Join Bridge Winners
All comments by Phillip Martin
You are ignoring the author of this comment. Click to temporarily show the comment.
Indeed I should. Now I have to ask myself why I didn't think of that. I suppose it's because my first thought was that I had to guess who had the doubleton. Once I found a play that improved on that, I was happy. Deciding when it's OK to stop thinking is one of the hardest problems in bridge. If we were to encounter this as a problem in a book (“Find a sure trick line for an overtrick, assuming split honors.”), we would solve it. At the table, no one tells you there is a sure trick line, so once we find a line that works much of the time, we often stop looking. I have no clear idea of how to avoid errors like this. I suppose the easiest way to avoid this exact error is to know the combination off the top of your head, which obviously I did not. Now I do. Thanks for alerting me.
Jan. 9, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
I am a notoriously slow player. (See https://sites.google.com/site/psmartinsite/Home/bridge-articles/the-huddle.) And I am slower now than I used to be from lack of practice. This particular hand, however, did not take long, since once you see that leading the queen picks up more holdings than leading low, it's unnecessary to go through the calculations to determine that East is more likely to have a doubleton heart than West. Note that, in the post, I postponed the discussion of that fact until after the hand was over. That's precisely because I didn't work that out during the play of the hand; it's something I thought about afterwards. (Although in rereading the post, I see that perhaps that wasn't clear. I should have omitted the “especially when you consider” phrase.)

I struggle to be honest that way: to discuss only what I actually thought about during the play, saving any additional points until the post mortem. Sometimes I feel guilty in leaving a train of thought incomplete (as in saying “the defense seems well placed at this point,” then moving on without actually verifying that declarer will go down in that position). Typically, bridge articles don't do that, and I have to keep reminding myself that my objective here is different than that of typical bridge articles. As you point out, you seldom have time to analyze a hand fully at the table, so shortcuts like that are necessary, and if I am to represent my thought processes accurately, taking such shortcuts is necessary.

I will say that reconstructing hands is absolutely essential to playing well. This is something Woolsey impressed upon me years ago. You don't need to construct a lot of hands. But if you find yourself saying things like “It looks right to tap declarer,” “It looks right to cut down on ruffs,” or “It looks right to lead diamonds through dummy,” you're being lazy. If the line of play you are about to embark on is truly correct, it shouldn't be hard to construct one specific hand where it is right. Many errors, particularly on defense, would be avoided if players would do just that much. It's difficult at first. But if you are disciplined about doing it, you will find yourself getting faster.
Jan. 9, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Thanks, Danny. I was about to comment about that auction, but I see you beat me to it. While the examples in the original Cooperative Pass article (https://sites.google.com/site/psmartinsite/Home/bridge-articles/the-cooperative-pass">https://sites.google.com/site/psmartinsite/Home/bridge-articles/the-cooperative-pass) were all at the two-level, I did not mean to imply that the cooperative pass can't occur at the one level. It is simply rarer because of the availability of a one notrump response. But if one notrump, a negative double, and a diamond raise are all inappropriate for one reason or another, I would certainly have no qualms about passing one spade with a moderate hand.
Jan. 6, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
(1) Yes, two notrump in that auction is a puppet to three clubs, presumably with the intention to sign off in a minor or invite with three spades.
(2) The link I supplied above goes into detail about why I play double as take-out and how I handle the subsequent auction. The short answer is: the one-notrump doubler reopens with a doubleton in the bid suit but not with length. If we both have length in the suit they run to, they sometimes escape. I'm willing to give up penalizing them when their trumps are three-three (in which case we may not have a substantial penalty anyway) in order to get the play-or-defend decision correct more often.
(3) The teacher of that class must have been confused. After one notrump and an overcall, penalty doubles were virtually universal well before the invention of lebensohl. The purpose of lebensohl was to allow you to make natural forcing bids. Before lebensohl, it was customary to play all new suits by responder as non-forcing and to use the cue-bid as the only force. Two notrump was natural but a shade lighter than normal. Opener tended to give responder leeway and seldom went on to three. This competitive two notrump had limited usefulness, so giving it up to give responder two ways to bid a suit was a sensible trade-off.
Jan. 2, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Jess: No, since I play negative doubles after one notrump-overcall (see http://sites.google.com/site/psmartinsite/Home/bridge-articles/countering-notrump-interference">sites.google.com/site/psmartinsite/Home/bridge-articles/countering-notrump-interference), I would play 1N-X-2H-X as negative also. If responder bids over the double, we simply pretend partner opened 1NT, and all the same agreements apply. I'm not sure I understand your statement that playing lebensohl “clarifies that a double would be penalty.” One can play lebensohl and negative doubles or one can play lebensohl and penalty doubles. Lebensohl is perfectly consistent with either agreement.
Jan. 1, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Yes, I play that lebensohl is on. You essentially have the same problems in this auction as you do if partner opens one notrump and your RHO overcalls with a natural two hearts. While there are some tactical differences (it is desirable to keep opener on lead if possible, for example), I don't think they are sufficient to merit the memory strain of playing different methods. So, whatever my methods are after one notrump–two hearts, I play the same thing here. I also play the same methods after one notrump–penalty double–two hearts by the way. For the same reasons.
Jan. 1, 2012
You are ignoring the author of this comment. Click to temporarily show the comment.
Aha! I guessed correctly who this comment from this from. (My email notifications list the comment but not the commenter.)

