Some problems from the Santa Barbara Sectional

All at IMPs, short matches

1. Both vul, you hold

    AQ10x A7xx A109x Q

   CHO	You
   1	1
   2	2
   4	?

   a) What now? What's your plan?
   

   ---

   MICHAEL

   Pass. I can't tell what I need to know.

   BARRY

   I pass 4 mildly reluctantly but prepared to concede this is the broken egg one has to crack in order to make an omelet—OK take off marks for convoluted style!

   DAVIDC

   Okay, I bid 2. (For the record, 3 by me would have been ok, too.) [Agreed. --Jeff] Partner bids 4 which is game but not slam interest. Partner could have  Kx KQTx Jxxx Axx which is just about reasonable slam. Could 5 be too high? Sure could. I am going plus and passing.

   CHRIS

   Pass. I consider it ridiculous to bid again with such weak trumps. On a bad day, we could be down in four.

   DAN

   Blackwood and then bid a slam if partner shows a KC and the trump Q. [Except that 5 shows one, which means... --Jeff]

   DAVIDM

   I am going to bid 6 and give partner a choice of slams.

   ROBB

   4NT. I expect to get to 6. Too many holes to try for 7.

   JOEL

   I pass. I dare not risk the 5 level or a bad slam.

   MIKE

   5. Presumably this implies a club control, since partner didn't cuebid 3 or 4 clubs en route to 4. This doesn't show shape, but it does show the diamond ace. Blackwood is wrong, as it doesn't find out about the necessary elements for slam—partner's secondary red suit holdings. Parner should know now that K, KQ, KQ is gold. Second choice is pass. Opposite the wrong hand, we could be in trouble at the 5 level (KJx,(KQJx,xxxx),Kx)

   BOBBY

   I pass—our distributions on this hand are likely to be close, and that makes me dubious about prospects for 12 tricks.

   DAVIDW

   Pass, avoiding the quest for a magic hand. Partner bid only 2; he pretty much needs  Kx KQxx KQxx xxx for 6 to be good. Any wastage in clubs will doom the hand. Bad breaks could doom us at the 5 level if he has any of the many wrong hands.

   KENT

   4, key card. Tough auction. Partner shouldn't have the club ace. [Why not? Would he cue on the way with crappy hearts? --Jeff] If partner shows one key, I will check for the trump queen, and we'll be in either five or six hearts.

   ED

   Pass. Partner will usually need all cards working to make slam a reasonable contract and I think there will be more -100 and -200 in reaching 5 and 6 than there will be +1430s in 6. Obviously the uninformative jump to 4 destroys any real chance to find out if slam will be good. On this hand, a self-splinter of 4 over 2 might be the best bid since the one card I don't want to find in partner's hand when the dummy comes down in 6 is the K. I'd like to get there opposite  Kx KQxx KQxx xxx but not get past 4 opposite  xxx KQJx Qxxx AJ, or  KJx Qxxx Qxxx Ax.

   MARSHALL

   Pass. To have a good slam, I need perfect cards (K, KQ, KQ) and don't know how to find out. Maybe I should have bid 4 over 2, but that would have given up on our chances to bid 6 in case of a 4-3 heart fit although I'm not getting there this way either. [The problem isn't 6 then, but not reaching 3NT when there's only a 4-3 heart fit. 4 would definitely be a poor choice. 3 might be better, though. --Jeff]

   JJ

   I think some slam try is in order. Any bid except for 5 or 4NT emphasizes pard's looking at his holding and bidding aggressively with a control, which is exactly the wrong message. If I’m going to blast a slam, offering 6 is a good idea. That allows us to play 6 if pard has something like  x Kxxx KQxxx KJx. I think I’ll choose 6.

   JEFF AT THE TABLE

   4NT, then 6 when we had enough keys. Even at the table, I hated this choice; I think a direct 6 is much better.

