Join Bridge Winners
All comments by Kit Woolsey
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve,

You have spelled out the problem very accurately.

Playing it your way, where returning low is either suit-preference low or ambiguous, it seems like partner may have a problem when low is returned. Let's suppose the suits are clubs and diamonds, and the opening leader has the following:

A dangerous club holding to lead from, one which would be most profitable if partner really wanted a club lead but costly if partner had no preference between the two suits.

A relatively safe diamond holding to lead from.

What does he do? If he leads a club, he may find that partner was giving the ambiguous signal and the club lead costs a trick. If he leads a diamond, he may find that partner really did want a club shift, and failure to do so costs a trick.

Perhaps you are saying: Ah, but if this were the case partner would work out that a club shift was disastrous, so he would give suit-preference high even if he didn't particularly want a diamond shift. And so he should. But this means that you aren't really doing anything different from what I am doing. You would be saying that if partner has equally poor holdings in the suits, he will give suit-preference for the suit which he judges will be doing the least damage. As would I.

If you want to say that if you really have no preference whatsoever (i.e. you can't make any determination about which shift is likely to be less damaging) then instead of flipping a mental coin you lead low, I have no problems with that. But in my experience this is a very rare situation. Third hand might get it wrong with his return, but from his point of view he virtually always has some preference. His suit-preference return is not a command. It is simply an expression of which return looks best from his point of view. It is the job of his partner to synthesize this information with his own hand and dummy and come up with the best defense.
March 5, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Stefan's line may well be best. However, it requires West to have the doubleton heart. If West has the third heart, Stefan's line has to fail since West can afford to play low on the spade – East wins and gives West a club ruff. And if West has 3 hearts and East's singleton spade is the jack, Stefan's line of play snatches defeat from the jaws of victory.

Understanding the existence of Stefan's line should have solved the defender's problem on the actual hand. Suppose declarer had 3 small spades. Declarer would think that East's singleton spade is the king due to the non-spade lead. If that is the case, drawing trumps and leading the low spade can't work, since with AQJ10xx of spades it would be trivial for West to crocodile since that would be his only chance. Therefore, declarer would definitely adopt Stefan's line of drawing 2 trumps and leading a spade, which succeeds when West has a doubleton heart.

Since declarer didn't do this, he is marked with the queen of spades. Hence, West should not have crocodiled.
March 5, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Okay, I think I finally have this thing figured out.

Presumably West doesn't have AK of hearts or KQ of either minor considering his lead, and East doesn't have two honors in any suit or he would have opened the bidding. So, the honors are probably divided in all 3 suits.

Run 6 trump tricks, keeping 3 clubs, 3 diamonds, and the jack of hearts in dummy. Both opponents have to hold 3 diamonds, else you can pick up the suit for 1 loser. One opponent has to hold 3 clubs, else you can ruff out the clubs for your tenth trick. Hence, one opponent has to come down to a stiff heart honor.

Now, cash the club ace, and lead a heart. The person with the stiff honor wins, and can't safely break the diamond suit. So, he must return a club which you ruff. You are now down to a 4-card ending, with each opponent having 3 diamonds, one opponent a high club, and the other opponent a high heart.

If East has the high heart, exit with your queen of hearts. East will be forced to break the diamond suit and give you your tenth trick.

If East has the high club, lead the jack of diamonds. West must cover. You win the ace, and end-play East with dummy's last club.

Note that this works regardless of which opponent has the 9 of diamonds. In fact, the 8 of diamonds is a red herring.

Of course at the table you would have to read the distribution, which might not be too easy. The diamonds might have been 4-2 all along, in which case you could make just by going right in the diamond suit. But if you know the enemy hands, I think the contract can always be made.
March 4, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Michael – On your AKQxxx example, where you want to show the queen so have only 4 possible orders, my plays would mean:

QKA – strong low preference – playing lowest card both times
QAK – mild low preference – lowest card first has priority
AQK – strong high preference – highest possible card all times given the conditions
KQA – mild high preference – since didn't start with queen

You ask how can having an I don't care default hurt? Suppose you have 95 left, and 5 means either no preference or SP low as you play. Suppose you have a mild preference for the higher suit. If you play the 5, partner may think you have a clear preference for the lower suit and do the wrong thing. However, if you play the 9 partner may really believe it (thinking that if you didn't have much of a preference you would have played your potential default card) and do the wrong thing.

