Thursday, May 29, 2014

The Hitchhiker’s Guide - part 4: the Tiger wraps it up

I have just finished reading a really excellent book on defense called Tiger and Fly by Krzysztof Martens. There are two parts, each with examples of the style of defensive play of one of the eponymous characters. The bulk of the book is devoted to the Tiger. He (or she) is a bold defender with a predator’s eye for deception and camouflage. He will cunningly ambush declarer by drawing a false picture of the deal then pounce at the last moment just as declarer is, he thinks, about to wrap up a successful contract.

Naturally, I was keen to show my feline mettle at my next club game. Here is the setting. I was sitting East, dealer, at favorable vulnerability and holding the following nice collection against one of the best pairs in the room:

♠ AK97  QJT62  K9 ♣ 85.
I opened 1 and partner raised to 2. We play Bergen raises so this bid showed only three hearts or a very flat hand. Unfortunately, this being the first round, I had momentarily forgotten that and was thinking we were playing constructive raises. Anyway, this was passed around to my LHO who doubled. Partner passed and RHO bid 3♣ which was alerted as showing five clubs. I only have thirteen points, but I am certainly better than a minimum opener so I doubled. By agreement, this is cooperative asking partner’s opinion about defending or bidding on. I usually have one or two clubs and will have something in each of the unbid suits. Partner passed and it was my lead. I do think I was a little over-aggressive in this action (especially given that we play Bergen) but this story isn’t about the bidding. It’s about the defense.

It seemed natural to lead a high spade and get a look at the dummy. Dummy came down with:

♠ QT2  A95  J85 ♣ AJ72
and the first trick was made up of A-2-8-4. We play UDCA so the 8 is probably a suggestion to switch. Now is the time to heed the usual warning for defending doubled contracts: DON’T PANIC!

I panicked. I led a heart and that was that: -670 and obviously a bottom.

Tigers don’t panic however. They own their territory and they know it. People don’t wander in with their reopening doubles and expect to get out alive. The tiger laid down the K, continued to partner’s ace and received a diamond ruff. Five tricks in for the magic 200 and a clear top.

Why did the Tiger succeed when I failed? Mainly because he didn’t panic, of course. But he carefully analyzed the hand, too. It looks like we have three sure tricks in spades and diamonds. We need two more. Based on the auction, we can count on partner for two of the missing honors. What could these be, practically speaking? The K, the AQ and the ♣K are the relevant missing cards, although the Q is a slow trick and will likely come too late to do us much good. Assuming that declarer really does have five clubs, the ♣Q cannot be a factor. If partner has the A, we should be able to get a diamond ruff, providing our trumps don't get drawn first. Let’s consider what pairs of cards will get us the five tricks we need:
Down one if we cash second spade (2S, 2D, ruff); down two if we switch to K immediately (K scores).
Cannot defeat the contract (2S, 1H, 1D).
K ♣K
Down one provided we switch to hearts before the ♠Q is established (2S, 1H, 1D, 1C).
Down one provided we switch to K before losing the lead (2S, 2D, ruff); most probably down two.
A ♣K
Always down one (2S, 2D, 1C); down two if we switch to K before losing the lead (2S, 2D, 1C, ruff).
Q ♣K
Cannot defeat the contract (2S, 1D, 1C)
Ignoring the sequences which result in down two (or making), we are left with a decision. If partner has the A (three cases out of four), the contract is always going down provided that we switch to diamonds before losing the lead. The only time this would be fatal would be when partner has the two round-suit kings (the remaining case) without the diamond queen too. That's to say that any time partner has three of the four honors, the contract is going down anyway when we lay down the K.

What if partner has only one of the honors? Well, obviously, if it's the A and we switch to the king, it will be going down. As Martens points out, this is an application of Occam's Razor. If we can defeat the contract by assuming only one honor, then that's the way to go.