   VOTES

   Action Votes Pass 9 4 1 4NT 3 5 1 6 2

   WINNING ACTION

   Pass, sort of. Partner has  Kx KQ108 KQ8xx xx. 6 is a fine contract. 6 is also a very good contract. Not this time. Hearts are 5-0 offside. 6 gets doubled and goes for 500. 6 just goes down two quietly.

   JEFF UPON REFLECTION

   Looks like a close call. 9 stop in game and 7 drive to slam. Upon lots of reflection, it seems that slam can't be much better than 50-50 on balance. Passing seems right. If we bid, I don't see anything that will allow partner to help judge at the 5-level (sorry, 5 doesn't do it, which is why I think we might have been better served bidding 3 a round ago...unless that wasn't forcing), I think shooting out 6 is best. Partner will pick a red slam based on his lower red suit honors, which is what we want. It'd be nice to stop lower when one of his red suits is truly awful, but I don't see how to do that anymore.

   Another possiblity is to bid 5. That has to imply that I am not worried about controls, probably just about trumps. That'll cover one of the problems. And we might even get to 6 after that if partner suggests playing there. Problem with that is we might play 5 down when 6 is cold.

   ---

   b) Do you play that 4 is ambiguous about slam or suggests lack of slam interest or neither? (Obviously, partner has few aces this time, but in general, should he cue bid if he's slammish, to cater to a slam try by responder?)

   ---

   MICHAEL

   I'll cue bid if I have a hand that wanted to make a limit raise but didn't.

   BARRY

   I believe other action than jumping to 4 should be made on no more than 15% of good hands. [Excluding counter-game tries and 3NT efforts.] Splinters (some), 4—with the appropriate hand, typically. Partner would bid 4 with e.g.  Kx Jxxx KQxx KJx because the spade holding is perfect (opposite  Axxx AQxxx xx xx you have to bid 4 or else you play 3 here).

   DAVIDC

   Partner bids 4 which is game but not slam interest.

   CHRIS

   I hate the methods entirely. I play that 2 is a relay, asking opener about trump length, hand strength, and distribution, 2NT and 3 are invitational, and all other bids are natural and forcing to game. In these methods, 3 is ideal- partner won't get excited without strong diamonds, and he won't go jumping randomly to game no matter what. A jump to 4 over 3 would be a picture jump, showing concentrated values in the red suits with four hearts.

   DAN

   Partner has just accepted a natural game try. I don't see that it says anything much about his feelings regarding slam. Three of a red suit would have been to play, and 3 would have shown four spades, so his available cuebids were all somewhat ambiguous: 3 sounds like a counter-game try, 4 sounds like a splinter and 4 sounds like good diamonds and might not even be perceived as forcing. [!!! --Jeff] I don't see him necessarily cue-bidding 3 every time he holds a game-acceptance and the club ace.

   DAVIDM

   The way that I play it, 2 would have been an artificial help suit relay. This would effectively prevent partner from keeping me from describing my hand. (unintended side benefit) Without that agreement, partner should not be slammish, but should probably have concentrations in +. He did not bid NT so he is unlikely to have a lot wasted in clubs.

   I believe in bidding out patterns when slam is possible. I like to get partner involved if I can (it tends to keep them happier in the long run if you occasionally pretend to value their judgement ) I think partners bidding says that he doesn't have anything to add to the description of his hand, but has sound values for what he has shown.

   ROBB

   Ambiguous. Most cuebids suggest uncertainty about strain at the 3 level.

   JOEL

   4 shouldn't be a super fitter. It's a nice hand, but not great.

   MIKE

   ...Presumably this implies a club control, since partner didn't cuebid 3 or 4 clubs en route to 4...

   BOBBY

   I play that it tends to show a lack of slam interest.

