home

Build your own algorithm

Jean-Paul Meyer ( France )


Born in 1936, JEAN-PAUL MEYER of Paris is a professional bridge writer. The best moment of his playing career was in 1987 when, in partnership with Gerard Le Royer, he won the European Pairs. He is the editor of Le Bridgeur and columnist of L'Expres.

 

Quite often I meet bridge players who ask me why chess computers are so efficient compared with bridge computers. There are plenty of reasons; you probably know most of them and this is not the place for such arguments. But one of the important reasons is that programmers in chess have succeeded in computing several steps in advance by building up what is called an 'algorithm'.

Until machines devoted to bridge improve  their performance (Zia's 1 million bet that computer could beat him still looks safe), you had better count on your own brain to win at bridge, and it would be a good idea to build your own algorithms. If it works perfectly (unfortunately there is no such thing as perfection) you should have no problem playing and defending efficiently.

 

To be more pragmatic, let me now give you my BOLS bridge tip:

When playing, either as declarer or in defence, your first concern should be
to foresee what will happen two or three, or some Top even eleven, tricks later.

Many articles have been written to emphasize the importance of making a plan at trick one, but often a new plan has to be launched in the course of play. Your own algorithm should warn you what is going to happen before it is too late.

Let us see one example:

5    

 Q 7 5 4

 

 

 Q J 10

 

 

 A J 6 5 3

 

 

 

N

 

 

 

 

 Q 10 9 4 3

 

W                          E

 K J

 

 

 A 4

 

 

 K 10 9 4

W

N

E

S

 

 

1♠

Pass

1NT

Pass

2♣

2

Pass

2♠

Pass

2NT

Pass

4

All

Pass

Partner leads a trump to your jack and declarer's ace. Now South plays the two of diamonds, seven from West. You take with the ace. You cash your king of hearts. All follow. What is going on?

Your opening bid showed a five-card suit. Partner has shown no support so should have at most two spades; that gives South a 5-5 major two-suiter. Partner showed an even number of diamonds, so you can construct the full deal: 

  East Dealer 5    

 

N-S Game

 Q 7 5 4

 

 

 

 

 Q J 10

 

 

 

 

 A J 6 5 3

 

 

 

 

 

N

 

 

K 8     Q 10 9 4 3

 10 3

 

W                          E

 K J

 K 9 7 6 5 3

 

 

 A 4

 Q 8 2

 

 

 K 10 9 4

 

 

 

S

 

 

 

 

 A J 8 7 2

 

 

 

 

 A 9 8 6 2

 

 

 

 

 8 2

 

 

 

 

 7

 

 


What next? You could play a spade 'safely'. Really? Let us check the algorithm:

 

Trick 4:        ace of spades

Trick 5:        ace of clubs

Trick 6-9: clubs and spades cross-ruffed Trick 10: club ruffed

Trick 11: diamond

 

Your partner takes his king and must play a diamond for two high cards in the dummy (diamond and club). So here you are! At trick four, after the ace of diamonds you should play back a diamond! Who said, 'Both sides playing the same suit, one is crazy'?

South was right to play a diamond and had not your brain done a little bit of work, no doubt declarer would have collected his reward for the work of his own brain. Both algorithms were good, but here the last word belonged to the defence!

A diamond return leaves declarer with no chance to make his contract.

 

When playing, either as declarer or in defence, your first concern should be
to foresee what will happen two or three, or some Top even eleven, tricks later.

home