[MFA]

In this situation, I play:

Low card = Continue please (no shift desired)
High even card = Shift to lowest ranking side suit
High odd card = Shift to highest ranking side suit

If I have, say, only odd cards among the big ones, the highest odd card would have it's usual meaning (shift to highest ranking suit), while a smaller odd card would indicate the lower suit.

[ArtK78]

Fascinating discussion.

When I first saw this problem posed, I thought that this was the easiest question that had ever been posed in this Forum. And the vote and the discussion, for the most part, confirmed this. A clear suit preference situation.

Lexlogan's assertion that the first signal is attitude even if the defense cannot win the next round of the suit is interesting. While I do not have a copy of the Bridge Encyclopedia or Kantar's book on defense in front of me, it is my recollection that the situation posed in both of those reference works is a singleton in dummy, not Kx after the lead of the Ace (please correct me if I am wrong on this). This is a much different situation.

However, even if it is Kx in dummy, it is possible to construct a hand where the defense needs to continue, win a trick and force the dummy to ruff (or force a ruff in one hand or the other, in the case that the declaring side is 2-2 in the suit). So, attitude is not out of the question.

The question really is, what card requests a continuation? I have always thought that an unnecessarily high card (the J in this case) is suit preference for the higher side suit, and the lowest card is a suit preference for the lower side suit, while an intermediate card expresses no opinion and tends to ask for a continuation (by the way, I fail to see how the presence or absence of the 10 is relevant to the meaning of the play of the J or the willingness to play it, absent some squeeze possibility against partner). Playing Obvious Shift, the situation might be different, depending on the appearance of the dummy. And, of course, if a shift to a heart on this hand is irrational (ex. - dummy has AK tight), then perhaps the play of the Jack has some other meaning (an alarm clock play, to use the lexicon of Obvious Shift).

Phil says that the hearts in dummy were Txxxx. I would like to know what the rest of the dummy was before coming to a definite conclusion about the meaning of the play of the ♠J, but at least the heart holding in dummy is consistent with a desire of my partner for a heart switch.

Without the entire dummy to help guide us, and absent some agreement to the contrary, the play of the Jack appears to be a suit preference signal.

[helene_t]

You should have made distinct options for "suit preference" and "reverse suit preference" to get some controversy.

[jdonn]

lexlogan, on Feb 20 2008, 10:43 PM, said:

bid_em_up, on Feb 15 2008, 07:38 PM, said:

Given that you cannot take another trick in spades, it is suit preference.

You never lead a suit in which you cannot take another trick? Declarers must love all those free gifts -- oh, wait, I don't have to ask, I get those gifts all the time. Hasn't anyone on this forum ever heard of passive defense?

I don't understand why you can't seperate that the signal is suit preference from the fact that opening leader is allowed to continue the lead anyway!!! He doesn't have to turn off his brain.

[lexlogan]

jdonn, on Feb 21 2008, 10:00 PM, said:

lexlogan, on Feb 20 2008, 10:43 PM, said:

bid_em_up, on Feb 15 2008, 07:38 PM, said:

Given that you cannot take another trick in spades, it is suit preference.

You never lead a suit in which you cannot take another trick? Declarers must love all those free gifts -- oh, wait, I don't have to ask, I get those gifts all the time. Hasn't anyone on this forum ever heard of passive defense?

I don't understand why you can't seperate that the signal is suit preference from the fact that opening leader is allowed to continue the lead anyway!!! He doesn't have to turn off his brain.

In the problem posed, opening leader holds AQ of the suit led. For all third hand knows, leader does not have the Queen. When that situation occurs, it is often important for third hand to signal whether he has the Queen. This would make continuing the suit relatively safe (passive defense). Continuing the suit with the King in dummy and an original holding of Qxx or better in declarer's hand could be disastrous. So, no, opening leader does not have to react mechanically to partner's puppet string signal, but he will make better decisions given better information.

Depending on the auction, dummy, and leader's hand, third hand is often fairly weak. If he holds a high honor in the suit led, he is less likely to have a useful holding in either side suit. Switching suits costs, on average, half a trick. Knowing that partner's hand is more suitable for a continuation than a switch can be crucial for the opening leader. There is no practical way to transmit this information if you assume suit preference signals here.

Since the proper goal of the defense on many (perhaps most) hands is to NOT BLOW A TRICK, waiting for what declarer must yield if you don't help him out, attitude is still important even if you can't win the next trick in the suit. Therefore, Eddie Kantar and I do not think the benefits of suit preference outwieghs attitude for the case where dummy has a singleton. I haven't seen him address this case, and perhaps it is different enough. Also, perhaps most experts think suit preference or count is more useful in such cases. But I have yet to see any justification of that. All the comments are along the lines of "of course, suit preference."

Here's a hand from the Suit Preference article in the Bridge Encyclopedia. Does attitude apply here only because suit preference makes no sense? (West, after leading a high diamond, must continue the suit to protect East's King of trumps.) The Encyclopedia merely says "the suit played is significant, so suit preference does not apply; East's Queen is a standard come-on." This sounds much more like attitude is the default, even with a singleton in dummy.

6♣ by South

---

[Echognome]

The example just given is pretty straightforward. You play a middle card = no preference. Thus West can figure out East cannot get a ruff. So the only hope is a trump trick and the continuation is clear.

[jdonn]

Echognome, on Feb 21 2008, 05:37 PM, said:

The example just given is pretty straightforward. You play a middle card = no preference. Thus West can figure out East cannot get a ruff. So the only hope is a trump trick and the continuation is clear.

The example is a perfect one to demonstrate why suit preference is right. Imagine being void in a major, discouraging diamonds without being able to ask for a particular major, then sitting there as partner is in the tank... As you say, if you are not ruffing a major you play a middle card, and in fact with his actual hand and the actual dummy partner will continue diamonds 100% of the time as anything else is hopeless, which reaffirms that attitude is pointless since partner can use his brain anyway.

[lexlogan]

Re-reading the Bridge Encyclopedia article, the next hand does feature Kx in dummy. North-South have bid to 6♣ off two Aces; West cashes the heart Ace and must switch to diamonds despite KQ10x in dummy else declarer (with two singletons) dumps his diamond loser on the ♥K. Suit preference (or an unlikely guess) is the only way to keep them from stealing the slam. I'm not convinced -- surely, it is more likely that switching could pickle partner's Jack than that he holds a second Ace (or can ruff) AND declarer has two singletons. I'll let those who believe they and partner can simultaneously determine when suit preference is "obvious" collect their good boards on such rarities. I strongly suspect they blow far more matchpoints or IMPs with untimely switches than they gain from saving partner a guess about which suit to switch to.

All in all, the Encyclopedia article is very unsatisfying since the final example, a supposed triumph for suit preference, is the sort of hand where logic seems to dictate the only possible defense and no signal from partner ought be needed.

[jdonn]

lexlogan, on Feb 21 2008, 05:54 PM, said:

All in all, the Encyclopedia article is very unsatisfying since the final example, a supposed triumph for suit preference, is the sort of hand where logic seems to dictate the only possible defense and no signal from partner ought be needed.

Just like the example that was a supposed triumph for attitude. So I think all we can really conclude is they don't pick good examples.

[Foxx]

A couple of people have hinted at it, but this question cannot be answered absolutely without knowing the key detail - - - - how many trumps came down in the dummy.

Of course, if it was any respectable number (like, more than two) then I would expect suit preference applies, similar to the most well known case where a singleton is in dummy and opening leader's high honor holds the trick.