The Tao of Gaming

Boardgames and lesser pursuits

Progress at the bridge table

The good news is I recognized a merrimac coup. The bad news is I recognized it too late. It could have been straight out of a textbook, too. Damn.

Advertisements

Written by taogaming

March 4, 2009 at 10:10 pm

Posted in Bridge

17 Responses

Subscribe to comments with RSS.

  1. Congrats. They don’t come up all that often, but the general idea of attacking the entries to a long suit comes up regularly. Merrimac Coups are only a tiny fraction of the general case.

    Interesting case—a couple of years ago, in a team game, I was able to diagnose that I had to lead a dry king against a contract to knock out the side entry to dummy’s long suit. It turned out to be a Merrimac and was needed to beat the contract. Push. Same bidding and same lead at the other table.

    JeffG

    March 5, 2009 at 12:30 pm

  2. Brian or Jeff, I’ve been looking at the wikipedia page and I don’t understand what makes the Merrimac coup special. I.e. what in particular makes it distinct from, as Jeff notes, the general case of plays that are designed to keep declarer from having an entry to a long suit? Is the key that you are leading a protected king into an ace?

    Lou

    March 6, 2009 at 12:59 am

  3. A Merrimac Coup usually costs a trick in the suit led. I have seen a variant, in which the Merrimac forced a duck by declarer, but would not have worked the next round (imagine Axx vs. Qxx), but allowed me to shift and set up enough tricks in a different suit, because the forced duck gave me a tempo. I don’t know if that’s really a Merrimac—it’s more like a Morton’s Fork Coup, but it had Merrimac elements. Anyway, the Merrimac really only prima facie costs a trick; for example with Ax vs. Qxx, if the king is under the queen, declarer is always due two club tricks, but he has to lose a trick to get them. So the defense sort of loses a trick along the way in the Merrimac.

    JeffG

    March 6, 2009 at 9:38 am

  4. Lou, I’ve actually not seen this situation referred to by name, but I’ve seen accounts of similar plays. What distinguishes this from the standard practice of knocking out an entry to dummy is that you have to lead an “unnecessarily” high card in order to make it work. So, as Jeff says, if you have an unsupported King and underlead it to knock out an Ace on the board, if declarer has the Queen he can just win the trick in his hand. If, however, the board holding is exactly Ax, leading the King will do the job, as ducking the trick will do defender no good (you win and just lead the suit again). This gives up a trick in the suit, but if the Ace is the only entry to a board with potential tricks, could well be the only way to defeat the contract.

    Larry Levy

    March 6, 2009 at 12:17 pm

  5. Lou, I’ve actually not seen this situation referred to by name, but I’ve seen accounts of similar plays. What distinguishes this from the standard practice of knocking out an entry to dummy is that you have to lead an “unnecessarily” high card in order to make it work. So, as Jeff says, if you have an unsupported King and underlead it to knock out an Ace on the board, if declarer has the Queen he can just win the trick in his hand. If, however, the board holding is exactly Ax, leading the King will do the job, as ducking the trick will do defender no good (you win and just lead the suit again). This gives up a trick in the suit, but if the Ace is the only entry to a board with potential tricks, could well be the only way to defeat the contract.

    Larry Levy

    March 6, 2009 at 12:17 pm

  6. I’m trying to understand the example from the linked article.

    I get how leading the king works to eliminate the board’s Ace (and only non diamond entry).

    After that, the board leads a diamond. If I take with the Ace, then South later leads a diamond and takes all the diamond tricks. If I duck, then I can take the second trick and the board is locked out. But this depends on the knowledge that the unseen diamonds are split 2/2 among South and West. If I dont guess correctly in that, then I either give up extra tricks, or dont prevent the diamonds from being taken. If the split was 1 for South / 3 for West, then I should take that first diamond. If 3/1, then I need to duck twice.

    Is it simply that this is the highet probability play for setting the contract?

    Alexfrog

    March 6, 2009 at 1:37 pm

  7. But this depends on the knowledge that the unseen diamonds are split 2/2 among South and West. If I dont guess correctly in that, then I either give up extra tricks, or dont prevent the diamonds from being taken. If the split was 1 for South / 3 for West, then I should take that first diamond. If 3/1, then I need to duck twice.

    That’s solved by defensive signalling. Standard signals work by having partner play high-low with an even number of cards in the suit, and low-high with an odd number. Usually, bidding can tell you how many cards declarer has to within one, so the only ambiguous case is if partner has a singleton (one card in the suit), and it’s not the lowest remaining (so declarer should hide the lowest if he has it by leading his middle card with three; it’s much harder for him to know which card to play if he has two, since he doesn’t know if the ace is doubleton or tripleton). Sometimes you still have to guess, but often there’s a big clue or two; for example, if you can tell that two diamond tricks are enough to fulfill the contract, you assume that declarer has at most two and take your ace accordingly. Sometimes you just have to let declarer go down fewer tricks than he might otherwise to ensure beating the contract.