Wouldn't it have been better to cash the ♠K to see what would come next? Partner would play the trey, suit preference for diamonds, the lower-ranking suit. Was there any risk that declarer would show out on the second round of spades and draw trumps? If declarer only had a singleton spade, partner would have had five. Given that my double guaranteed at least three spades, we would then have had a double major suit fit of 16 cards. Far too much offense to risk passing for penalties.

But while it would be safe to cash the king and get the happy news that a diamond shift was right, in the one case where diamonds was wrong, we would now have given up the possibility of getting a heart trick. So, since the Tiger was planning to play on diamonds anyway, there was no need to continue spades.

In fact, partner didn't have two "cards." She had only the A! This, together with the spade jack, made up her entire hand. Was it right, then to pass the double? In the long run, probably not. In 3, most pairs were making 140 although they should be down one. But minus 50 would be preferable to minus 110 of course (and much better than minus 670.

But partner actually got us to the theoretical top spot: all set for 200. Now, if only she'd been playing with the Tiger!

Friday, May 23, 2014

The Hitchhiker's Guide to Bridge - part 3

Part two of this series added some thoughts on safely cashing out in doubled contracts. Today I want to add a useful principle which I call "The Principle of the Disappearing Trick." Not the kind of trick whereby rabbits disappear under handkerchiefs  but the situation where a cashing trick becomes uncashable. This hand just came up with a random partner from the internet about whom I know nothing at all. Matchpoints.

First of all, you might ask yourself if you should sit for the double. If North has just the two red aces, they might make 3 while we will be down only 1 in 4♣ (may be better to bid on). If North has the ♠A and one other trick, then 3 is down just 1 (100)  unless we can ruff some spades  but we will be making 4♣ (130) (again, may be better to bid on).

You (random partner) decide to pass the double. If you step through the first couple of tricks, you will find yourself on lead at trick 3 with the exhortation in the comment area "DON'T PANIC!"

Ask yourself "what can go away?" If partner has a natural trump trick, it can't go away. If he has a promotable trump trick it will go away if we don't act quickly enough. What about the side suits? If partner has the A, it's not going anywhere. What about the spades? Declarer presumably has five so they can't all be pitched on those diamonds. Is there any other way that our spade trick(s) (assuming we have some) are going away? Yes, if we try to promote partner's trumps by playing a third round of clubs. Declarer may be able to ruff high in hand and pitch one spade from the dummy. That defense, therefore will at best break even unless partner doesn't have either the ♠A or the ♠Q. But in that case, what is he doubling on? Five points?

So, let's look at this another way. First of all, there's an inference that declarer doesn't have much extra, either in points or distribution  he was content to let us play 3♣. Let's assume that North has the A and that it will cash (we don't know if West is 5512, 6412 or 5422 yet). If North has the ♠A too, we need to cash out, starting with the ♠K for down one at least. If he only has the ♠Q then he'll have to have a natural trump trick. Based on all of these considerations, the ♠K is by far the best continuation as, if all goes well, we can also get some spade ruffs!

Here's the complete hand. If you step through again to trick 3, you can now press the GIB button to see what will happen with various continuations. Notice how the ♠K will result in down 3 for a magnificent 500 which would have scored 15/17. A diamond is OK providing North now under-leads his ♠A to give you just one spade ruff. Anything but a club results in at least +100 for a 7.5/17 matchpoints. Needless to say, random partner continued with a club giving us a total goose egg.

Sunday, May 11, 2014

The Hitchhiker's Guide to Bridge - part 2

Part one of this series was published over a year ago. I don't recall if I had a sequel in mind at the time, but here is a totally new sequel. The key mantra, which every fan of the increasingly inappropriately named Hitchhiker's Guide trilogy (there are five parts), is of course DON'T PANIC!

