Some Problems from the Reno Region 12/00: Answers

Today's Panelists: Barry Rigal, Robb Gordon, Mike Shuster, Dan Hugh-Jones, Ed Davis, Rolf Kühn, Bobby Bodenheimer, and a couple of guests.
  1. BAM, unfavorable, you hold

     S:xxxx H:Kxx D:Axx C:Axx

    LHO CHO RHO YOU
    2H: 2S: Pass ?

    BARRY
    I drive to game, so 3H: then 3NT seems right. offer the choice. Note: one of Mark Feldman's ideas (via Martel?) is that 3H: in the sequence 1S:-(2H:)-3H: should be limit or slam interest while 4C: should be game values for spades. does this apply here too by extrapolation? [Doubtful. I play 4C: is clubs and spades, even though the opponents do not appear to be showing any enterprise yet. That is probably not best, but at least it's a definition. --Jeff]
    ROBB
    3H:. Different vul I would bid only 3S:, but the H:K could be worth something.
    MIKE
    The choice is between 3H: and 3S:, but I go with 3H: for two reasons:
    1. the fourth trump and
    2. my LHO's favorable preempt doesn't necessarily make my H:K worthless.
    DAN
    3H:. What else? Sure, 3NT may be the only game that makes, since it will play from the right side, but it is just as possible that it fails with 4S: making and, with 3 prime cards and four trumps, 3NT will cause us to miss slam a significant amount of the time since pard will never believe we have this good a hand for him. These days, and at these colours, nothing says the heart king is even a wasted card. [No, but should we be betting on it? --Jeff]
    ED
    2NT. It doesn't look like it is worth more than an invitational bid to me. I'll advance with NT although I don't mind 3S:. One advantage of NT is that we can still get back to spades whereas a 3S: bid makes it difficult to reach 3NT when it is right.
    ROLF
    3S: - I see no real alternative.
    BOBBY
    3H:. Partner could have a really big hand here, and even though my shape is bad, I do have support and values, so I'll show that on the way to 4S:.
    JEFF AT THE TABLE
    It was an opponent's problem.
    CONSENSUS
    3H:
    VOTES
    3H:5
    3S:1
    2NT1
    WINNING ACTION
    3NT. Partner has fast winners, but a heart lead generates five tricks for the defense in a hurry.
    JEFF UPON REFLECTION
    At the table, I thought 3H: was a mild overbid; when LHO's partner signed off in 3S: and LHO bid 4S:, I was sure it was after seeing dummy.

    It seems wrong to suppress spade support, but 2NT really seems like the right bid. It's dead-center on values and protects the H:K. If partner bids anything other than 3NT over 2NT, I'll get to bid spades. I don't see how to get to notrump after starting with a raise, so at BAM, I think 2NT is the best bid. Come to think of it, that's pretty much Ed's argument. My second choice is 3S:. At BAM, just because the H:K might still be good isn't enough; it's likely to be bad so I think it's right to go with those odds.

    Put it another way: If we held  S:xxxx H:Kxx D:Kxx C:Axx, and partner opened 1S:, would we make a limit raise or a single raise? I think this is just between them. Improving the D:K to the D:A and making the H:K questionable should diminish the hand slightly overall. Therefore, I think it's not quite a limit raise. But it's close enough, I suppose.

  2. BAM. West deals.

    S: QJx
    H: x
    D: AQxxx
    C: AKxx    		 S: A1098
    H: AKQx
    D: Kxx
    C: QJ

    Come up with a good sequence to 7NT, playing generally standard methods. (E.g. no big club.)

