Slam Bidding Mindset

I have mentioned in many conversations with friends and students that I had to write a book on slam bidding. I will write a series about it here on the blog - comments are highly appreciated!

Introduction

“Should we bid this?” and “How could we avoid bidding it!” are how many players approach me after they have gone wrong in a slam deal. Double dummy, slams make up for around one out of every eight deals. This is a proportion that is sure to surprise most players, even the most experienced ones, since our lived experience of slam deals is surely rarer. Half of those deals belong to the opponents; and many double dummy slams are unbiddable (requiring many finesses, or a guess of a short honor offside, etc.).

Luckily, we are not concerned about double dummy here. My own estimate of the expected number of slam deals in a typical matchpoint session (21-24 boards) is from two to three. Not very far from 1/8, but remember that “a slam deal” is not a hand in which you bid and make a slam; from the viewpoint of the bidding partnership, it is any hand in which the question of whether or not to bid a slam arises. In some cases the answer to the question will be negative, but this is still a slam deal, particularly if the exploration takes us to the 5-level (with a major-suit fit). In this series we are concerned with slam auctions rather than double dummy slams.

Since bridge is a probabilistic game, absolute accuracy is impossible through legal means, and the best that we can hope for is to reach theoretically correct contracts. This means that we want to bid slams with a chance of making higher than X% (in which X is defined by many circumstances, including the most well-known: method of scoring and vulnerability, but also strength of the opponents, state of the match considerations, etc.). And we also want to avoid slams in which the chance of making is lower than Y%.

Note that X and Y do not need to coincide! There is a gray zone between them, in which both decisions (to bid slam or to not bid it) are reasonable, especially if we consider that the auction A can have revealed information that auction B did not reveal, and vice versa. To give a very simple example, one auction may suggest one opening lead to the opponents, and another auction may suggest another, and this may have a very significant impact on the chances of the slam making or going down.

So, the goal of a bidding partnership is to develop the best mindset for solving the problem of any given slam deal with as high a rate of success as possible. This includes agreeing on efficient tools for slam exploration.

Hopefully, you and your partner will be better equipped to do this after reading this series of articles.

A note on grand slams

The decision to bid a grand slam is a momentous one. You jeopardize the large bonus of the small slam (and of game) to reach for the big prize. I am not planning on dealing with this issue throughout the series of articles. My expectation is that the mindset being described (aimed at solving the problem of small slam decisions) will also be of help in solving the rarer problem of grand slam decisions. Once you are looking for the right data to determine whether to bid a small slam, to decide on whether to go for the grand is basically an extension of the same methods. My general advice in this area is not unlike that of many other experts: judge the field you are playing in, try to count your tricks before bidding a grand slam, and keep in mind that the other table(s) may be having trouble solving the small slam issue. Don’t be too greedy.

A typology of slams

The human mind works by categorization, and slam bidding is no exception. Although every categorization has borderline cases, it is useful to think of slams coming in three kinds:

•             High card power

•             Ruffing power

•             Running tricks

Let us examine each kind in turn. Keep in mind that different tools will be used for deciding what to bid with each kind of slam.

High card power – If you look at typical high card guidelines for slam bidding with two balanced hands, you find that most authorities suggest 33 hcp as a good benchmark. If you can find a 4-4 fit (which often provides a ruffing trick), 31-32 hcp may be ok if the hands fit well. A trick from length (if you have a 5-card suit) can also compensate for 1 or 2 missing hcp. This, obviously, assumes that the opponents cannot cash 2 tricks immediately.

As experienced players know, the most common methods for bidding slams include, in many hands (though not all), the determination that the two hands have sufficient controls in every suit to guarantee that the opponents cannot cash 2 tricks in a suit, or 2 Aces.

A typical example of this family of slams was created randomly by asking a bridge simulation program to deal 2 balanced hands with 33 hcp:

North-South have 33 hcp and the potential to develop 12 tricks (4 clubs, 4 diamonds, 3 spades and a heart) even if they misguess the Queen of spades. Unfortunately for them, East-West have the potential to develop 2 tricks (one heart and one spade) if they can find the heart lead, which is easy from East, and perhaps also from West when the auction allows for a cheap 1♥️ overcall. North-South will have to guess spades after this start (and should probably get it right, but we will not discuss the play here).

All in all, a worthwhile slam, laydown against the wrong lead and (at worst) 50% if they get to the best lead. In many cases the heart honors will be divided and so it won’t be so easy to find that lead. Looking at only the NS cards, we would like to be in slam, which is the theoretically correct contract.