    JeffG

    March 6, 2009 at 5:59 pm

  8. Lou, I’ve actually not seen this situation referred to by name, but I’ve seen accounts of similar plays.

    Merrimac Coup is a standard name for the play of leading an unsupported honor to knock out the entry to a long suit at the cost of a trick in the led suit. It’s a very old term; it predates Contract Bridge, maybe even Auction. (It can’t be older than 1898, of course.) The play was known and used earlier than that, during the days of Whist. I don’t know what it was called before then. Wikipedia says it was called a “Hobson’s Coup,” but I think they are mistaken. I’ve never heard it called that, and I have heard what is now called the “Morton’s Fork Coup” called the “Hobson’s Coup,” from “Hobson’s Choice.” Anyway, the Merrimac name is probably over 100 years old and is in universal use these days.

    A Deschapelles Coup is somewhat similar, except that instead of knocking out an entry to a hand (note that this does not have to be dummy, thought that’s most common), it is used to create an entry to partner’s hand. It’s also an old term from the days of Whist.

    JeffG

    March 6, 2009 at 6:14 pm

  9. That’s solved by defensive signalling.

    That’s how I was going to answer Alex, Jeff, but then I took a closer look at the problem. In that case, I really do think it comes down to an intelligent guess.

    The problem is that I (the defender) have to play before my partner does. So I duck the first trick, of course, and my partner plays the 6. Is this the beginning of a high-low or did my partner start with a singleton in diamonds? If it’s the former, I need to take the Ace on the second trick; if it’s the latter, I must hold up until the third trick.

    I’ll probably hold up twice, as the 6 is a relatively high card (although if the declarer plays the 2 on the first trick, as he should, there will be only one lower card that I can’t see) and it makes more sense to give up one too many tricks than three too many. But it’s possible that that second trick in diamonds could give declarer the contract. Switch the King and Jack of hearts in the example and declarer only needs two tricks in diamonds; my second holdup play gives it to him. I’m just not sure there’s enough information to figure anything else out. So I’d hope the 6 showed a doubleton and play accordingly, but I really can’t be sure it’s the winning play.

    Larry Levy

    March 6, 2009 at 8:52 pm

  10. That’s solved by defensive signalling.

    That’s how I was going to answer Alex, Jeff, but then I took a closer look at the problem. In that case, I really do think it comes down to an intelligent guess.

    The problem is that I (the defender) have to play before my partner does. So I duck the first trick, of course, and my partner plays the 6. Is this the beginning of a high-low or did my partner start with a singleton in diamonds? If it’s the former, I need to take the Ace on the second trick; if it’s the latter, I must hold up until the third trick.

    I’ll probably hold up twice, as the 6 is a relatively high card (although if the declarer plays the 2 on the first trick, as he should, there will be only one lower card that I can’t see) and it makes more sense to give up one too many tricks than three too many. But it’s possible that that second trick in diamonds could give declarer the contract. Switch the King and Jack of hearts in the example and declarer only needs two tricks in diamonds; my second holdup play gives it to him. I’m just not sure there’s enough information to figure anything else out. So I’d hope the 6 showed a doubleton and play accordingly, but I really can’t be sure it’s the winning play.

    Larry Levy

    March 6, 2009 at 8:52 pm

  11. That’s solved by defensive signalling.

    That’s how I was going to answer Alex, Jeff, but then I took a closer look at the problem. In that case, I really do think it comes down to an intelligent guess.

    The problem is that I (the defender) have to play before my partner does. So I duck the first trick, of course, and my partner plays the 6. Is this the beginning of a high-low or did my partner start with a singleton in diamonds? If it’s the former, I need to take the Ace on the second trick; if it’s the latter, I must hold up until the third trick.

    I’ll probably hold up twice, as the 6 is a relatively high card (although if the declarer plays the 2 on the first trick, as he should, there will be only one lower card that I can’t see) and it makes more sense to give up one too many tricks than three too many. But it’s possible that that second trick in diamonds could give declarer the contract. Switch the King and Jack of hearts in the example and declarer only needs two tricks in diamonds; my second holdup play gives it to him. I’m just not sure there’s enough information to figure anything else out. So I’d hope the 6 showed a doubleton and play accordingly, but I really can’t be sure it’s the winning play.

    Larry Levy

    March 6, 2009 at 8:52 pm

  12. That’s how I was going to answer Alex, Jeff, but then I took a closer look at the problem. In that case, I really do think it comes down to an intelligent guess.

    Sometimes it does. Sometimes a signal will be clear enough, or the odds will be strongly in your favor that the signal is what you think. Sometimes West has the DA, in which case, he’ll usually get to see two of his partner’s cards before he has to commit. Sometimes logic of the hand will prevail. And once in a while, it’ll come down to a guess, but you usually get odds.