    BARRY
    WestEast
    1D:1H:
    2C:2S:
    3S:4D:
    5C:5H:
    6C:7NT
    Opener shows strong 3154 and over 4D: (setting the suit) bids 5C:, denying second round spade card. 5H: is a grand slam try promising S:A and H:AK now 6C: is obvious and now East knows partner has at most 4HCP in majors, with 7 points in clubs, so must have D:AQ? ["Must"? He can't have  S:QJx H:x D:A10xxx C:AKxx? Why not? --Jeff] Might I have bid it this way? Don't hold your breath! [Why not? Every action seems pretty down-the-middle until the last one, which is a bit of a guess. --Jeff]
    ROBB
    I am clueless about this kind of thing. How about
    WestEast
    1D: 1H:
    2C: 2S:
    3S: 4D:
    4H: 4S:
    4NT 0/3 KC:
    5NT 7NT
    MIKE
    The start of the auction is fairly staightforward with
    WestEast
    1D: 1H:
    2C: 2S:

    West's and East's third bids are where the world will turn. Should West continue 3S: (whether or not this shows extra values is a good question, [Depends on whether 2S: is game forcing. If not, then yes it does. Otherwise, I don't see why it must. --Jeff] but it certainly pinpoints the distribution) the auction could proceed
    ...
    3S: 4D:
    4S: 4NT
    5S: 5NT
    6C: 7NT

    [Why would West bid 4S:? I can barely see 4H:, but I don't like to cue shortness in a suit where partner probably wants stuffing. Most likely, over 4D:, West will bid 5C:, ending the Blackwood story. Indeed, when faced with that exact situation, I figured partner would be bidding 5C: and therefore avoided the 4D: call. --Jeff]

    Should West continue 3D: over 2S:, the auction is much easier:
    ...
    3D: 4NT (4H: redwood would be nice but you've bid hearts already so it isn't redwood)
    5S: 5NT
    6C: 7NT

    DAN
    WestEast
    1D: 1H:
    2C: 2S: (4th suit)
    3NT4D: (3NT = extras)
    5C: 5H:
    6C: ?
    Easy enough so far, but at this point, I give up on science. I think East has to take a mild flyer. His pard is cooperating in a try for a grand [Or 6NT? --Jeff] missing an awful lot. East can count the tricks as long as his partner has the D:AQ, and I know of no way to ask. I confess I would be tempted to just bid it hoping for it to be at worst on a diamond hook.
    ED
    WestEast
    1D:1H:
    2C:2S:
    3NT4D:
    5C:7NT
    From East's point of view, 7NT is reasonable (and very good when West holds the D:J) if West is 2-2-5-4 or 3-1-5-4 with any of the following high card combinations:

    a) D:AQ + C:AK OR
    b) S:K + D:A + C:AK and either S:Q or D:Q OR
    c) S:KQ + D:AQ + C:A

    I don't see how a cooperative standard auction can reasonably go any differently through 4D: and, although West would bid differently over 4D: with each of the above example set of high cards, his willingness to cue bid 5C: means that 7NT is going to have a good play.

    [I wonder how much extra 5C: shows. Clearly, 4NT would be regressive, but, more or less, every other call is forcing to slam, because 3NT is known to be playable. Does that mean a cue below game is a grand slam (or 6NT) try showing substantial extras? I've never really thought about that inference before. Obviously, it does not apply at IMPs. --Jeff]

    ROLF
    WestEast
    1D: 1H:
    2C: 2S: (4SFG)
    2NT 3D: (strong SI)
    4H: (spl) 4S: (relay)
    5S: (single + 2KC + D:Q) 5NT (relay - spiral scan)
    6D: (C:K no S:K) 7NT
    BOBBY
    WestEast
    1D: 1H:
    2C: 2S:
    2NT3D:
    4D: 4H: (RKCB in D:)
    5D: 5H:
    6C: (C:K) 7NT
    JEFF AT THE TABLE
    PartnerMe
    1D: 1H:
    2C: 2S:
    3S: 4NT
    5H: 6NT
    JEFF UPON REFLECTION
    It seems easy if we can get diamonds agreed at a low enough level:
    WestEast
    1D:1H:
    2C:2S:
    3D:4NT
    5S:5NT (specific kings)
    6C:7NT
    or
    WestEast
    1D:1H:
    2C:2S:
    2NT3D:
    3S:4NT
    5S:5NT
    6C:7NT
    The problem, of course, is that West isn't going to bid 2NT or 3D:. He'll bid 3NT or 3S: to show his values or shape. This time, those interfere seriously.

