Let's use a fairly simple example to demonstrate the difference between tournament and money play.
If this position arose in a money game Black would double and White would take. In 36 games White would lose 26 games and win 10 for a net loss of 32 points - better than the 36 points he would lose by dropping.
Now let's suppose that Black leads White 5-4 in a match to 7 points. Again Black will double; should White take? Let's work it out. If he takes and Black throws one of his 26 (72 per cent) winning rolls he will win the match; if Black throws one of his 10 (28 per cent) losing numbers then White will lead 6-5 with the Crawford game to be played. As already noted above, White will then be a 70 per cent favourite to win the match. From the start position above his chance of winning the match is therefore the product of the likelihood of the two events occurring: winning this game and then going on to win the match. This likelihood is thus .70 x .28 = 19.6 per cent.
What if White drops the original double? Then he will trail 6-4 with the Crawford game to be played. His winning chance from that score is 25 per cent. This is easy to calculate as basically he must win the next two games to win the match. The probability of this happening is one in four or 25 per cent. So if White takes, he wins the match 19.6 per cent of the time while if he passes he wins 25per cent.
This is a huge difference and the conclusion is obvious - White must drop the double, whereas for money he would happily accept. This one example just touches the surface of the complexities of tournament doubling - a fascinating but difficult area.Reuse content