Yes, your solution is superior to mine. Not only does it allow you to delay your decision about the lie of the heart suit, it avoids the crossruff variation, which fails on best defense if East has the club eight. So why didn't I see that? I suppose it took me so long to think of not ruffing two hearts that I wore myself out, and I gave insufficient thought to the best way to reach my hand. My instinct said to ruff a spade to force East to discard, and I didn't stop to question my instinct.
Dec. 26, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
The most complete discussion of third-and-lowest leads is in Rosler & Rubens' book Journalist Leads. It's worth reading, because there are a few corollaries that you will not find discussed elsewhere. As far as 3/5 vs. 3/low, I've always assumed that the practice of leading 5th from a six-card suit arose simply from confusion. The original Journalist proposal was “3rd from even, low from odd,” the idea being to distinguish parity as quickly as possible. Rather than say “3rd from even, low from odd,” people starting saying “3rd and 5th” as a shortcut. Some who didn't understand the point of the leads but wanted to accommodate their partners (“Want to play third and fifth leads?” “Uh… well… Alright.”) starting leading 5th best from six simply as a result of this shorthand description. Viral effects ensued. We have since learned to say “3rd and lowest” instead “3rd and 5th,” but it was too late.
Dec. 14, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Let's assume hearts are four-two, since either line works if they are three-three. Going for the ruff gives you one chance: the hand with long hearts has long spades. Going for the squeeze gives you two chances: the hand with long hearts has the club king or the hand with long hearts has the sole diamond stopper. In addition, the squeeze does not require four-two hearts. It works if hearts are five-one as well.
Dec. 6, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
I agree. It doesn't follow that responder needs more spades than hearts. But I don't think opener can assume he doesn't either. And, yes, I too prefer to play RKB for two suits (when the opposite hand is known to be balanced), which would make this hand much easier. My point was simply that you can sometimes find out what you need to know by lying, by switching from “describe” mode to “discover” mode. Joe Grue's recent “splinter” with Kx of diamonds falls into this category. If you're willing to be creative, there are all kinds of things you can do to ferret out the information you need. Many players get stuck in “describe” mode and don't even consider solutions like this. Lowenthal was much better at this sort of thing than I am. I wish I had kept better records when we we're playing. I'm sure I would have lots of amusing examples to offer you if I had.
Dec. 6, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Since I like to start on a positive note, let me say that I completely agree with the final pass. Now let's look at the rest of the auction.

Two clubs: You state that the room saved by opening 1 rather than 2 is an illusion, since, in either case, opener will be showing his second suit at the three level. This is true so far as opener's bids are concerned. But opening 2 does prevent responder from making descriptive, limiting bids. After 1-2, responder's hand is sufficiently defined that I doubt there is any serious danger of reaching a slam. On the 2 auction, responder had no chance to make a limit bid at any point, which is precisely why the auction was so difficult.

I don't dispute there is a danger that 1 will end the auction. But holding a two-suiter minimizes that danger somewhat. And 2 auctions are so awkward in general that I think it is worth the risk.

Two diamonds: The idea that you don't need to limit your hand so long as you keep making forcing bids is a fallacy that permeates Eastern Science Fiction. And the meaningless 2 response to 2 is one of those permeations. If 2 were a genuine negative, denying as much as an ace and a king, opener could give up on a slam immediately. Yes, there are some patterns where you can make slam opposite just the c!A. But to fish for those hands is probably a bad idea.