    Another try is:
    WestEast
    1D:1H:
    2C:2S:
    3S:4D:
    4H:4NT
    5S:5NT
    6C:7NT
    The problem with this is that West isn't going to bid 4H: over 4D:; he'll cue clubs and we won't be able to get to the grand with confidence. Moreover, East knows that West isn't going to cue a major, so bidding 4D: is pointless.

    In fact, that's exactly the problem with which I was faced, so I tried instead to find the S:K/D:A/C:AK combination via 4NT over 3S:, found a missing key card and stopped in 6NT.

    I don't see a way to cater to S:K/D:A/C:AK and also to D:AQ/C:AK without West's misbidding or East's guessing. Ed thinks he'll be guessing with the odds, but bidding grands on guesswork isn't my style.

    This suggests, however, a refinement to 4th suit forcing. Let's assume it's already forcing to game. (Other tinkers are probably best if it can be invitational.) There are three big messages that need to be sent:

    1. 3-card support for partner's major
    2. stoppers in the 4th suit for NT
    3. neither of those.
    (3) includes a bunch of other possibilities; for example, opener might be able to show extra values or show his shape, but how important are those? When responder bids 4SGF, he already knows his rebid most of the time. The exceptions occur when he's missing stoppers for 3NT or he's looking for a 5-3 major suit fit. I propose, therefore, that it might be best to play that in reponse to 4SGF, opener support partner with 3 (but not with 4333 and good stoppers in the 4th suit, a rare occurrance), bid the cheapest NT with a stopper, or bid the cheapest other bid otherwise. This will likely let responder get in his minor suit forcing raise when he has one. John Swanson plays "Extended PLOB" which has as the default action a rebid of opener's second suit. Most play that the default action over 4SF is for opener to rebid his first suit. I guess the only difference here is that I propose the default rebid to be the cheapest non-preference that isn't a notrump bid. Some examples:

    1H:-2C:; 2D:-2S:

    2NT = natural
    3C: = natural
    3D: = default
    PLOB and not helpful.

    1D:-1S:; 2C:-2H:

    2S: = natural
    2NT = natural
    3C: = default
    PLOB and helpful if responder has diamonds (likely)

    1H:-1S:; 2C:-2D:

    2H: = default
    Standard and helpful.

    Since the cheapest non-preference non-notrump bid is one of opener's suits, that means the default should probably simply be cheapest of opener's two suits. Seems reasonable to me.

  3. BAM, none vul, you hold

     S:Jx H:J10x D:AQxxx C:Axx

    You LHO CHO RHO
    1D: 2H: 3C: 3H:
    PassPass3S: Pass
    4C: Pass4S: Pass
    ?

    BARRY
    6C:. It is not clear that playing 5C: making would be bad but my minor aces and black length make this action reasonable--- no wasted (or indeed any) values... a typical excuse!
    ROBB
    5C:.
    MIKE
    Presumably partner is 5-6 in the black suits. If partner holds  S:AKxxx H: x D: x C: KQxxxx he will likely make 4S: (ruff the second heart, cash S:AK, run clubs) and 6C: will be touch and go. If partners spades are a little worse: AQ10 or KQ10 we could be in deep trouble at 4S:. [3-3 or maybe a hook? --Jeff]

    Likely making 4S: should be compared to the likely result at 5C:. If partner has the first hand we are about 50-50 to make the same score in 5C: via the overtrick. That compared with the increased safety when partner's spades aren't headed by the AK lead me to believe that 5C: is indicated.

    DAN
    5C: Again, this doesn't seem all that close. I suppose we might miss a slam if pard has a good hand but feels he has done enough, but he should be aware that with no heart control, I may not have wanted to get past 5C:. The fact that I didn't try 3NT or double over 3H: may encourage him to think I have little wasteage in hearts.
    ED
    5C:. My hand hasn't gotten any better and partner is just showing 6-5 distribution, not cue bidding to try to reach a slam. 5C: looks like a better spot than 4S: to me.
    ROLF
    5C: - I know this is BAM but I don't think that 4S: is a good spot.
    BOBBY
    Pard is going to face an immediate tap if we stay in 4S:, and I don't think our chances there are that good, as he would have bid this way with  S:AQxxx H:x D:x C:KQxxxx. So I correct to 5C:.
    JEFF AT THE TABLE
    Pass
    CONSENSUS
    5C:
    VOTES
    5C:5
    6C:1
    Pass1
    WINNING ACTION
    6C:. Partner has  S:AKQxx H:x D:x C:KJ10xxx. The C:Q is easily found. Spades are 5-1, though, so 4S: probably goes down while 6C: makes. Ed thinks partner underbid.
    JEFF UPON REFLECTION
    I'm convinced. 5C: is better than 4S:. When 4S: makes only four, sometimes 5C: is making an overtrick to tie the board. I figured that I'd already taken a preference to clubs; if partner wanted to play clubs that badly, he'd've bid something other than 4S:. Hence, I placed him with good spades. But he doesn't know I'm not 3-3 in the blacks.