    Switch the King and Jack of hearts in the example and declarer only needs two tricks in diamonds

    Switch the HK and HJ in the example and declarer has a strong NT. Assuming they are playing some naturalish system, you ought to know if declarer has 15 HCP or not. Or if he does, he has a singleton somewhere. Guess where.

    The raw cards can give you odds, too. For example, from East’s perspective there are four cards out there: the 7, 6, 3, and 2. His partner plays the 7 on the first round. He knows that it’s either a singleton or a doubleton. Of the six doubletons possible, three will play the seven. Of the four singletons possible, one will play the seven. Odds are pretty good that partner has two.

    If partner plays the 6, he has 63, 62, or stiff 6. 1/3 of the doubletons, 1/4 of the singletons. It’s pretty close, but other clues might be helpful. If he plays the 3, he either has 763, 32, or 3. 1/4 of tripletons, 1/6 of doubletons, and 1/4 of singletons. But declarer played a card; if he played the 2, you know partner doesn’t have a doubleton; if he played the 7 or 6, you know partner does not have three. If partner plays the 2, he has one of three tripletons (out of four), or one singleton out of four. That’s roughly 3-1 in favor of the tripleton. (Roughly—if declarer opened 1NT, it’s nearly 100% that partner has one. And you aren’t beating this unless it was a 12-14 NT, and probably not even then.)

    Yes, a non-beginner declarer will try to falsecard you into getting the guess wrong, but it’s not easy for him, because he doesn’t know what information he’s trying to conceal; he doesn’t know who has the DA or how the diamonds are divided.
    Sometimes, therefore, he’ll guess wrong and play a card that tells you what’s going on.

    All in all, you may still have a guess, but it’s not a straight 50-50 guess. A good defender will get it right between most and nearly all of the time.

    JeffG

    March 7, 2009 at 12:41 pm

  13. …Oh—for what it’s worth, a very good pair of defenders can get it right nearly all the time with a useful hack. They can play that their card in CLUBS helps with the count in diamonds. If you use that approach, West plays the C9 at Trick 2, and East has no problem at all in diamonds. A fair number of pairs actually play this way; the idea is that your defensive signal is what partner desperately needs to know. Note that this time, by the way, the C9 will cost a trick in clubs, but the combination of the C6 and the D6 makes it pretty likely that diamonds are 2-2. Sometimes West has both of the C98, and then East will know for certain what’s going on.

    Don’t expect cooperation this good, however, except at the highest level.

    JeffG

    March 7, 2009 at 12:49 pm

  14. …Oh—for what it’s worth, a very good pair of defenders can get it right nearly all the time with a useful hack. They can play that their card in CLUBS helps with the count in diamonds. If you use that approach, West plays the C9 at Trick 2, and East has no problem at all in diamonds. A fair number of pairs actually play this way; the idea is that your defensive signal is what partner desperately needs to know. Note that this time, by the way, the C9 will cost a trick in clubs, but the combination of the C6 and the D6 makes it pretty likely that diamonds are 2-2. Sometimes West has both of the C98, and then East will know for certain what’s going on.

    Don’t expect cooperation this good, however, except at the highest level.

    JeffG

    March 7, 2009 at 12:49 pm

  15. …Oh—for what it’s worth, a very good pair of defenders can get it right nearly all the time with a useful hack. They can play that their card in CLUBS helps with the count in diamonds. If you use that approach, West plays the C9 at Trick 2, and East has no problem at all in diamonds. A fair number of pairs actually play this way; the idea is that your defensive signal is what partner desperately needs to know. Note that this time, by the way, the C9 will cost a trick in clubs, but the combination of the C6 and the D6 makes it pretty likely that diamonds are 2-2. Sometimes West has both of the C98, and then East will know for certain what’s going on.

    Don’t expect cooperation this good, however, except at the highest level.

    JeffG

    March 7, 2009 at 12:49 pm

  16. Switch the HK and HJ in the example and declarer has a strong NT.

    That’s a good catch. I still use the old fashioned 16-18 pt 1NT, so I didn’t think East could tell if the cards were switched, but 15-17 is definitely more common for most pairs these days, so that should give East all the information he needs to make the proper play.

    The club signal is dicier, but still possible for an expert pair. However, I can see West thinking that he’d rather just signal in diamonds rather than possibly give up a trick in clubs.

    Great game, isn’t it?

    Larry Levy

    March 8, 2009 at 12:28 pm

  17. West will signal in clubs if and only if it won’t give up a trick. Some days he’s dealt good enough spots. Sometimes the diamond carding is clear. This was a tough one for standard carding. It’s possible that E/W were playing upside-down carding, but the situation is the same, assuming declarer concealed the D2. In any case, if someone is good enough to find the Merrimac Coup, he should be good enough to get the diamonds right!

    JeffG

    March 9, 2009 at 1:36 pm


Comments are closed.

%d bloggers like this: