Over the board I made the fateful (and common) mistake of choosing a move without first deciding my plan, and compounded this error by forgetting to consider the position of the cube. I should have remembered Goulding's first law: "Have a plan, consider the cube, evaluate the candidate plays".
So let's do it the right way: What is Black's plan? He is way ahead in the race - 57 pips to 102 pips. As he has no priming potential and his home board cannot be strengthened it seems that Black's plan should be to play for the race. This indicates either play (a) or (b). What about the cube? Although Black leads the race by a long way he cannot double yet as his man on White's 5-point still has a long way to travel to reach his own home board. However, if Black can advance the man part of the way without being hit then he will have a strong redouble.
What about candidate plays? The three identified above are really the only possibilities. Given that we have decided to play for the race, play (c) can be discarded: it does nothing to make progress in the race and at the same time weakens the home board. Play (a) brings the man closest to home but leaves 26 shots. Play (b) makes slightly less progress in the race but leaves only 21 shots. After either play (a) or play (b) provided he doesn't get hit, Black will have a strong redouble which White might be able to take depending upon what his own roll was.
As Black will have a redouble after either play it is correct to leave fewer shots and therefore the correct play is (b). Over the board I made the awful play (c) ignoring one of my own lessons: in backgammon you must play to win rather than play not to lose.Reuse content