  4. BAM, both vulnerable, you hold

     S:KJ10 H:Qx D:AJx C:9xxxx

    Partner You
    1S: 1NT (semiforcing)
    2C: 3S:
    3NT ?

    BARRY
    Has partner promised four clubs---i.e. with 5332 14 could he bid this way? [He has not; he could. --Jeff] If so I pass, but otherwise I bid 4S:, too tough to play 3NT with two such good black fits. [Therefore, Barry passes. --Jeff]
    MIKE
    This one is a tough cookie. Partner mearly needs to be balanced to try 3NT here - some red suit honors, but not necessarily a lot of stoppers. Most of our values are in the offsuits, offset by the presumed double-fit in clubs... Since it is BAM, partner may be inclined to offer 3NT as a choice more often than at IMPs, but by the same token, I shouldn't go out of my way to reject that offer. I have too much in the reds to presume we can't make the same number of tricks in NT and the possiblilty of not having an 8 card club fit [Mike plays that 2C: could be a 2-card suit once in awhile. --Jeff] ices the deal. I pass.
    ROBB
    Pass.
    DAN
    Pass and pray the opponent on lead will talk himself into believing your partner is prepared for the other major...
    ED
    4S:. I'd pass with one fewer club and one more heart (but then it wouldn't really be a problem). It seems silly to leave it in 3NT when partner might not have more than H:Jx.
    ROLF
    Looks like pard has about 10 HCP in the reds, which does not leave very many HCP for the blacks. As we will probably have all the black losers in S: but will only need one black suit in 3NT I pass.
    BOBBY
    It's a close choice, but I think I'm likely to get 10 tricks in spades easier than 10 in NT, so I correct to 4S:.
    GIB
    Pass. Worth about 10% more matchpoints on a 133/167 deal sample. (The first number is the number of unique hands generated for partner; the second is the number used all told, perhaps by shuffling the opponents cards.) [I thought this would be a good candidate for simulation, so I asked GIB (actually Matt Ginsberg). --Jeff]
    JEFF AT THE TABLE
    4S:
    VOTES
    Pass5
    4S:3
    WINNING ACTION
    Hard to say. Single-dummy, 4S: is the winner, as partner has  S:AQxxx H:AKx D:xx C:KQ10. Double- dummy, however, 3NT is the winner, as the D:KQ are in the slot and clubs are 4-1 offside.
    JEFF UPON REFLECTION
    I think it's a guess. Does partner have  S:Qxxx H:AJx D:Kx C:Axx? Or does he have  S:Q9xxx H:Jxx D:AK C:AKx? I have no way to tell, so it's an odds issue. This is probably an excellent hand for a simulation, but it's going to be a very hard one to perform, since overtricks are crucial. If he's 5233 in contrast to 5323, it'll almost always be better to play 4S:, and that swings the balance, I think. Barely.

    Upon reflection, I'm sorry I asked GIB; it depends so much on what hands partner will bid 3NT. If he can have H:Jx, as Ed thinks, it's pretty clear to bid 4S:. If he has to have stoppers in both red suits, it's probably right to pass, but still quite close, as a heart doubleton will be trouble an awful lot of the time.

  5. Matchpoints, unfavorable, you hold

     S:10xx H:xx D:QJ9xxx C:97

    LHO CHO RHO You
    1C:* 2C: 2H: Pass
    3H: Pass4H: Pass
    4S:* Pass5D: Pass
    6H: All Pass

    1C: was strong, 16+.
    4S: was key card in hearts.