   DAVIDW

   Since 2 is only a try for game, any of partner's bids below 3 are bounceback game tries. Therefore, only 3 or 4 are viable as cue bids on [Not 3. That is needed for 4342 acceptances. --Jeff] this sequence, Partner's 4 bid shows a good hand, including 4 trumps, for 2, since he accepted. If I were to try for slam, I would bid 5, and raise 5 to 6, catering to the hand where we have a heart loser ( KJ Kxxx KQxxx Jx). [I assume this means that partner should cue if some bids are defined as such. --Jeff]

   KENT

   4 doesn't deny slam interest, but does deny the ability to cue on the way. Yes, he should cue if he's accepting cheerfully. I don't need a very good hand from partner if all of his stuff is outside of clubs. [I'm calling this "4 is nonslammish," as the real question is "do you cue when partner might just have a game try or do you cater to the rare case of a slam try?" --Jeff]

   ED

   I play methods that distinguish between responder's holding game-only hands and slam hands. Opener's bids depend on what responder is interested in knowing — a jump to 4 has no specific meaning opposite game-only interest and shows a poor fit opposite slam interest.

   MARSHALL

   I'll admit that I don't have good methods for this type of sequence. But since reaching the right game is more important than trying for slam, I wonder whether a cue bid shouldn't show a maximum with 3 trumps, suggesting notrump if partner has the right hand. [2/3NT shows three hearts over 2. Each show a maximum (only 3 shows a min) but with different surenesses about 3NT. 2NT leaves room to judge about stoppers; 3NT does not. --Jeff]

   JJ

   In terms of general methods, in game try situations I try to bid according to the principle that hands that evaluate to more than the expected maximum (in this case, about 15 evaluation points) should do something to try to tell partner about the extras, but that normal maximums should not. Thus, I wind up bidding something like this:

   • 90th - 100th percentile of hands: try to find a way to show extras
   • 60th - 90th percentile of hands: accept game
   • 40th - 60th percentile of hands: try to find a return try if possible
   • 1st - 40th percentile of hands: decline game

   Bids which would clearly show extras in this auction:

   • 4: shortness
   • 4: good

   Bids which might show extras:

   • 3: normally a return game try, possibly better

   Bids which accept game:

   • 3: shows four spades
   • 4: shows four hearts normally
   • 3NT - max without four hearts

   Bids which are NF:

   • 2NT: min with three hearts
   • 3: min with three hearts, 5+ (this could be played as forcing)
   • 3: signoff

   [Not me. 2NT is forcing; 3 is forcing. Each are natural. 3NT suggests playing there fairly strongly. 2NT is ambiguous to strength. If we play a partscore after raising a major, we play it in the major. --Jeff]

   The actual hand is interesting. It’s not clear to me as to what the best bid over 2 is. 2 and 4 (shortness with a slam try) are both good calls. 2 has the advantage of allowing for alternate strains other than hearts. 3NT and 4 are both realistic possibilities after 2. On the other hand, 4 gets across the slam possibilities right away. If pard has the right hand for slam, (a maximum with good , little or nothing in ), he’ll bid RKC and you’ll get there.) Given that this is an IMP problem and that 4 will make even if it the wrong strain, I think 4 is best.

   [That seems to me to be after-the-fact thinking. I'd never bid 4 with four so-so hearts. Not only do we get to the wrong game often, we can sometimes dope out club duplication below game with 2. Perhaps the auction might go ...2-3-3-3NT. Are we not thrilled to pass? Also, if we bid 4 and partner has Kxx, it won't occur to him that we rate to lose two trump tricks; he'll drive to slam with a hand otherwise suitable. --Jeff]

   JEFF AT THE TABLE

   I thought partner was denying slammish cards, but since I have all these aces, I already knew that.