Two hearts–two spades: If I were assigning blame You-Be-The-Judge style, I would have to assign 10% of the blame to Eric Kokish for inventing this screwy convention in the first place. There is some gain in giving opener two ways to rebid 2NT. But the cost of depriving responder of a meaningful rebid over two hearts is enormous. If 2 is natural, responder has enough ways to raise hearts to make even Marty Bergen happy. He can bid 3 or 4. He can splinter. He can bid 2NT (or 3 second negative) followed by a jump to 4. Further, after 2NT or 3 and a 3 rebid by opener, responder can show support via a cue-bid. After 2–2–3, responder has only one way to raise hearts below game: 4. (3 doesn't count, since it could be a mere preference.)

Three diamonds: OK. I exaggerated in my opening remark. I agree with this bid, too.

Three hearts: Since 3 doesn't promise real support, it cannot promise slam interest. Thus, even if you normally play fast arrival, you can't play it here. 3 must be either a bad hand or a hand willing to drive past game. 4 must show some slam interest but not enough slam interest to merit bidding past game unilaterally. This hand would appear to fall in the latter category, so responder should bid 4.

These are fuzzy criteria. A serious partnership should set out these criteria explicitly. If each partner knows precisely what responder needs to bid past game unilaterally, the auctions become much better defined. A possible approach would be to assign point-count values to slam-positive assets. For example, an ace or a key king is two points; a king or a key queen is one point; various useful features that might or might not be useful could count as half a point. Responder is expected to drive to slam with four points and to bid past game with three. This hand would count as two points: one for the club king, one for everything else combined (queen-jack of spades, four trumps, ruffing value in partner's second suit). So it is not worth driving past game but is worth an aggressive move below game (hence 4). Obviously this is just a starting point. Such a scheme would need considerable fleshing out. But I think it would be well worth the effort. The more explicit you can be in defining what various slam moves require, the more comfortable your slam auctions will be.

Four diamonds: Even without any explicit agreements, I think opener should count on responder to bid past game with an ace and a king. To bid anything other than 4 at this point should suggest opener is willing to play slam opposite less. Further, even if opener were to make a slam move, 4 is a poor choice. The first slam move by the declaring hand should be like a help-suit game try. 4 should indicate that opener is looking for fitting diamond honors. If I judged opener's hand worth a slam move, I would bid 4, so partner would know that the club king is a working card.

Five hearts: Given responder bid only 3 on the previous round, and given opener's further try with 4, responder has a huge hand. 5 seems like the only aggressive move available to declarer. It would be nice if he could cue-bid the c!K, since whether that is a working card or not is probably key. He could afford to cue-bid the c!K if he knew that opener would expect him to drive to slam with the c!A and the diamond fit. Once again, clear agreements about what responder needs to drive to slam are necessary. In the above-mentioned point-count scheme, for example, one might say that, after the 4 try, the four trump/doubleton diamond combination (presumably exactly what partner is looking for) should count as two points. The fit plus the c!A would then be four points, a hand with which responder is expected to drive to slam. Thus the cue-bid must show a worse hand (either the c!A without the diamond fit or the c!K with the diamond fit). Whether this particular set of agreements makes sense or not is open to question. But some explicit agreements, even if they are not the best possible agreements, would go a long way toward making auctions like this meaningful.

Six hearts: This final call I don't understand. Surely partner would have cue-bid a black ace if he had one. Surely he would drive to slam with both aces.
Nov. 23, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Thanks, everyone, for these insightful answers. As some of you have surmised, this is an upcoming Gargoyle Chronicles deal. But this was not the auction that actually occurred. It is an auction I contemplated, and I was curious to see if a human expert would divine my intentions. Before I reveal the hand, I do have one follow-up question. What if you had two small-three small in majors? Over Smolen, you bid four hearts. Partner bids Blackwood and the auction proceeds as above. Now what you is your call and what do you think partner has?

One notrump is 12-14, by the way. I intended to “alert\” that, but I see that I didn't. Sorry.
Nov. 19, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
I tried to cast a vote for 3, but it doesn't seem to be taking. I'm not sure what ‘3rd bid’ and ‘4th bid’ means. In any event, I don't see any alternative to 3. 3 should show a partial stopper; you can't bid it with three small. A preference to partner's first suit with a doubleton is the customary punt in many auctions, this one included. Partner is not entitled to assume that you are 6-3 in the majors. If you are and chose to rebid your suit in preference to raising on the previous round, it is up to you to get that message across, either by bidding 4 now or by bidding 3 and taking aggressive action later.
Nov. 17, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Penalties. What more does North need to know?
Nov. 11, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
You're right that two spades can be defeated two tricks. I had thought that declarer could get out for down one by playing hearts early, but I was wrong. Say the defense starts clubs. Declarer ducks the first one, wins the second, and plays ace and a heart. If South continues clubs, declarer will get out for down one. He must abandon the tap and switch to a trump. If South hops with the ace and plays another heart, North must play low, allowing his partner to ruff and draw dummy's last trump. Thanks for getting me to look at that contract more closely.
Nov. 2, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
I agree with this principle (and my partnership notes state this explicitly): When third hand wins the opening lead and cashes a side winner, the leader's card should be attitude, indicating whether he wants his partner to shift back to the suit he led or not.