    What do you lead?

    BARRY
    C:9. Partner did not double so he must be prepared for this. I can't see why I have to be the hero every time.
    ROBB
    The pedestrian C:9. The alternative diamond honor will give away as many unmakeable slams as it will beat otherwise makeable ones.
    MIKE
    Partner went out of his way to bid clubs vulnerable. I see no reason not to lead his suit. [Other than LHO's bidding slam when he expects a club lead? --Jeff]
    DAN
    D:Q — unless the state of our game suggests that winning the post mortem/keeping partner happy is more important than beating the slam. I think double would suggest a non-club lead and at IMPs might therefore lead one, but at MPs partner might not have felt he/she could risk it.
    ED
    C:9. Very close between the D:Q and the C:9. The D:Q needs to find the D:A with partner and the D:K in dummy with no stiff diamond in either hand or the D:A with the opponents and the D:K with partner and stiff diamond. The C:9 needs to find partner with a major suit winner and a cashing C:K (or, less likely, C:A). Given the absence of a splinter by LHO and no 4C: cue bid over 3H: by RHO, I think the second round club is likely to cash and thus the club will be right.
    ROLF
    This depends 80% on what kind of suits you bid over the big club. As I think that the suit does not have to be of good quality I lead D:Q following my usual aggressive policy.
    BOBBY
    C:9. I could lead a diamond but I don't have a strong preference for it, so I'll lead partner's suit.
    JEFF AT THE TABLE
    Wasn't there.
    JEFF WHEN ASKED
    Club. A diamond never occurred to me, even though I know LHO is probably prepared for a club. It seems like a shot in the dark, a huge parlay, whereas a club could be right for a number of different reasons. There's more to be said for a diamond lead at IMPs; we might be defending a normal contract with the club lead holding down the overtrick.
    CONSENSUS
    C:9
    VOTES
    C:97
    D:Q2
    WINNING ACTION
    A diamond. Dummy has D:K10 tight and partner has the Ace.
    THE WHOLE STORY
    This was an appeals case. 4S: and 5D: were not post-alerted. Our problem hand holder claimed that he was therefore barred from leading a diamond by the misinformation.

    The rules for this are difficult to use. If it is accepted that misinformation (MI) is present, and if the misinformation seems to lead to the non-offending side's (NOS) bad result, then we adjust the scores as follows: we give the NOS the best "likely" result and the offending side (OS) the worst result "at all probable." The laws carefully refrain from defining the quoted phrases in the last sentence, but good rules of thumb are that "likely" means 30-35% of one's peers will take that action, and "at all probable" means about half that number will take the action.

    The director did a very good job on this. He polled five players whom he thought were the leader's peers, giving them both auctions. All led a club with the wrong information. One changed to a diamond with the right information, but he had no reason for it. One said that he "had sympathy" for a diamond lead with the correct information, but would still lead a club. The rest thought clubs was the only choice in either case. As a result of this, the director ruled that the diamond lead was not "likely," but was "at all probable," and gave the NOS -1010 and the OS -50. He added a 1/6th board procedural penalty (PP) to the OS for the failure to post-alert; this was a Flight A pairs game and they should know better.

    That's a pretty good ruling. I wish all of our directors would do as good a job.

    The committee, however, felt differently than the experts polled. One thought that leading a diamond was a 0% action—a shot in the dark. Another thought that it was about a 20% action; I thought it was somewhere between 10% and 20%, but closer to 10%. Hence, all three felt that a diamond lead was not likely, so we left the NOS's score at -1010. The trouble was the OS's score. With one feeling strongly that a diamond lead was not at all probable, one feeling that it barely was, and one feeling that it was right on the borderline, the consensus was that it was, therefore, not at all probable, so we gave the OS +1010. We felt that this was a bit of a gift to the OS, and I strongly felt that this pair should be fully responsible to avoid procedural errors that cause committee hearings, so we increased the PP to half a board. I, personally, would have felt quite comfortable letting the director's ruling stand. This was pretty close to it, however, so I found it acceptable.

