New from MPP Part 1

As we approach the holidays, this is one of the busiest times of the year for us at MPP, although probably not for the reasons you think.  We’re a publisher, not a bookseller, and people do insist on interrupting our real work to buy books from us.  Some of them even come to the office to pick them up for gifts!  Meanwhile, we’re trying to get three new books off to press, which is what I really want to write about here.

Anyone following Linda’s blog knows about the new edition of Love’s Bridge Squeezes Complete.  We ‘re all very proud of this one; sometimes you just have the feeling you’re working on something important, and this is one of those occasions.  I know Linda feels that way, and I suspect Julian Pottage (who helped enormously) does too.  Our second book on our Spring list, though, may also turn out to be a major contribution to the literature:  Bob MacKinnon’s Bridge, Probability and Information.

I constantly see people misuse numbers and statistics, which irritates me.  There’s an old saying about there being ‘lies, damned lies, and statistics’ (something ‘climate change’ fanatics might do well to remember), and it’s a pleasure to see a professional explain what conclusions you can and cannot legitimately reach from a particular collection of data.  There was an article published about 4 months ago in which statistics were relied upon to support some very dubious conclusions about the usefulness of the Multi two diamonds convention.   In a happy coincidence, Bob describes in the book the fallacies in the arguments that were used, using a different example, the Flannery two diamonds.

Bridge is not chess — it’s a game of inference rather than complete information, so whether we like it or not, there are certain basic mathematical concepts and numbers that we have to know in order to play it well.  I’ve been kicking around some of these ideas with Bob for a few years now, ever since he wrote Samurai Bridge for us.  And if you read Bob’s blog, you’ve had a glimpse of some of what he talks about in this book.  Bob is a retired mathematician who happens to be very literate, very well-read, and an excellent writer.  What I told him when we started the project was that it had to be aimed at bridge players, not mathematicians, and I think in the end he’s pulled it off.  When Stephen Hawking wrote A Brief History of Time, the story goes that his editor warned him he would lose 25% of his readers with every equation he put in the book.  In the end, he insisted on only one, e=mc². I think Bob has managed to outdo Hawking, in that there isn’t a single equation in his book — the ideas are the thing.  What we have left, after all the heavy math was eliminated, is a fascinating and readable account of the ideas of probability and information theory, and a host of practical applications of them to bridge.

The book starts by tracing the history of the theory of probability, with the young Blaise Pascal throwing dice and pondering the results.  Quickly, though, we get to the bridge table, where after briefly looking at the kind of numbers that any bridge player knows (the 3-2 break is 68%, for example), we are gently blown out of the water.  Those simple numbers we all know and love, it seems, merely represent an approximation at the start of each deal, when we have little information.  As bids are made and cards are played, the amount of information we have changes — and so do the odds.  We may know this instinctively (for example, when someone preempts we tend to suspect that other suits may not be breaking well), but we don’t know how to apply it in any kind of quantitative sense.  But of course, the known splits in one suit do affect the probabilities of splits in another — and we all make the mistake of looking at these things in isolation, when we should not.

Once we get into the habit of looking at suit splits holistically, many interesting conclusions can be drawn, with obvious practical applications.  Let me give one example.   We’re looking to pick up trumps, missing four of them to the queen; normally we would play for the drop.  But what if RHO has preempted, so diamonds are likely to be 2-7, say.  That leaves LHO with 5 more vacant places in his hand for the queen.  So now, you say, the odds must favour a finesse.  True, but by how much?  At what point, in terms of an imbalance in Vacant Places, do we switch from playing for the drop to finessing?  Do you see what I mean by practical applications?

There’s much more in this book, though.  Having laid the groundwork, Bob goes on to discuss a host of (to me at least) fascinating bridge issues, these among them:

• The idea of visualizing ‘sides’, the complete combined holdings of both defenders, and not just the splits in individual suits
• How a known split in one suit affects the odds in another
• Empirical rules to help make decisions when there is incomplete information or the situation is too complex to analyze accurately
• How a priori probabilities (the ones with which we are all familiar) change with each card played
• How an imbalance of vacant places in the defenders’ hands affects the odds – and when to change your line of play as a result
• The ‘Monty Hall Problem’ and its bridge cousin, Restricted Choice
• HCP distribution – what partner’s bidding tells you about where his high cards are
• Information versus frequency: the trade-off in choosing conventions
• Losing Trick Count – does it work, and if so, why?
• Probability, statistics and the LAW of Total Tricks – how far can you rely on it?
• Cost versus gain: information theory as applied to bidding systems
• Using statistics to help you choose a bidding system that works for your style of play

This is a book I’ve wanted to publish for some time; I hope readers out there are as enthusiastic as I am about it.