I'm not saying my approach is necessarily better. If I truly don't have a preference I have to decide which lie I think will be least damaging, and if I decide wrongly it won't be good. I'm still more comfortable with this approach, since partner will always know what I'm trying to say.
March 4, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Michael and I have somewhat different signaling philosophies. I do not believe in giving murky signals unless I am sure that partner will know (or find out in time) what I am doing. If I don't have a preference, I give what I believe will be the least damaging signal.

In my AK72 example, I do truly believe that there is no “neutral” signal – king, ace, 7 in my mind is mild suit-preference for the lower suit. The fact that the 7 is the “normal” play does not matter to me. I do agree with his comment on the ethical implications of a quick king. This is just another illustration of the importance of thinking before you play, and third hand always pausing at trick 1.

A more common example is as follows: In notrump, declarer leads up to a possibly entryless KQJ10x in dummy. You hold 952. Playing standard signals, obviously you play the 2. Declarer wins, and continues the suit. Partner wins. You have the 95 left. What do your plays mean?

Obviously this is a suit-preference situation – everybody would agree with that. But what do you do if you don't have a preference? Many players play that the 9, the “normal” card, might be either SP high or neutral, while the 5 is definitely SP low. Michael's comments indicate that he may go along with this philosophy. I do not believe there is a neutral play. For me, 9 is SP high, 5 is SP low, period. If you don't have a preference, too bad – you find a preference.

I'm not saying that one approach is better or worse than the other. But it is important to realize that these are different approaches.

March 4, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Bob – It is true that if North knew declarer was ruffing the opening lead his play would be suit-preference. But North doesn't know that. South might have 4 clubs and 5 diamonds.

It is true that if declarer were known to be short in clubs that North's play would be suit-preference. But that isn't known. South could have 4 clubs, and North might have been forced to bid a 3-card suit. Therefore, from South's point of view a second club might be cashing, and it might be necessary to cash it.

This is why with 5432 of clubs North's play cannot be suit-preference.
March 4, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Chris – Of course you should never embark on a sequence which might lead to a misunderstanding from which you can't recover. That is why I said that the approach to choose depends on partnership agreements (or non-agreements). Thus, as you say, if there is any danger that partner would pass 4NT that is a bid you do not make.

Still, isn't it better to probe with 3? If he bids 4, you bid 4NT RKC, and he bids 5 showing 1 or 4, you are allowed to pass rather than getting to 6 off two of three critical cards. It might not be a thing of beauty, but perhaps it is your best chance for a plus score with that information. Or you can bid 5 and hope he passes. If he doesn't, worst case is you get to the same 6 you would get to by leaping there – 6 he will definitely pass.

Even if you take the approach of bidding RKC should he bid 4 but leaping to 6 if he bids anything else, you are still as well off as if you bid 6 immediately with perhaps a chance to improve and no downside.

My point is that there is always some approach which is as good or better than leaping to slam. It doesn't have to work, but it can't cost. 6 will always be there. It won't run away from you.
March 4, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
I agree with the evaluation that 6 is probably the best contract from South's point of view, and if South were allowed just one call 6 would be the best bet. However, I strongly disapprove of the leap to 6. What's the rush? 6 isn't going to run away from you. As long as you don't make a non-forcing call partner will keep bidding, and the moment you mouth the words 6 the auction will come to a grinding halt. So why not try and find out something about partner's hand.

What might you do? How about simple Blackwood. Partner figures to have at least one ace, but if he happened to make a light takeout double on J10xx KQJxx KJ10x at least you will avoid a stupid slam off two cashing aces. If he shows 2 aces, you can ask about his kings. If he shows 2 kings, wouldn't you be willing to take your chances on a grand, hoping he has a singleton spade or enough winners to dispose of spade losers. You have nothing to lose. 6 will always be there.

Another possibility might be to bid 3. If partner bids 4 you can bid RKC, and if he shows 3 keycards you might shoot out a grand while if he shows 1 keycard you can stop at the 5-level. If he bids 4 you can bid RKC, and then if he shows 2 keycards follow with an ask for specific kings so you will locate the king of clubs for grand purposes. If he bids 4, you can't do much but bid 6.