    Given the new information from the group here, 25% of the respondents chose a diamond lead. Ignoring other circumstances, that strongly suggests that a diamond lead is at all probable, so upon reflection, I think that the committee got it wrong and the director got it right. There are, however, other circumstances that I had not considered until after the deliberation. The opening leader was playing with his wife. She (and he as well) has a bit of a temper, and doesn't hesitate to announce her dissatisfaction with his (usually excellent) play. Given that, I would never lead a non-club, just so that I would not lose the post-mortem. I thought of this point between the committee deliberation and the announcement of the results, so I was rather more satisfied with our choice than I was during the discussion.

  6. Matchpoints, favorable, you hold

     S:Q H:KQ D:x C:AKQ109xxxx

    Pass, pass, ?

    BARRY
    5C:. mundane—that's me!
    ROBB
    5C:.
    MIKE
    5C:. A preempt known to be weak is a [blunt] sword.
    DAN
    3NT, no second choice.
    ED
    I think I'd like to open 4C: Gerber. Barring that, I'd bid 1C:.
    ROLF
    3NT - to play.
    BOBBY
    1C:, followed by a large number of clubs on my next bid (4 or 5). The problem with opening more than one club, as I see it, is that partner will never be able to correctly evaluate his hand. Granted that it will be hard to show this hand in any event, but at this vulnerability I doubt the opponents would push to 5S:.
    JEFF AT THE TABLE
    5C:
    CONSENSUS
    5C:
    VOTES
    5C:4
    1C:2
    3NT2
    WINNING ACTION
    beats me. Open 6C: or something like that. Partner has  S:xxxx H:AJx D:Axxxx C:x. 6C: is laydown; on a non-spade lead, 13 tricks are claimed. I don't know what the normal lead vs. 3NT would have been; I never saw their hands or a hand record.
    JEFF UPON REFLECTION
    Shrug.

  7. Matchpoints, favorable, you hold

     S:xx H:xxxx D:xxxx C:Kxx

    RHO You LHO CHO
    1S: Pass1NT Pass
    2C: Pass2S: Dbl
    Pass?

    1NT was forcing.

    a) What does partner's double mean?

    BARRY
    Spade stack in my seat, but here I think T/O. Sorry if I am wrong—they deserve to catch me if partner has five spades.
    ROBB, MIKE
    Penalties.
    DAN
    This feels ugly. Sounds like takeout of spades although why partner couldn't double 1NT is hard to fathom. Maybe pard is pre-balancing light and the opps actually do have a fit, i.e. the 1NT bidder has three spades. Pard is under the bidder and he is unlikely to be stacked since I have two.
    ED
    Takeout, e.g.,  S:Axx H:KQJx D:Axxxx C:x.
    ROLF
    He has the red suits.
    BOBBY
    Partner is balancing.
    KIT WOOLSEY (WHOM I ASKED AFTER THE SESSION)
    This is NOT a balancing auction. It is 100% a penalty double.
    JEFF
    Penalty. This is not a balancing (or pre-balancing) situation. The opponents have not found a fit. They are still in a wide-ranged auction (LHO has 0-9 or so and RHO is 11-18). LHO might well have a big balanced hand, itching to double you. If you want to make a takeout double of spades here, you double 1NT, not 2S:. You can't play two-way doubles, either; the opponents have between 7 and 10 spades.
    VOTES
    Penalty4
    Takeout5
    b) What is your call?
    BARRY, DAN, ED, ROLF
    2NT scramble seems obvious if I read the double right.
    KIT, MIKE, ROBB, JEFF
    Pass.  S:KQJ10 H: Ax D: Axxx C: Axx. Down 2.
    WINNING ACTION
    None. You can either go for 800 or 870.
    JEFF UPON REFLECTION
    This is a 50/50 split, which provides only the information that low-level doubles are still a province of possible partnership misunderstanding. When a pair tells me at committee that they've discussed all of them, this is a problem I can whip out. Not that I would have believed them anyway.

    As I see it, here are some arguments for playing the double as penalties. Ed gets to respond editorially this time.

