not being able to play a transfer and pass as a bail out.

2NT 3
3 = five spades and only two hearts

2NT 3
3NT = five hearts and only two spades.
Oct. 31
You are ignoring the author of this comment. Click to temporarily show the comment.
Playing double dummy over 1000 boards bidding on with 4D and then playing 4NT if partner had two hearts and playing 6D if partner had three or more key cards and playing 5D if partner hand 0-2 key cards I obtained the following scores on average (per board) compared with passing 3NT:

+3.886 IMPs or
64.4% at MPs.

The real numbers will undoubtedly be worse but how much worse I do not know. I suspect these margins are wide enough that it is clear to bid on whatever the scoring.

Here are the IMP frequencies (positive is 3NT better than trying for 6D - the reverse of the above numbers)

-15 1
-14 69
-13 0
-12 0
-11 6
-10 454
-9 111
-8 0
-7 0
-6 0
-5 0
-4 0
-3 0
-2 0
-1 0
0 3
1 0
2 97
3 33
4 5
5 0
6 0
7 1
8 0
9 0
10 42
11 176
12 2
Oct. 30
You are ignoring the author of this comment. Click to temporarily show the comment.
True if as is most common the simulations use double dummy analysis.

For more accurate analysis it is possible to count the card or at least some of them where there is a double dummy bias and discount the results accordingly.

For some problems you can also do single dummy simulations using some computer bridge program to play the cards. That eliminated double dummy issues but does not eliminate all problems.

You can also count obvious making layouts rather than do double dummy as a lower bound with a bit more coding.
Oct. 29
You are ignoring the author of this comment. Click to temporarily show the comment.
That seems to be a question for the partnership.

I open that hand 1c most of the time.
Oct. 29
You are ignoring the author of this comment. Click to temporarily show the comment.
I am surprised that there are that many people in New Zealand playing Ghestem.
Oct. 29
You are ignoring the author of this comment. Click to temporarily show the comment.
so it should say >=3 I will check my numbers

Thanks
Oct. 29
You are ignoring the author of this comment. Click to temporarily show the comment.
I did a simulation with the following constraints for partner:

hcp(north)>=22 and hcp(north)<=24 and
shape(north, any 4333 + any 4432 + any 5332 + any 5422) and
spades(north)<4 and
hearts(north)<4 and
(aces(north)>=3 or (aces(north)>=2 and hascard(north,KD))) and
diamonds(north)>=3

That is I assumed partner will cooperate always with three diamonds and not with two diamonds and that partner showed the required number of key cards for slam.

The result was that double dummy the following numbers of tricks were available with frequencies out of 1000 hands to the right.

7 1
8 1
9 2
10 37
11 211
12 560
13 188

Edited and updated numbers thanks Craig
Oct. 29
Wayne Burrows edited this comment Oct. 29
You are ignoring the author of this comment. Click to temporarily show the comment.
So called convention disruption will normally be bad for the side making the offending call. You have to live with the occasional bad result.

If I am honest most of my good scores come from the opponent's mistakes. I don't want to have rules that do not let them make mistakes.

We do though need the rules to be enforced when they subsequent to a mistake use UI to get out of jail.
Oct. 28
You are ignoring the author of this comment. Click to temporarily show the comment.
It is impossible to tell if the players on here are peers - I don't know them all. Some of the names in the poll of the 76% that I know I would consider ball-park peers of the player.

The player in question was a competent player ranked Silver Grand Master in New Zealand which until recently was our second highest ranking - there are now two new higher rankings.

I do not know who was polled at the tournament but I imagine some would be of lower rank.

I seriously do not think that fewer than the small number (of peers) required to determine a logical alternative would have chosen.

Clarification: The insanity I referred to was intended to mean that the call was determined not to be a logical alternative.
Oct. 28
You are ignoring the author of this comment. Click to temporarily show the comment.
The ambiguity is that there is no objective way to determine a result after these unauthorised information situations. Moreover, the stakes are high or at least can be but with a very high variance. Sometimes the UI can have no effect and sometimes it can cost a bucket load of IMPs or MPs.

Even if we could take away the UI the results for different pairs making the same wrong bid can be wildly different. Here, for example, after 3H one pair might think and have the agreement that new suit by a passed hand implies a fit or tolerance for our clubs and another not. So one pair may be justified in bidding 4C and another not. Even for regular partnerships those system details might not be documented.

Then from 4C one pair might be able to pass and another treat 4C as a further cue. Although the pair that passes might have additional disclosure problems as it appears that 3C was not majors but a two way bid that should have been disclosed at the 3C bid.

And so we go on at every bid. Sometimes the auction will spiral completely out of control.

To realise what can happen without the UI when there is a bidding mistake, imagine an online practice where you do not have to explain and see what happens to the auction. I know that occasionally under those conditions the auction ends uncomfortably high or in completely the wrong contract. Those results seem to happen much less often at the table where there is more potential for UI (although occasionally you see them when there are screens and rarely without).

