Saturday, September 16, 2017

You can squeeze, too

One of the first things we learn about the playing of the cards in Bridge is how to take a finesse. If you lead towards your AQ combination and the person on the left has the K, you will win two tricks. Otherwise, one. Of course, there's a little more to finessing. There are also ruffing finesses, two-way finesses, intra-finesses, “obligatory” finesses, and so on.

The situation is similar with squeezes. What was that you said? Surely, squeezes are much harder than finesses? Well, they are a little bit harder but it really isn’t rocket science. The first bit of added complexity is that, whereas in the case of the finesse, where your LHO must follow suit if able, a squeeze operates on two, or sometimes three, suits at a time. In fact, when you lead the "squeeze card” to the squeeze trick, your victim will be out of that suit and have to discard. But that’s where the fun (for you) begins. If the squeeze is working, your victim will have to discard either a winner, a guard to a winner, or a card that he could have led to his partner’s winner. Either way, he loses out.

Note that I didn’t say “LHO” but victim. That’s because sometimes your victim is on your right, although that’s a bit harder to pull off because he will be discarding last, after everyone else. But that’s when you play your “ace in the hole,” so to speak. More on this later.

For now, let’s stick to squeezing LHO, because that’s the easiest situation—hence its name: “simple squeeze”. In the following situation, either the enemy trumps are all drawn or it’s a no-trump contract and you are in your hand. RHO doesn’t have any cards in the three critical suits (he can be playing with Happy Families cards for all you care). Here’s the layout at trick 11 when you play the 2, the squeeze card:

         A3
         2
 
         
54              
A               
—               
—               543
         2
         
         2
         2

According to the rules of bridge, LHO will have to play one of his cards before you play from dummy. His A is a winner, of course, and if he discards it, your lowly deuce will be promoted to master rank. What if instead he decides to pitch the 4? Now—and here’s the fun part—you pitch that little deuce of hearts because it’s no longer useful. When you cross to the A, with your deuce, your trey will now win the last trick. So, we just swindled LHO out of his useful cards simply by virtue of playing after he does. These cards which can be potentially promoted into tricks are called “threats” or, “menaces”—the word the Brits use to describe them. Note that LHO didn’t make a mistake—he had no legitimate choice because all his plays were going to give something away. That’s the essence of a squeeze! Sometimes you can gain a trick by a defender pitching the wrong card just because they don’t remember what’s out. That’s called a "pseudo-squeeze" and it's better than nothing (although don’t run a pseudo-squeeze against an expert—it’s a little bit insulting).

So, let’s look at this situation and see what really happened. There’s a reason that there are only three cards each in this layout. If there were any more, LHO would have a spare card to throw and you wouldn’t know which card to pitch from dummy. Basically, you have to be in the situation where you’ve lost all the tricks you are willing to lose but you don't want to claim yet because there’s a chance the squeeze will operate for an extra trick. You also had to have a (single-card) entry to dummy's two card holding. Your threats have to be positioned in the upper hand (the one that plays after your victim). And, finally, your victim has to be the one guarding the two suits where you have threats—he said to be busy in both suits.

From these conditions described in the previous paragraph, we get the famous BLUE law for squeezes: Busy-Lost-Upper-Entry. In this short article it’s impossible to go into all the details of why you need these four conditions. But deal out some cards yourself and try it. You’ll soon see what happens.

I will just add some observations on situations that are a little more complicated than what we just had (you knew it wasn’t really that simple, didn’t you?). You might have noticed the presence of the 2 in your hand in the situation above. It didn’t seem to play a part, certainly not a starring role. It’s a spare card. In this case, it could equally have been a small heart. And RHO’s cards didn’t all have to be clubs: they could have been any small cards as long as there weren’t two spades (the suit of your two-card threat). If the spare card happens to be in the upper hand, and you have a threat in your own hand, you have what’s called an “automatic” squeeze. The title doesn’t mean that it will play itself (although actually, if you’re paying attention, it will). What it means is that it operates against either defender. So, in the diagram above switch your 2  (the spare card) for dummy’s 2. Now, when you play the squeeze card from your hand, you know ahead of time that you’re going to pitch from dummy—the spare card. But you do have to look out in case someone (either player) discards the A because if that happens, and you've been paying close attention to the hearts, you now cash the 2 in your hand, and then cross to dummy’s A for your last trick. In other words, in this situation, you could swap all of LHO's cards for RHO's cards—in an automatic squeeze you don’t care who your victim is—lefty or righty.