   VOTES

   Style Votes 4 ambiguous 4 4 nonslammish 10 In Between 1

   JEFF UPON REFLECTION

   Where did this come from, you might ask? Ed and I have been having a discussion about whether it is right for a game-try acceptor to show something about his hand in case partner was making a slam try. Ed avers that he never ought to, that nearly all the time, that's just giving information to the enemy. That sounds like sensible theory, but standard methods then have a very serious problem trying to reach slams after simple slam tries. So I wondered how many subscribe to this theory. Looks like very few. Seems to me it is really good to do at matchpoints, less critial at IMPs. The right theoretical answer is to disambiguate game and slam tries. How to do that effectively is less clear, as there are other things we need to do with the space: find 3NT when it's right, change strain (say with opener's 4342 vs. responder's 4423), and have counter-game-tries. There's room for experimentation, that's for sure.

   ---

2. None vul, you hold

    xx Q10x KJ10x K9xx

   RHO	LHO
   1	2 (Bergen)
   4	Pass

   What do you lead?

   ---

   MICHAEL

   J.

   BARRY

   Trump. J might be equally attractive, but if dummy has no source of tricks then the trump looks better and I get a second chance in all probability, and only take finesses for declarer he is favorite to get right for himself.

   DAVIDC

   I'll go passive, lead a trump and pick up partner's Qxx. Who knows?

   CHRIS

   Trump. Totally random, as any of the four suits could be right. Second choice diamond jack.

   DAN

   Yuk. I lead the J with trepidation. I probably would have led a trump at MPs. Maybe I should here, too.

   DAVIDM

   I would tend to lead the small heart. It is unlikely that I will be clearing up the problem of declarers 4-4 heart fit. Playing Bergen, they might have explored that possibility. I don't play Bergen so, I am not completely familiar with followups after 1-2. If they were not playing Bergen, I would probably lead a trump.

   ROBB

   trump.

   JOEL

   trump.

   MIKE

   x. There is no special reason to believe that dummy will have a source of tricks. If I pickle partner's QJ, that is just unlucky, but declarer was likely to pick up the lone Q there anyway.

   BOBBY

   I opt for the passive-aggressive lead of a trump. This lead probably won't hurt much, and with a strong RHO on the right, I dislike underleading one of my kings (and the queen even more).

   DAVIDW

   This is a classic trump lead auction. [1-2-4? Really? --Jeff] In addition, trump leads are most likely to be effective when dummy has three trumps. At worst I'll lose a tempo, which may not be fatal when I have cards everywhere. My rather distant second choice is the diamond jack.

   KENT

   A trump. This may pickle partner's trump holding, but I have all suits stopped, no clear alternative, and would like to reduce ruffs in dummy. If declarer has a sixth trump, this may well blow a tempo.

   ED

   T (coded) or J (standard). A mild preference for this over a trump lead — it would be a stronger preference if I didn't hold the Q and K.

   MARSHALL

   A spade. No reason to be aggressive after this sequence.

   JJ

   A trump. Second choice is the correct in my methods. A trump might be right when I cut down ruffs (dummy is 3424 or similar) or when being neutral is right. At IMPs it’s close. At matchpoints I really favor a trump lead over the aggressive leads.

   JEFF AT THE TABLE

   trump

   VOTES

   Lead Votes Trump 12 Heart 1 J 3

   WINNING ACTION

   Trump. Declarer needs to ruff hearts in dummy and only an initial trump lead is fast enough to stop him. Dummy has long clubs, but our K9xx is sufficient to prevent the use of the suit.

   JEFF UPON REFLECTION

   I think the odds favor a trump lead because we have so much stuff, which means declarer's finesses rate to be offside and our chance of hitting partner's honors is lowered by the strength of our hand. If declarer isn't making this on force of high cards, he may need either a long suit in dummy as a source of tricks or ruffs in dummy. If dummy has a long suit, perhaps I have it bottled up.

   Bergen helps the defense here—I'd not lead a trump if I knew dummy had four, because it'd be useless. I couldn't stop a ruff. But with dummy's having only three, I can do something useful with trump leads.

   Before the panel vote, I thought this was a tough problem, but I guess it is not. We did win a game swing when I found a trump lead and my counterpart did not.