I have had at least two memorable examples at the table where our auction went uncomfortably to high levels. One fortuitously good and one extremely bad. The fortuitous one was soon after we adopted exclusion key card. My tired partner did not recognise my exclusion and that was compounded when our opponent asked and my partner's explanation was “I have absolutely no idea”. The auction complicated when the opponent doubled and my partner signed off in her suit. Systemically the signoff showed two with the queen (P0D1) but the UI told me that I had no idea what she actually had. I closed my eyes and bid slam as two with the queen was exactly what I needed and by a miracle of fate it was exactly what she had and the slam rolled home.

The second one with a different partner I made an obscure but clearly agreed splinter. Partner forgot and raised and then returned to the splinter suit when I bid on. I had no idea where the auction would end but judged it was likely to be outside the range of positive MPs so I bid Seven something and redoubled for -3400. MPs took the pressure off as you can't do worse than a zero so I could play around with bidding seven and redoubling. At IMPs it would at some point become, perhaps, a difficult judgement call what to bid as there would likely still be IMPs at stake.

I wish there was a more objective way to determine results with UI. The method of polling to determine UI situations, which is in vogue, is often hopeless in my view. After one recent poll, I was told by the chair of the committee (one of the very best players in New Zealand) that pass was not a logical alternative but the BW poll that I conducted has, as I write 76%, of respondents passing. So an opponent was allowed to act on UI in a situation that I think and the BW poll seems to concur with was completely insane.

In addition to adjustments I favour significant penalties under Law 73 for blatant uses of UI. Again, however some work needs to be done to make sure that such penalties are more objective. Subjective penalties inconsistently applied will only lead to understandable dissatisfaction.
Oct. 27
You are ignoring the author of this comment. Click to temporarily show the comment.
Peter, what do you consider equity when a pair has gone completely off the rails in their bidding?
Oct. 26
You are ignoring the author of this comment. Click to temporarily show the comment.
“I would have thought any strong club pair or above …”

I think this applies even to lesser pairs, for example, those that play some version of Standard American.

As the Hideous Hog put it on being told he didn't know Precision (strong club)

“True but I know Standard American and what better reason could I have for playing Precision?”
Oct. 22
You are ignoring the author of this comment. Click to temporarily show the comment.
If we didn't throw away so much information in this game the question would not be needed. A player could look up a database of hands played by his expected opponents and prepare to play against them.

Instead we do not record that information most of the time which allows players to vary their actions based on extraneous information and not be easily caught and for partnerships to not disclose accurately agreements and styles with no way of verifying the inaccuracies.
Oct. 22
You are ignoring the author of this comment. Click to temporarily show the comment.
“If ”what we consider an honor“ is anything other than any ace, king, queen, jack, or ten, we're wrong.”

What is defined as an honour in the laws of bridge is not necessarily synonymous with the usage of honour in the vernacular?

Both definitions can exist simultaneously.
Oct. 21
You are ignoring the author of this comment. Click to temporarily show the comment.
“as in even bidding over 3H”

This auction is much more nuanced than that. It can easily be wrong to bid over 3H for some pairs and wrong to pass 3H for others.

For example, one pair might have an understanding that 3H here shows or even is likely to show club support (Robson and Segal's fit non-jumps) and another have no such agreement. You can't treat them all the same. And you can't decide for any of them without investigating carefully their agreements.
Oct. 21
You are ignoring the author of this comment. Click to temporarily show the comment.
“Suppose your regular partner is on lead against the auction above, holding …”

1. Obviously you do not have to disclose to declarer what you know about partner's holdings that has been deduced from declarer's bidding and play and your own cards.

2. So I know what partner would lead from a particular holding but there might be thousands of holdings that could be relevant. How precisely do I disclose all of that information?

3. Declarer has additional information that I do not have - their own cards and inferences that they have been able to deduce from that information and the play and defence. How does declarer ask such a question without tipping off their own hand or misleading the defence? It seems wrong to ask about leads from holdings (in four suits) that include cards in your own hand and also wrong to only be able to ask about holdings that do not include cards in your own hand.

While I think such questions are legitimate it seems much less clear to me how precisely the question should be asked and how precisely it should be answered.
Oct. 21
You are ignoring the author of this comment. Click to temporarily show the comment.
It is a huge leap from an answer that causes declarer to misguess a suit to the defender has lied.
Oct. 20
You are ignoring the author of this comment. Click to temporarily show the comment.
I would think that most serious partnerships discuss lead tendencies. If not then those partnerships are not exploiting very strong inferences from knowing your partner's lead style.
Oct. 20
You are ignoring the author of this comment. Click to temporarily show the comment.
I was just being dismissive of those results as I wanted to concentrate on the play at this table.

At four of the five tables where 4 was the final contract, that is except the Chinese table where the auction was uncontested, north had made a takeout double and not bid after their partner had bid spades at the three level or below.

Not sure if that is enough to deduce for sure that spades are 2=2 but it points in that direction.
Oct. 20
You are ignoring the author of this comment. Click to temporarily show the comment.
Double shows extra values but not the extra distribution or enough strength required to bid at the five level or above.

It does not bar partner from bidding. It is encouraging if he has extra distribution or enough extra values to think there might be a slam.

It does not show a heart stack or indeed any particular holding in their suit.

Similarly, if we were weaker and passed then double from partner would show extra values with no clear direction. We would bid out with distribution and spade support or pass for example.
Oct. 19
.

Bottom Home Top