What you can do with two suits, you can sometimes do with three. These situations are a bit more complicated, but not much. For one thing, their end positions (i.e. at the squeeze trick) will have more cards in each hand. Just as with the two-suit squeeze above, there are two major forms: the triple squeeze where only one player is busy in all three suits and the double squeeze where both players are busy (each of them in just two suits). The triple squeeze itself has at least two forms: in one, you can actually gain two tricks! Triple squeezes aren’t common but they are even more fun than a two-suit squeeze. Some double squeezes are easier to pull off than others. There are even ruffing squeezes. But, for these you need to do some reading. There are lots of good books on squeezes. The classic (and not too long) is Clyde E. Love's Bridge Squeezes Complete or, if you can find it, its predecessor: Squeeze Play in Bridge. Perhaps the ultimate is Hugh Kelsey's four volume set (now published in one very thick volume as Kelsey on Squeeze Play). But my favorite, with the inimical style of David Bird—and with lots and lots of examples—is Squeezes for Everyone.

But whether you want to read about squeezes in books or not, here is a pretty simple way to try for one of the simpler squeezes on your next hand. Don’t worry too much about preparing the ground (yes there are some techniques you can use but they are a bit more advanced). Just follow this checklist:

  1. Do you have exactly one loser? If you have more, try to lose the tricks you must lose early on, for example ducking from a suit where you have Axx opposite Kxxx. If you no losers, claim: don't torture your opponents.
  2. Are there any trumps out? If there are draw them. If you can't, or if you need the trumps yourself for a cross-ruff perhaps, then this is not a squeeze hand.
  3. Run all your winners—Yes, even your last trump—I know you'll feel naked without it, but it's essential.
  4. If your single threat (2 in the example above) has come good, cash it and claim. If not, use your entry to the other hand and hope that its companion will be good. If it isn't, then there was no squeeze but you've done your best. Note that in the example where we had Axx opposite Kxxx, and we ducked a trick early, if that suit splits 3-3, you will end up making the same number of tricks as everyone else. But if that suit was 4-2 and the one with the four had to guard another suit, Bingo! you just made an extra trick. Believe me that this will score a lot of matchpoints in your average club game.

I'm sure I'm going to regret publishing this article when you are making extra tricks against me! But you deserve to have more fun at bridge. Happy squeezing!

Monday, September 11, 2017

A leading question

The problem of which card to lead from a sequence or near-sequence of honors has been with us since the early days of whist. "Standard" leads are the top of any sequence, except that from AK, the K is led. At first glance, this may seem inconsistent but of course there was a good reason, historically speaking. In whist, there was always a trump suit so, typically, only one or two rounds of a suit would be cashing (it's unusual in a suit contract that you happen on the lead of a suit where all follow to three rounds). So, therefore, it was important--if you held the lead--that you don't give up a trick on the second round in case there was a useful third round. That's why the "come on" signal was developed to help you know whether to continue or to switch.

Originally, this was called a "peter" after "Blue Peter," the flag that told sailors in port that their ship was about to set sail. Failure to heed the Blue Peter would result in a charge of desertion. Not good. At bridge (or whist) the consequences might be expensive but you wouldn't (normally) lose your life over it. These days, in North America, we call it an echo or, less poetically, an attitude signal.

A minor diversion on a subject I think I've addressed here before: sequences. Bridge (the play) and whist are, to a large extent, a game of sequences and position. If you are the last to play to a trick, you have the luxury of being able to cover whatever is the highest card (assuming you have such a card). You don't have to worry too much about it. But if you are one of the early position players, especially leader and third-hand, you want to try and make the best use of your sequences if you have any. You don't normally lead unsupported honors, at least not unless you have good reason to think your partner might have a touching honor (thus giving our side a sequence).