   ---

3. Favorable

   	K943 J953 J42 A3
   QJ52 Q10 98 Q10986

   South	You	North	East
   1	Pass	2*	Dbl
   4	All Pass

   2 was a 3-card (!) raise.

   Do you pass over 4?

   ---

   MICHAEL

   No, no, no. God deals you favorable vulnerability for a reason.

   BARRY

   I probably [pass], slowly and reluctantly but I do. Too flat in the red suits to act—obviously we all bid with just a tad more shape.

   DAVIDC

   Surely I do if dummy is going to table K9xx. But, no, I don't bid with my quacks and 8 losers which may take tricks on defense.

   CHRIS

   Yup.

   DAN

   Very close. I think I would have bid 4. Maybe double-dummy everyone is going down at the four level, but we have been known to let contracts make that shouldn't. Also, at these colours, it may be that we can't touch 4 and partner's double was a purely shape-based effort to allow us to save.

   DAVIDM

   Yes, however, if my Q were elsewhere, I would bid,  QJxx xx Qx QT98x

   ROBB

   I probably would have bid.

   JOEL

   Yes. If I was 4-2-1-6, I'd bid. But this isn't good enough reason (might be a 4-3 fit, for instance).

   MIKE

   I would have bid 4 over 4. A bit pushy, but follows the competitive rule, bidding 4 over 4 if it is at all reasonable.

   BOBBY

   Yes. Doing otherwise seems rash.

   KENT

   No. 4 at favorable, in spite of the crummy distribution. Even favorable at IMPs, I expect partner to have a very good hand to come into a live auction with an unlimited hand on his left. I think the 4 bid was based on a lot of shape—seven trumps wouldn't surprise me at all. I really don't like the looks of dummy—either partner or RHO is really light, as this is a stone max.

   ED

   Yes, I pass.

   MARSHALL

   yes.

   JJ

   I’m a 4 bidder. The say bid, the QT suggest passing. The 5-4 suggests bidding. If they are making the save must go for less than the value of game. I think I'll get out for down 2 frequently, and down 3 most of the rest of the time. I think I have a better expectation if I bid, especially given that sometimes the opps bid on to 5 if I try 4. [Odd. I think the 5-4 suggets passing. No singletons, only four trumps. --Jeff]

   JEFF AT THE TABLE

   pass. I don't like going for big numbers, but perhaps I should have saved in 4 this time.

   VOTES

   Action Votes Bid 6 Pass 9

   WINNING ACTION

   depends on your defense.

   JEFF UPON REFLECTION

   Looks like it is close. I'm a passer still. No aces, no kings, no singletons, no voids. Does partner really want me bidding on that kind of hand? On the other hand, I have a bunch of high card schmoints and everyone's bidding a lot. That suggests declarer is bidding on long hearts or some quirky shape. He might just bid 5 over 4, a huge win for our side. This time, of course, LHO saws off 4, stopping RHO from getting too high.

   ---

   You lead the Q.

   T1: Q-3-6-8
   You are playing upside-down count and attitude.

   Over to you.

   ---

   MICHAEL

   shift to a club.

   BARRY

   10. Hoping to set up a club trick before it gets away on diamonds—though I had to work to find a hand that fit the target there are some. How likely is a club to be important—only if we have to cash AK and ruff I think now. Not impossible for sure—but might declarer cover T1 then ?

   DAVIDC

   Answer 1. My monkey said to lead a second spade so I do. Could he have a "problem" with ATx or ATxx and try to give me a ruff by flying A? No, because declarer would have likely covered the Q holding the J (a Grovesner but I guess there are hands where it would work; not thought about it before.) Anyway, I don't see any other switch being all that attractive.

   Answer 2. Low . It looks like declarer has at least 4 diamonds and we need to get our club trick in before partner's card is dislodged. If partner has a second spade trick coming, it won't go away. And from ATx, partner couldn't very well play the T.

   Answer 3. . Partner has AK and you need to get your ruff right now to beat it. Partner couldn't know you need a ruff so signalled low, again possibly from ATx.