The exact approach to take isn't clear, and would depend a lot on partnership agreements. However, any kind of exploratory approach has to be better than directly bidding 6. Leaping to a slam is almost always wrong if there are any feasible alternatives.
March 3, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Brian – When the kibitzers 5 tables away and the janitor know it is suit-preference. That's how obvious it has to be.
March 1, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Bob – I strongly disagree with your concept about “giving partner the signal you think he needs”, unless you are 100% sure that he will interpret it correctly. I have seen many defensive disasters, including at the expert level, where one partner thought the signal didn't have the default meaning while the other partner thought it did. The deal under discussion is a perfect example.

On the actual deal, I think we all agree that Eric made a very good play of the queen of clubs, and a very bad play of the 9 of diamonds. Perhaps Geoff should have gotten it right anyway, but he certainly would have gotten it right had Eric played the 6 of diamonds. Eric was trying to give his partner the signal he thought his partner needed, but it wasn't interpreted that way.

On the actual deal there was no reason for Eric to deviate, since the queen of clubs play was loud and clear. But suppose Eric's clubs had been 5432. Now an alarm clock play would not have been available. His play at trick 1 would be either count or attitude depending upon partnership agreements, but it definitely would not have any suit-preference connotations. In this scenario, what diamond should Eric play at trick 2?

I strongly believe he should play the 6, giving the default count signal. He knows he wants a spade shift. He would like to give a suit-preference signal. But he doesn't know that his partner knows that, or that his partner will interpret this as a “what he needs to know” situation. On this hand the diamond count may be all that Geoff needs to know to find the right defense.

Some pairs may play that suit-preference rather than count is the default signal on suits declarer leads, and they give count only when it is obviously necessary (such as telling partner when to win his ace with an entryless dummy). That agreement is fine, and would have worked well on this deal. But unless that is the partnership agreement, count is the default signal and that is what Eric's hand should show.
March 1, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Thanks for the input Geoff. It is especially valuable to the readers when the player describes what his thinking process was.

On the issue of conflicting information, I believe the partnership agreement should be that the first signal has priority. For example, against a spade contract West leads a doubleton heart and hits East with AK72. East cashes his top hearts and gives West a ruff. East has two chances to give a suit preference signal – his order of heart honor plays, and the low heart he leads on the third round. I think his carding should mean:

ace, king, 7 – strong preference for diamonds
ace, king, 2 – mild preference for diamonds
king, ace, 7 – mild preference for clubs
king, ace, 2 – strong preference for clubs
Feb. 28, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Greco's queen of clubs play was quite imaginative. From his point of view Hampson's shape could be 2-2-5-4, in which case only a spade shift will defeat the contract. If Hampson does have 4 clubs he will know that a second club isn't cashing. Greco couldn't afford to play the queen of clubs from Qxx, since if Hampson had AKxx he would assume his partner started with QJx and underlead. Thus, Greco was telling Hampson both that a second round of clubs wouldn't cash and that he is interested in something other than a club continuation.

Unfortunately, Greco threw it all away with his 9 of diamonds play. It is a fundamental principle of defensive carding that a signal has its default meaning (attitude on partner's lead, count on declarer's lead) unless it is 100% clear that the attitude or count of the suit is known, about to be known, or clearly of no value. Only then can a signal be suit-preference instead of the default meaning. The diamond count was not known, and it could be of value to Hampson, therefore the signal must be count, not suit-preference.

Should Hampson have worked it out anyway? Perhaps. It would be pretty weird for declarer to be attacking diamonds with xxx in diamonds. Perhaps he thought declarer had something like Qxx A109xxxx xxx – and would misguess the hearts from the biding unless he found out the spade count. In practice I'm sure he simply believed his partner's count card and his thinking stopped there – he knew there was no discard coming.

