Sunday, October 05, 2008

When the Experts Are Wrong

You have to make a lead against 3NT after your RHO dealt at Both Vul and opened 1NT (15–), which was raised on your right to 3NT. You hold ♠ J97643  J3  J9862 ♣ —. What do you lead?

I picked this up from page 8 of the second issue of the youth bulletin from Beijing Mind Games. Defenders at both tables in the match discussed had led a diamond and been rewarded (the diamond lead was the only lead that would've succeeded). Why a diamond, you ask? Well:

Recently, I've been trying to train myself to lead better with the guidance of GIB. I have examined a lot of leads recommended by it and, though I can't honestly claim that I can predict reasonably well what GIB would recommend, I have learnt enough to be suspicious in cases like this. So I fed GIB this problem. Guess what? The correct lead is  J.

This is what is wrong with much bridge writing: denial of human fallibility. Ridiculous explanations are frequently concocted to justify opening leads. Do these people really not have doubts, or will expression of doubt hurt sales (or readership)? I don't want to go into too many details, but suffice it to say that I'm utterly convinced that a bulk of the theory of opening leads we learn is nonsense.

Let me give one example of the ridiculous explanations that I mentioned earlier. Matthew Granovetter in Murder at the Bridge Table strongly chastises the lead of the unsupported ace against suit contracts. In one example, he criticizes the lead of the ace from a doubleton in partner's bid suit (though partner is known to be weak) against a game contract. In the actual layout, the game is made purely because of the lead. Though is auction is a bit involved, I wanted to see what GIB had to say. Not surprisingly to me, GIB said that the lead of the ace was second-best, the best lead being a neutral trump lead from three small. Anyway, Granovetter then says that the lead of the unsupported ace against trump slams is in fact attractive, because—want to guess? Because when you have found one trick, it's easy to find another. It's the most ridiculous explanation I've ever heard, even if the thing recommended has merit (I don't know).

For the hand given here, GIB in fact thinks that the diamond lead is considerably worse than the lead of the jack of hearts. A small spade is the second-best choice, it says.

I wouldn't say that GIB is always right about leads. GIB does not do deception, and GIB does not do concealment of high-card strength and distribution from declarer. It also cannot understand many auctions. But for cases like this one, I think it would be foolish to disagree with it.

2 comments:

Ashok said...

This board was played in the under-26 category. Of the 18 times the board was played, on 12 tables West declared three notrumps (you are North in the post) and on 4 tables East declared 3N. The lead of a low diamond was popular (Guthi led it too). Only Susan Humphries (of New Zealand) led the jack of hearts.

Anonymous said...

I haven't read any such rules. Its not hard to reason that Spade lead is no good. But I'm surprised low diamond was a popular lead, I was considering a heart lead, bur rejected it in favor of small diamond. I didn't make note that the same thing happened at the other table.