Playing around with these cards we see that some hands will be easier to make, some will be harder, sometimes we will find a 4-4 fit and be able to exploit it usefully, etc. In general, 33 hcp is good enough for us to bet that slam will be worthwhile, so much so that purely quantitative bidding (not stopping to look for controls in all suits) is recommended with this kind of hand. When partner opens a 15-17 NT and we have 17-18 and a balanced hand, we simply bid 6NT, unless we have very sophisticated mechanisms to look for 7, and even so it might be wiser to just forget about it unless you have other positive features (e.g. a 5-card suit).

Ruffing tricks - The other side of the coin is the slam that is made primarily out of ruffing tricks. One of the most famous examples of the type is the Moonraker hand (from the James Bond novel of the same name), in which Bond tricks the villain into losing a lot of money by bidding a grand slam with only 8 (!) high card points:

In 7♣️, South can establish diamonds while finessing in clubs and makes 13 tricks without any trouble. Well, at least when he knows the hands, as in the novel. In real life playing the club suit would not be so simple.

This extreme example is fun but also instructive if we try to imagine how the bidding would have gone. Note that E-W can make slam in a major (played by West), and that the par is 7♠️ or 7♥️, down one. What this shows is that slams of this family (lots of ruffing tricks, not a lot of high card power) are always accompanied by very competitive auctions. The opponents, who own the high cards which are missing in your hungry slam – as well as a very good fit of their own (a good rule of thumb is, the better our two hands fit, the better their two hands fit) – are not going to keep quiet during the auction, and the most likely scenario is one in which the bidding climbs very high very quickly and nobody is quite sure of what is going on.

That said, most slams are not made out of pure high card strength (not even in notrump! Length counts for many tricks there), or out of pure ruffing power. We must be prepared for a mix of weights of each factor in everyday deals. Slams on 8 hcp happen only on staged deals, but slams in the 24-28 range are quite common. They require a good fit, to compensate for the high cards that have been taken out (compared to the typical balanced 33 hcp hand), and good controls too, but we have all seen hands such as this:

Each individual hand here is completely within the bounds of our experience. All of us have seen them hundreds of times in our bridge lives. North-South have 26 hcp, divided 14-12. They have a 10-card fit, surely above average, but nothing outrageous, in a 5-5 division. But despite the apparently mundane character of the deal, North-South have 12 absolutely solid tricks (5 trumps, 3 spade ruffs, 3 diamonds, 1 club), regardless of the opening lead. Even a club lead would avail nothing; declarer would simply unblock the ♦️K, draw trumps ending in North, discard his clubs and lose a spade, setting up the ensuing crossruff.

In subsequent chapters we will investigate this deal more closely, to train our minds to look for the decisive characteristics of the deal that would enable us to diagnose (without peeking at partner’s hand) whether a slam would be a good bet or not. Before we go there, we must look at the third kind of slam.

Running tricks - Ruffing tricks are not the only way in which a deficit of high card power can be overcome; we can also substitute long cards for high cards. Of course, there must be some conditions, most importantly the ever-present requirement that the opponents cannot cash 2 quick tricks (slam bidding would be so much easier if the declarer side also had the opening lead advantage…).

The typical example (constructed randomly):

North-South, once again, have only 26 hcp, but they have 12 top tricks playing in one of their suits, and they have enough controls on the side (a singleton and an Ace) to ensure that the opponents will not cash 2 quick tricks.

This kind of slam is very difficult to diagnose in real time because we can never be sure that partner’s limited high cards are in the right place. If the ♥️Q were any other Queen (spades or diamonds) the slam would become much worse. And since our bidding methods are geared towards finding the best game and bidding it without a big exchange of information when our combined strength is in that range, we often stop bidding (at the game level) before any slam move is ever attempted.

The possibility of one of the two other kinds of slam is usually perceived in the early bidding very quickly; either we diagnose that we have extra high cards beyond the minimum requirement for game, enabling us to exchange slam-going information secure in the knowledge that our extras will bolster our chances even if we stop short of slam (in other words, we are not too concerned about information leakage when we are in the 29-30 range), or we diagnose the excellent fit (extra trumps, shortness) at a sufficiently low level to enable slam investigation. The “running tricks” slam almost always requires some systemic luck. But it is important for us to have this category in our minds to identify the lucky situations.

This series will be continued at an approximate rate of one article per week. If you are enjoying it, please comment and share the word with your friends.