It is possible that the offbeat 1NT opening contributed to the disaster. After trick 1, Greco was sure that his partner's shape was 2-2-4-5, making declarer's shape 4-7-2-0. Thus Greco thought he knew that a diamond discard wouldn't matter, so he went out of his way to signal for a spade shift which would certainly defeat the contract. Of course if that were the hand declarer would have drawn trumps first, but it is difficult to think of everything.
Feb. 28, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve – Strength isn't important. If 2nd seat bids opener finds out immediately roughly how strong responder is, and opener can take it from there. If you get to a 23 point 3NT or miss a 25 point 3NT, it isn't a big deal. Sometimes the 23 point 3NT will make. Sometimes the 25 point 3NT won't. While of course you can lose a 6 or 10 IMP swing if you are wrong, you can also gain the 6 or 10 IMP swing by being “wrong”. The average swing from these decisions tends to be quite small, even though the swing itself may be large. The average swing from an accurate slam auction is likely to be much greater.

Heavy competition can do plenty of damage to a strong 1, but usually only when opener has an unbalanced hand. When opener's hand is balanced, competition is relatively easy to deal with.
Feb. 27, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Henry – It is definitely worth having 1 as natural (and forcing). Responder gets to show a spade suit if he has one, heart support if he has it, as well as whether he is minimum (0-5) or non-minimum (6-8) for the 1 response. These gains easily outweigh the loss of getting to 2NT on the infrequent 20-21 hands.

It is fundamental for good constructive bidding that the strong hand should be captain, with the weak hand describing. If the 1 bid is natural, this can happen. If the 1 bid is artificial the weak hand can't make a descriptive call, so the weak hand is forced into the captain's chair.
Feb. 27, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
I don't consider the 2NT opening showing both minors particularly valuable. The only reason we play it is that we prefer to open strong balanced hands 1, since slam auctions (when responder has a positive response) are considerably more accurate than after the bulky 2NT opening where the weak hand has to take control. However, we have gotten unusually good results when it has come up, although the sample size is too small to signify anything.

You may have gotten your defense against 2NT showing the minors from us, but we no longer play anything like that. We find that having 3 takeout for the majors is sufficient, since partner can bid 3 if he needs to know which is our better major. We use 3 to show a light overcall in one of the majors, with the obvious pass or correct responses. A direct 3M overcall shows a sound overcall. This strength distinction with the overcall we consider to be the most important information responder needs. We use this structure against any artificial 2NT or 3NT opener which shows 1 or both minors.
Feb. 27, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Certainly that is the way to start. If nothing exciting happens in diamonds, you will need the queen of clubs on and either the clubs coming in directly or a club-diamond squeeze. You need to ruff the second heart along the way for the squeeze, of course.

But suppose something exciting does happen in diamonds. There are 4 exciting things which might happen, and each of these may lead to a different approach in the club suit. What are all these?
Feb. 25, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Is 7 easier to play than 6? How exactly should one play in 7? Be very specific.
Feb. 25, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Nice analysis Jonathan. I like the relevant holdings approach. In addition to making the mathematical computations a lot easier or unnecessary, that approach will often get you away from a clearly inferior line of play or defense. In order to justify any play or defense, you should have at least one layout in mind where your play wins and play B loses.

At the table I would imagine that almost all experts would choose line A. It is easy to find, and easy to see that it succeeds if one of 3 cards is right. While line B is just about as good, seeing that is more complex.
Feb. 25, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
A limited hand cannot logically suddenly have a game force when partner might be very weak. When Hamman chose to rebid 1 rather than 2, for better or worse, he made the decision to limit his hand. That limitation does not change. Sure, Hamman expects Zia will probably bid over 3, and Hamman will be disappointed if Zia passes. But that is the chance Hamman took when he rebid 1. He limited his hand, and he can't change that.
Feb. 24, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
I'm confused about this discussion of 3. Either Hamman-Zia play a Walsh-style responding structure or they don't (I would bet that they don't). Are they permitted to respond 1 on a hand such as Jxxx xx KQ109x xx or not?

If Zia cannot have such a hand, then the 1 call must be at least invitational. In that case, obviously 3 is forcing.

If Zia can have such a hand, why should 3 be forcing? Hamman's hand is limited by his failure to jump shift. 3 clearly shows some 4-4-1-4 or 4-4-0-5 hand in the 17-19 range. If Zia doesn't think there is a game opposite that, which he wouldn't with my example hand, why shouldn't he be free to pass? He know what his partner has.
Feb. 24, 2011
.

Bottom Home Top