   Answer 4. T. The entire hand comes down to taking two trump tricks with QT opposite A2. If left to his own devices, declarer can't go wrong. Playing the T gives declarer a chance to play partner for AQ2.

   And the winner is (da-da-da-dut-da-da; fanfare of trumpets) "Answer 2." Seems most likely.

   CHRIS

   Diamond nine. Declarer is a heavy favorite to have six hearts, in which case there is no need to shift to clubs now. If declarer has one spade, we need at least two red tricks to defeat the contract, so we'll have time for clubs later. If declarer has two spades, a club shift only wins against a hand like  T8 AKxxxx KQx xx, in which case partner would presumably have played the spade seven at trick one.

   Discounting a ridiculously bad signal from partner, a club shift gains when declarer has  xx AKxxx KQTx xx, which would be a pretty random 4 bid. I'll play for  x AKxxxx QTxx Kx instead.

   DAN

   I'll play for a trump promotion and shift to the 9. Maybe partner has something along the lines of  Axxx — AKxxx xxxx and declarer has  x AKxxxxx Qxx Kx.

   DAVIDM

   If partner has 4 spades, it is probably safe to continue, however, if he only has 3, it may set up a pitch. Partner's 6 looks encouraging, however, what is he supposed to play holding AT6. I think I am going to switch to the 9.

   ROBB

   Club. WTP? Do I think I am tapping somebody? How am I getting 4 tricks if I don't have a club trick?

   JOEL

   I shift to the 8. Time to set up partner's king before his diamond card gets knocked out. I'm assuming that if partner wanted a diamond shift, he'd have played discouraging. I'm hoping declarer has something like:  xx AKxxxxx KQ xx or one of a variety of hyper-aggressive hands.

   MIKE

   Partner probably doesn't hold the AK or the red aces or he would have discouraged. In that case, I shift to a club to get that trick established before declarer sets up a pitch on the diamond. Perhaps partner has only 3 spades.

   I changed my mind. Partner's encouragement merely shows the 10, which is information I already had. That doesn't change the fact that there are more cases where a diamond switch wins than a club switch. (club gains when partner holds  A10xx K H10xx Kxxx, diamond when partner holds AK or the red aces)

   BOBBY

   I switch to the 9, trying to assault dummy's entries. A spade continuation won't effect the K as a menace or for its throw-in potential, and it seems right at IMPs to be aggressive.

   DAVIDW

   I switch to a low club. This looks like the right thing to do against declarer's 1-6-4-2 (or the less likely 2-5-4-2) pattern, where we need to ensure that we get our club trick. I am worried that declarer will be able to shed a club from dummy on his diamond suit. My defense caters to partner's most likely distribution, 4-1-4-4. In an emergency, partner may have to play me for the club queen to avoid an endplay, so I lead a low one to make that easier.

   KENT

   Diamond nine, playing partner for either AK or A, K.

   ED

   6. This can't cost unless declarer has KJx and a diamond loser to discard. There is a very good chance that a club loser will go on declarer's long diamonds if I don't play clubs now.

   MARSHALL

   Play J. Tempting to lead a trump, but leading Q could lose if partner has stiff king. Leading 10 might lose if partner failed to play the king.

   JJ

   Given that I’m on defense, I shift to the 9. With the 2 undisclosed hands having done all this bidding on 24 HCP, it’s extremely likely that hearts are divided 6-1 if not 7-0 between pard and declarer. I came up with a couple of constructions where a or a could be right, but they are rare. [What are they? --Jeff] It’s not always going to beat the hand to cash a 2nd spade even if it does happen to cash. By far the best chance to beat the contract is to catch pard with the AK. I also win on A, A. BTW, pard’s signal is almost meaningless; even if he has the AK he won’t know that you’re ruffing. Nor will he know that the 2nd spade isn’t cashing. [I thought about that problem at some length, and I concluded John is mistaken. If the defense is careful, 3rd hand will know the 2nd spade isn't cashing because partner didn't continue spades. More on that later. --Jeff]

   JEFF AT THE TABLE

   Shift to a club. I trusted my partner, but I think that was probably wrong.