    1. There's a very narrow range of hands that won't bid before and want to now.

      [So what? The frequency of 2 or 3 spades and 5-4 in the reds and an opening hand is considerably higher than that of a penalty double. --Ed]

    2. Such a hand might well be in big trouble even if everything is perfect for action.

      [You are thinking -870 and -800. I am thinking +200 and -50/+110/+140. Even if we might be in trouble by bidding, this is an argument against entering the auction, not against playing double for takeout. --Ed]

      An argument against entering the auction with a takeout double is the same as an argument for not playing the takeout double in the first place. Taken to the extreme, to play 7C: here as a light takeout for the reds is absurd, of course. Why? Because we'll get a bad result by using the convention nearly all the time. The magnitude of the argument is not similar, but the argument applies here: if we are going to get a bad result fairly often when using a takeout double, that suggests a takeout double isn't the right choice of agreements.

    3. Our matchpoint expectation isn't bad yet, so bidding isn't "free". At IMPs, I doubt there are any hands worth the risk, since the upside is 5 IMPs and the downsize a dozen.

      [I don't buy this as indicating we should not act with appropriate hands. Why settle for +200 or -110 when we could be +500 or -50. It is losing bridge to enter earlier or to fail to double 2S: with hands like  S:xx H:AKxx D:AJxxx C:xx or  S:Axx H:Axxxx D:Axxx C:x or  S:xx H:AJTxx D:Axxxx C:x. --Ed]

    4. If we have a business double of 2S:, we can't take action before and the change in matchpoint expectation is quite large. The odds of one, however, are slender.

    5. 2NT is available for takeout.

      [That is fine for hands with sufficient playing strength but without defense such as  S:xx H:KQTxx D:Kxxxx C:x. But it is a poor choice to bid 2NT on 5-4 hands and hands where you hope partner's spades and clubs are good enough to pass the double for penalty. The position that we should all easily agree on is that it is wrong to bid 2NT with those hands. --Ed]

      It never occurred to me that partner is going to pass for penalty. I don't think he'll know enough to judge when to do it. When he can do it succesfully, I suspect, moreover, that the opponents will not stay in spades. They'll run to clubs or notrump. Converting +200 to -110 (or -670?) isn't going to happen too often, but it's another way to lose for the takeout double.

    6. If they bid on, I suspect that (a) they would have anyway, so we don't gain by pushing them, and (b) we convert some sets into makes for them.

      [You gain whenever they can't make their bid. But that is just a small part of the gain from competing. You gain, and maybe lots in matchpoints, by denying them +110 in 2S:. --Ed]

      We only gain if they weren't going to bid again anyway. If we were getting +100 against 3S: anyway, then we've simply broken even if we push them to a spot they were going to bid on their own. And telling them how to find black cards may be a disaster.

    7. One is allowed to bid a five-card suit the first time, even with a side four- or five-card suit, particularly if there's significant lead value difference. A delayed double can get the other suit in possibly. For example, you hold  S:Axx H:Kxxx D:AQJxx C:x. After 1S:-1NT, bidding 2D: is certainly an option. Make the H:K and a low spade into low hearts and 2D: is still not awful. The expectation difference between such an action and a delayed takeout double seems rather small.

      ["allowed," "can," "possibly," "still not awful," "seems"... add them all up and it looks like you don't think too highly of this as an argument for playing double of 2S: as penalty. If so, I agree with you. --Ed]

      Agreed. It's only another minor argument---that some of the time, hands that qualify for the double can be bid another way without a severe loss in expectation. It'll be a loss, but not much, I claim, so we can survive on those hands.

    AFTER ALL THIS
    I'm not convinced either way. In this particular case, I think the double should be penalty without partnership agreement to the contrary. Many partnerships, however, play "undiscussed low-level doubles under the bidder are takeout, and undiscussed low-level doubles over the bidder are penalty." In fact, I do with some partners. If so, this is easy; it's takeout. Having that rule is surely better than having to guess, regardless of which treatment is more effective.

    In any case, partner held  S:Jxx H:KQxx D:AKxx C:xx. Neither use of a double would occur to me on that hand. Not surprisingly, nothing good could happen. From the examples he gave, Ed would double if a low black card were a red card, so partner was only a trick off. My hand is a bit of a disappointment, but I do have shortness in spades and double trump support. We still would go for a number vs. a partscore.