Nowhere in bridge does this exhortation have more relevance than in defending a doubled part-score. When you're in this situation, two things are absolutely essential:
  • you must have confidence that you are in the right spot and that the contract is going down (usually at the part-score level, you only need the contract to go down one but sometimes, when the opponents are not vulnerable, you actually need them to go down two);
  • you must not panic.
I hope my partner from the following hand will forgive me for bringing this up. But it seems to me that both of us to some extent panicked on this hand which swung the entire matchpoints against us.

I have annotated some thoughts after the first trick. But basically, West knows almost exactly what my hand is  this is one of the hidden benefits of the cooperative double [there will be more hidden benefits in the next blog]. I don't have four spades (I would have raised to three) but I have stuff in the minors (and, based on the lead, the SK). In addition, East could have the DK, CQ, or a high heart, perhaps two of these. It's possible that declarer has a singleton in a minor, although the double would be somewhat questionable with a five-card suit on the side. One thing about playing doubled contracts at matchpoints is that it is much more like playing IMPs or Rubber bridge. You know precisely the number of tricks needed. In this case, exactly five. These will be two spades, the two minor suit aces, and either the DK or a high heart. Can any of these go away? Not the trump obviously, not the spade. A minor suit ace could disappear conceivably but if it's the club, there's not much we can do from our side. No, the only danger is losing a diamond trick if we have one.

Cashing out is not always as easy as it might be. If at this point, the DA is laid down and, in response to an encouraging signal, a small one continued, East may assume a doubleton diamond. Better perhaps is a low diamond, although there's a small chance that declarer will score a stiff K! I think the best approach is to play DA and a spade. Partner will not assume a doubleton diamond and will probably cash the DK and switch to a club (the ace will be the setting trick so there will be no need to play low). However, West continued a spade (not fatal but essentially wasting a tempo). I won the K and now had to switch to a minor suit, although I suppose another spade would not have been fatal either. If partner had the DA, why had he not played it at trick 2? Probably he had AQ of clubs, although that would leave him only a 10-count (possible for a third-seat opener I suppose). I led a small club and here is where the exhortation DON'T PANIC comes in. Trust me to have an honor in clubs and put in the T. This would automatically hold declarer to six hearts, and two black suit queens. But partner went up with the ace. Now, the tempo is critical  we are actually behind schedule in cashing our tricks and given time, declarer can get off a diamond. So now it is essential to start the DA. As you can see, this didn't happen so we went from +200 (a top) to -730 (a bottom).

This sort of thing has been the unfortunate result in many hands. I like to get those 200s so tend to make aggressive doubles. Against a doubled part-score contract, it doesn't take a genius to recognize that letting them make will be a bottom. So there is never any danger of making an even worse error that gives away an overtrick. You'd suffer the same bottom. So, always assume that the contract will go down with relatively normal play: building and cashing tricks. Trust that partner has his bid, because if he hasn't and they make overtricks, it can't get any worse.

Tuesday, May 6, 2014

Fun in the Smoky Mountains

Kim, CJ and I spent a very enjoyable week in Tennessee, mostly at the Mid-Atlantic Regional held in Gatlinburg, on the edge of the Great Smoky Mountains National Park. The main attraction was playing in the bracketed knockouts at the invitation of our friends Bruce Downing and Mark Conner. I had been once before nine years ago, Kim had never been, but Bruce and Mark are regulars at the tournament.

One of the things you can do at this tournament is play five sessions a day. In truth the "dinner-bell" Compact KO is only half a session, and the midnight KOs are whatever you can manage to stay in for. But Kim and CJ indeed played all five sessions that day! And CJ managed to add significantly to his masterpoint total by playing in I/N events -- although he was the only player in the whole tournament under 20, I think.

