Backgammon

JUST OVER 20 years ago, the American backgammon expertBarclay Cooke published a book of problems entitled Paradoxes & Probabilities. In his time, Cooke was viewed by his peers as one of the leading authorities on the game.

Reviewing his solutions with the aid of modern theory and our silicon friends Snowie and Jellyfish, it is staggering just how much our understanding of the game has progressed, though in retrospect I suppose this is not surprising. In the late Seventies, backgammon theory had evolved through trial and error, computers were massive, and members of the backgammon community reinforced each other's views.

The result was a very pure style of play. Slotting key points was very much the order of the day and blitzes were rare. The position above is Problem 116 in Cooke's book. Black has a 61 to play. The recommended solution was 13 7, 5 4 looking to build a five-point prime if white doesn't roll a 1 or a 2. The correct play is 8 2 * 1, attempting a blitz and keeping white off-balance. The equity difference between the two plays is a massive 0.3.

Cooke shouldn't be overly criticised for his solutions. He admitted that he didn't know the right play in many cases. It is only with the work done by Magriel and Robertie, and the advent of computers, that we can now state with any certainty what the correct play is in any given position.