  8. Matchpoints, favorable, you hold

     S:QJ9x H:KQ8xx D:J10 C:KQ

    You Partner
    1H: 2D:
    2NT 3C:
    3NT 4C:
    4NT 5NT
    ?

    2D: wasn't 100% game forcing; 2NT showed a little extra. 4NT was natural, non-forcing. Feel free to excoriate previous actions.

    BARRY
    6D:. All the previous actions seemed right; I expect 6D: to play as badly as 5NT—down one or so. But it is worth more to make it!
    ROBB
    6D:. I wouldn't know where to start.
    MIKE
    Previous auction is alright. I might have stalled with 2H: over 2D: to give partner room to bid 2S:, confident that I can handle the auction and that it likely doesn't matter from which side we play the hand, other than issues involving skill at declarer play.

    My first inclination was to bid 6D:, but it turns out that was just a knee-jerk preference. More carefully considered: Partner has between 11 and 13 minor suit cards, most likely  S:K H:— D: AKxxxx C: AJ10xxx

    If partner's diamond's aren't AKQ, we are going to be in trouble off one of the major aces, so we must bid 6C: to ruff the diamond loser.

    DAN
    Pass. Having failed to raise partner on the last round I think I have to live with my decision. 4D: on the last round looks much better than 4NT to me.

    [Excuse me, but I think this is dead wrong. To bid 4D: on the previous round commits the hand to slam. If 3NT is playable at matchpoints, you don't play five of a minor, and if you bid 4D:, the next natural notrump bid will be 6NT. Given the total dog we have, one last attempt to stay out of slam seems prudent. --Jeff]

    I am not going to guess now. Despite the great spot cards, I have one fewer ace than pard thinks I have.

    ED
    6C:. We may have diamonds to ruff (e.g., AKxxx) but are not as likely to need to ruff clubs. The given auction looks fine to me. It is not stated but I assume that 5NT is "pick a slam." [Up to you, but that's what I think 5NT must be. Not everyone agrees. --Jeff]
    ROLF
    The bidding sounds a little bit strange but everything seems fine. As I already showed some extras I actually should resign in 5NT, however although the hand is ugly D:JT and C:KQ are too good. I bid 6NT mostly because of the D:T.
    BOBBY
    I don't like my hand and don't really feel like I've got a "little extra". I don't really know what 5NT is—I suppose it's invitational to 7, since partner would have clearer bids to tell me to pick clubs or diamonds. Assuming that, I don't have anything extra to go to 7, so I'm going to bid 6NT and hope for the best.
    JEFF AT THE TABLE
    6C:
    VOTES
    6C:3
    6D:2
    6NT2
    Pass1
    WINNING ACTION
    6NT. Partner wasn't on the same wavelength and had a hand that doesn't resemble his auction. It's still a good problem.
    JEFF UPON REFLECTION
    I'm pretty sure 6C: is the right choice and evidentally convinced Mike at the site. If partner's diamonds can play opposite J10 tight and he is missing the C:KQ, how can he be asking me to choose without at least C:AJ10xx? If so, clubs must be playable, but diamonds need not be; he could easily have D:AKxxxx, in which case we need to ruff a diamond. If he only has five clubs, let's hope he has the 9, too, so that he can afford to overtake my honor, or that they lead his winner, or that clubs are 3-3.

    It's also possible that partner is asking us to choose between diamonds and NT. If so, he'll bid 6D: over 6C:. Given our majors do not include any aces, the suit slam has to be right, as he'll usually have a void somewhere or a stiff spade or will have committed to 6NT himself.

    Three thought that 5NT was natural, inviting 6NT (or 7NT). I can't agree with that; opener's strength is known to within a fairly tight range (roughly a crummy 14 HCP). His attitude about slam is already known (NONONONONO!!!!!) So why ask him again? One could get a choice between the minors, however, with either 5S: or 6C:. Or 5NT. What's the difference? I have no idea.

Jeff Goldsmith, jeff@gg.caltech.edu, Jan. 8, 2001