This is why defense is so much harder than declarer play. Defenders cannot directly see where they might, between them, have a sequence. A suit might be distributed KJ53 opposite AQ42. If declarer has that suit, he knows for sure that there are four tricks to be had, however he plays it. But if the defenders have that suit, each partner may be reluctant to lead the suit for fear of giving away a trick. Just last week I gained two overtricks in a suit contract because my LHO was twice reluctant to lead from KJTx (his partner having AQxxx).

Returning to opening leads: if the K is led, partner knows, holding the A, that he always wants to encourage a continuation because the K promises the Q (or perhaps Kx). This guards against the so-called Bath coup (a whist term) where fourth hand has AJx of the suit and holds off the first trick. But what if opening leader's partner holds the J but not the A? If partner led K from KQ, then the card will form a sequence with leader's remaining card and, therefore, another lead in the suit is safe. But what if partner holds the AK and declarer has the Q. A continuation would now be unfortunate because the declarer has the Q for the third round. But as mentioned already, the rules were set up to keep us relatively safe for just two rounds of any suit. Recall that in whist, there is no declarer. The hand playing last to the opening lead may not have sufficient high cards ever to draw trumps and cash that queen.

The problem with this cozy little scheme is that, in bridge, we do have a declarer. That player, fourth to play on the opening lead, likely does have sufficient high cards and/or trumps to eventually make use of that queen on the third round of a suit.

So, various schemes over the years have tried to clarify opening lead sequences. One rather complex scheme was developed by Norwegian Helge Vinje and was later reported by the Bridge Journal in 1964. Such leads became known as "Journalist" leads. They tend to be more popular as leads against notrump contracts, although Vinje actually described two somewhat different schemes for trump and notrump contracts. There is much to be said for these lead conventions, but they are too complex for most casual partnerships.

A very common revision of "standard" leads is always to lead the highest of any sequence, even ace from AK. One problem of course is that when you want to lead an ace, you don't always have the king and partner may well go wrong (this is more typically true of higher-level contracts). It's common, especially in Europe, to agree that against suit contracts at the five-level or above, ace still asks for attitude (but now it's only the king that is good) and the king asks for count. There's also a perhaps minor problem that has to do with UI (unauthorized information) when you agree to play ace from AK. How long does it take you to select a lead from a hand with an AK combination in a side suit? About half a second. So, when you quickly lead an ace, your partner knows that you have the king to back it up. But when you think for thirty seconds and then lead an ace, partner knows that you don't have the king. He's not authorized to know this and so must go out of his way to avoid taking advantage of such information. Is this really a big problem in reality? I don't know. But experts like to avoid giving their partners any kind of problem so that brings us to one more opening lead convention: Rusinow Leads.

Originally banned by the ACBL, they were adopted by Blue Team greats Belladonna and Avarelli and sometimes therefore called "Roman Leads." The principle is simple: you lead the second of touching honors. Typically, it is agreed to play this only against suit contracts and not in partner's suit. When asked about honor leads, the defenders will say "shows the honor above or shortness." Usually, the ten is considered the lowest "honor" for this purpose so that the 9 could be from T9(x) or it could simply be the 9 from 9(x). There will always be a little ambiguity with any such convention but isn't it much better to have the ambiguity over the 9 than over the ace?

It's easy to forget that you're playing Rusinow. We have all done it. Even though I've played Rusinow (or Vinje) leads perhaps more than any other lead convention, I've still forgotten it occasionally, sometimes with disastrous consequences. But, if you are part of a regular partnership that wants to improve, Rusinow leads are a very sensible and simple lead convention.

Incidentally, whether or not you play Rusinow yourself, you should be aware that some people play it. If you are declarer and you need to know the location of a particular high card (that's to say most contracts), you may need to know what their lead conventions are. You can simply ask "leads and carding?" or you can pick up their convention card and look for yourself. You probably shouldn't ask "is that lead standard or Rusinow?" as that tends to give away the fact that you don't have one of the adjacent honors.