   VOTES

   Action Votes Club 9 Spade 1 Diamond 5

   WINNING ACTION

   Diamond. Partner has  A1076 — AKxxxx xxx.

   JEFF UPON REFLECTION

   I'm with Joel. At the table, I didn't analyze the hand beyond "partner encouraged spades, so he can't have the AK or he'd want a shift real bad." Problem is, I couldn't find a construction on which a club shift was necessary to beat it. Fortunately, Mike and Joel found some, albeit unlikely ones.

   I'm a little surprised that so few of the panel even mentioned partner's carding. It mattered a lot to me at the table, convincing me to play a club rather than a diamond. As a matter of fact, I think this problem is all about partner's carding. I wonder why so few others agree.

   What ought partner play at T1? One might think that it depends on the partnership signalling style (note that Mike and I disagree about the meaning of the 6). Nope. Partner should play the 10! That's not an upside-down signal. It's not a right-side-up signal. It's not attitude, not count, not suit preference. It's a pure signal. If partner can afford to blow up the spade suit (if I have only QJx, he's given declarer a trick) he must have fast winners outside of spades. Where can they be? Diamonds. Therefore, I logically have no choice but to shift to diamonds. He also cannot have any combinable honors in clubs or we might be getting them too slowly. Therefore, he has roughly the exact hand he has and my defense becomes clear. Those things never work at the table, though. Partner usually says, "that's a very clever play. Perhaps I ought to have worked it out." What they really mean is, "that so-and-so keeps giving me goofy problems like this and he's complaining because I don't solve them all??? He can just ... "

   Ignoring the issue of the 10, what ought partner's card at T1 be? Attitude? When the Q holds, his attitude is known. Count? We'd be pretty surprised if he didn't have exactly four spades. And even if he didn't, do we care much how many spades he has? It's not directly going to influence our action. What we need to know right away is to what to shift. If the hand were played the other way around, with dummy's having the singleton spade, most would play partner's card as suit preference. Here, everyone has a good idea that declarer has a singleton, so I think the T1 signal should be suit preference. If you play "Obvious Shift," it comes to the same thing—encourage spades if you want a club, discourage if you want the obvious shift, diamonds. More simply, 3rd hand knows opening leader's problem. Opening leader knows his partner knows his problem. Therefore, 3rd hand's card should be an attempt to solve that problem. If one has strict partnership rules about signals (sometimes a good thing), one might have to find the 10 signal. If your partnership is more flexible, then I think this should be a suit preference case.

   What about later defense? Let's say partner plays the 10 like a good boy. What should we do? I don't see how the 8 will help; I think it'd deny a doubleton diamond when we follow with the 9 later. That sequence would suggest a spade is cashing. On the other hand, since we know partner's diamonds are fast winners, we can do better. If we have only three spades, we can continue spades to cash partner's second spade when we are 3-3 in the pointed suits. Then partner cashes his diamonds for down one. How does he know we have a third spade, not a doubleton? We continue with the small one, not the J, keeping the card we are known to hold. With QJ tight, we continue the honor and partner gives us a ruff, gets back with a diamond, and judges from the hand count and our diamond spot whether to try for a trump promotion or to cash another diamond for down two. What if declarer drops the Q under the first diamond winner? Partner has to cash another. We aren't beating this if only one diamond cashes. OK, that's the complicated partnership defense. Here's the simple logical defense. After one spade and two diamonds cash, partner is free to continue diamonds no matter whether he trusts us or not. Why? Because the spade trick isn't going anywhere. The trump promotion might disappear, but if declarer has two spades, partner will take trick thirteen anyway.

   All in all, it's a fascinating hand. And the panel blew it just as badly as I did at the table!

   ---