But a ruff is only one of the reasons he might want his partner to shift back. Most of the time, if a ruff is in question, the alternative to a singleton is a small doubleton, not honor third. In general, defenders tend to worry about one-card ambiguities in length and assume that partner can work out two-card ambiguities from other clues. So I suspect it didn't even occur to Wijs that his partner might worry that he had a singleton club.

It is possible, however, that Muller wants to know whether his partner has the club queen or not. Suppose Muller had KJ93 of clubs, for example. It's possible from his point of view for his partner to have Q74. It's also possible for him to have 8742. In the former case, it might be necessary for him to switch back to clubs (Qxxxxx QJx x 862). In the latter, it might be necessary for him to cash the other diamond (Qxxxxx QJx xx Q6).
Oct. 29, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Yes, Larry did say that. But he was simplifying. Verne's original article suggested counting notrump as a six, seven, or eight-card fit depending on whether the other side (the side with the trump fit) had a doubleton, singleton, or void respectively as their shortest suit. A later study, I believe by Ginsberg, showed that Verne was correct to make this distinction but had understated total tricks by almost a full trick. So you should actually count notrump as seven, eight, or nine trumps depending on the other side's shortest suit. One isn't generally even tempted to compete over notrump with a balanced hand. So, to keep things simple, I just call notrump “eight trumps” and keep in mind the fact that a void (like a double fit) may cause total tricks to be higher.
Oct. 25, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
At your table, I would say partner was duly punished for his overcall. Still, I think you have to share some of the blame. I would judge that three notrump is going down any time partner has a diamond trick. While that's not enough of a chance to double, it's enough to defend. You can reach this same conclusion by applying the LAW. Notrump counts as an eight-card fit. So, if partner has six clubs, there are 17 total trumps, and it is probably wrong to bid four over three.

Of course, it might be right to bid four clubs for constructive rather than competitive reasons. It's scary to hold a good hand in support of partner's suit and never clue him in to that fact. If you pass and are cold for five clubs, you're going to look pretty foolish. That's precisely why, when faced with a choice of introducing your suit or raising partner in a competitive in a competitive auction, raising partner is frequently the better choice. But I would say I dug my grave when I decided to bid two spades. While it's not impossible it's our hand for five clubs, it's sufficiently remote and I have a good enough chance for a plus score defending three notrump that I wouldn't want to risk going minus trying to get there.
Oct. 25, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Yes, one diamond isn't especially pre-emptive. But most partnerships bid better when they get to open the bidding, so there must be some gain in taking that away from them. And it generally can't hurt to let your partner know where 100% of your high cards are if possible. The opponents do rate to have a game, so it's hard to get into too much trouble. In fact, I don't recall the last time I got into trouble because we bid too much. It always seems to be partner's penalty doubles that bite you. So they do tend to work better when you aren't playing with someone who uses penalty doubles as a means of expressing disdain for the opponents' bidding.
Oct. 3, 2011
You are ignoring the author of this comment. Click to temporarily show the comment.
Concerning the Q86 problem: Restricted choice is a shortcut. As with most shortcuts, sometimes it helps you to come up with the right answer quickly. Sometimes it just muddies the waters. The surest way to solve problems like this is to forget about restricted choice and go back to the a priori possibilities. There are ten possible doubletons West can have: four with the queen and six without. If we always finesse regardless of what cards we see, we will be right six times out of ten. So the question becomes, can we formulate a strategy that will gain in more than six cases? Kit suggests the following strategy: Cash the ace. If West drops the ten or nine, play to the king. Otherwise, finesse. This strategy will always work in three cases: Q10, Q9, and 86. It will always fail in three cases: Q8, Q6, and 109. How it fares in the other four cases (108, 106, 98, and 96) depends on the card West plays. If he plays high, the strategy will fail; if he plays low, it will succeed. Since you must win at least 75% of the time against these four cases for this strategy to be superior to “Always finesse,” all West has to do is play high from these holdings at least 25% of the time to make your strategy a loser. There is no clear answer. You must make a judgment about how West will card and act accordingly. Personally, I think West will play high from these holdings at least 25% of the time, so I would stick with “Always finesse.”
Oct. 2, 2011
.

Bottom Home Top