If winning matches and masterpoints in the company of friends was our only goal then we succeeded handsomely: we won 7 of 8 matches and earned almost 48 masterpoints. Still, we came away feeling that something wasn’t quite right. Our total master points came to just under 10,000. But our competition was, in most cases, significantly outgunned. Each team surely had to have at least one gold life master, or four who are very close. Yet, the matches were not as competitive as we are used to at home even in flight B events. Nobody seemed to be able to bid slams and they were fairly ignorant of alerting procedures, how to deal with unauthorized information and general bridge etiquette. In the match we lost (the finals of the first event), we had bid a very good slam. The opponents were not in the auction. The opening lead went straight to their partner's ace in one suit followed by a switch to another suit, ruffed (my RHO had seven but had not bid the suit). Other than that, the slam was odds-on to make. But they didn't bid it at the other table of course. There were other bad boards for us too as we all made mistakes. The first event where we played in the sixth bracket of 21, offered significantly tougher competition than the second event (fifth bracket of 12). In the latter event, we felt a little bit bad for raining on their parade. In the semi-finals we found ourselves up by 70 imps at the half and our opponents decided to withdraw [how had they managed to beat two other teams?]

Possible explanations for this disparity of skills are (1) the tough competition in New England; (2) the fact that we usually play flight A events at home rather than collecting points in B flights; (3) simple randomness at work (a sample of two events, eight teams is hardly statistically significant); (4) a space-time warp. Or perhaps it's simply that, as Harold Feldheim supposedly has said, "as you get better, you won't notice it. Instead, you will think that other players are worse."

One of my favorite hands came in the final (winning) match. Kim has been a little skeptical about the efficacy of fit-showing jumps. It’s true they don’t come up very often but when they do…

All white, my hand was something like Qx xxx AQTxx ♣AJx. Dealer on my right bid 1 and I made a questionable overcall of 2. LHO bid 2 and Kim surprised me with 4, showing a diamond fit, a good club suit and a willingness to be at the four level. Thinking that her clubs might easily prove better than my ragged diamonds [not to mention Kim’s excellent declarer play], I raised to 5, which she wrapped up for 400 while our teammates were in 3 making four (11 imps). My judgment of strain proved fortuitous as my RHO had a void in hearts which would likely have been led against a diamond contract.

How do you get to be up 70 at the half? Let me quote from teammate Bruce Downing...

This board was a killer:

West opened 1D and I overcalled 1H. 1S from East and West (balked of the planned 2H reverse) ended the auction with 3NT. East passed slowly as well she might. It turns out not to matter which heart I lead. I can't stop her from getting 2 heart tricks. However we held it to 3 for a pedestrian imp result of 3NT making their way. Not so calm at the other table!!!

Kim also opened 1D but apparently the North hand does not meet the standards of their opponents for an overcall. North passed with my cards. 1S from Robin. Pass from South and 2H Reverse from Kim. Now double from North (lead directing? penalty? bizarre?) Whatever North's reasons for doubling 2H, the unfortunate woman had no idea with whom she was dealing. The next to bid would be Robin Hillyard who has spent many years refining theories of doubles. Having mastered the red card, he was ready to move on to the blue card. Redouble is Robin's bid. He likes his chances across from a red suit reverse. Pass from South (what can she do?) Pass from Kim (nice trust for partner) and it is back to North who has nowhere to go either. The contract is 2H**. Kim gave up only 3 tricks in the play, obviously playing her 4-2 fit brilliantly, and 2H made with 2 redoubled overtricks rang up +1040 for our side. Good times!

In truth, I probably shouldn't have redoubled with six spades. But when you get ahead in a match you can take a few liberties to see how things will work it. The hand always makes 9 tricks in hearts, 10 tricks in spades or notrump and 12 tricks in diamonds. So we beat par by only 120 (3 imps) but in practice it worked out as 12 imps.

We also did a lot of touristy stuff. We spent a day at Dollywood and I rode five (!) roller-coasters with CJ including the old scary Thunderhead (wooden coaster) and the sleek new Wild Eagle. There's also a kind of zip line on wheels and track called the Smoky Mountain Alpine Coaster which we enjoyed. The weather was good, the dogwoods were out in abundance. Good times indeed.