The simplest lesson in match tactics is: win the match.
Keep your focus on your goal, not the best chance to win the current game
or score the most points or even to avoid being embarrassed. Maximizing
your chance to win the match means making each doubling decision and many play
decisions with the match score (and time if it is a timed match) at the
forefront of your thinking. An honest evaluation of relative skills will
make difference in many situations.
When playing for money, each game stands on its own. You want maximize how much you win and minimize how much you lose over the long run. Leaving aside managing your money and emotions, the mathematical odds should rule your decisions when playing for money. In the long run math works out. In a match, there is no long run. First to the goal wins, the margin doesn't matter. That you were a favorite in given position doesn't matter.
Let's look at some situations:
End of match, you're behind:
When your opponent is within reach of winning, doubling decisions are easy. If your opponent will win the match if he wins the current game then you have nothing to lose by doubling. This is what the Crawford rule recognizes. When the score leader reaches match point, the Crawford rule protects him from an immediate double from the one who is behind for one game.
At other times late in the match, the one trailing has an advantage. If there is a possibility of a gammon, doubling early may cause a drop in an otherwise even game as the leader does not wish to give up the big score.
Playing the pips can also be influenced by being behind. Active moves that may lead to a gammon for either player are much more attractive for the player behind. Keep in mind your opponent is more scared of you making a comeback than you are of making things worse. When you are behind, if things go on evenly you will lose so shake things up. At worst you'll be able to go drown your sorrows quicker.
End of the match, you're ahead:
Obviously this is the flip side. After the Crawford game, your opponent should double at the first opportunity. Usually you will just take and then be careful not to let the emotion of the moment ever make you even think of doubling back. There is however one concept to keep in mind called the free drop. If your opponent needs an even number of points to win, you may drop once without changing the number of games he needs to win. For example, if the score is 10-5, it takes 3 two point games for the player with 5 to win the match. If you drop making the score 10-6, three games are still needed to win. That means if your opponent gets a good first roll and you roll a mediocre response, you can, just once, go on to the next game at no cost by dropping the double.
If ahead but not yet at match point, reasoning is similar but not as clear cut. If you are several points ahead, what you'd really like is to have a series one point games where you win your share and slowly turn a 6-2 advantage into a comfortable 11-7 win. Even 8-6 and out of time works. Your opponent needs 2's and 4's to get back in the match so you should strive for straight forward running games that result in one pointers. Avoid plays with good up sides but that risk disaster if they fail. Better to take a 50/50 chance at 1 point than a 60% chance to win but with a 20% chance of losing a gammon. Especially if a gammon for you doesn't matter. When doubling you want to wait until you want to play the current game for the match. Only when you are sure your opponent should drop do you want to double because if the cube is taken, it will come back immediately if two points is enough for you to win and will come back quickly if things should turn around. If you are up 6-1 in a 9 point match it can be very hard to take a cube back to 4 that could put your opponent back in the match even if it is a clear money take.
Early in the match, you are the weaker player:
If you believe your opponent is very likely to win a long match, you want to shorten it as much as possible. It is to your advantage to take marginal doubles and redouble as soon as the tide of the game turns. Suppose you estimate you have 1 chance in 10 of beating a better player. If you can, make the match be decided on one big game. If you have a 25% position on a 4 cube, your opponent is more afraid than you are. Your best chance to win is to get lucky in one game. The more games you play, the more likely luck will even out and skill will prevail. In a way, you should play like you are behind right from the start. You will be feared!
Early in the match, you are the stronger player:
Lots of games where technique overcomes the whims of the dice is one plan. The other is to take advantage potential gammon situations where your opponent doesn't recognize the danger. Even if behind early in the match, you still want to grind out the win but risky complex positions become more attractive, complexity favors you while the risk is more worthwhile when you trail.
Playing slowly when time is a factor is, in my mind, socially unacceptable. Chess time clocks have been used in backgammon tournaments where this has been a problem. I would hope a group of gracious congenial players like ours would not have this become an issue. On the other hand, favoring long drawn out games when you are ahead and time is short is sound tactics. Even taking a double so as to play out a long game for two may be better than dropping two quick cubes and starting a third game. Again you shouldn’t change your pace of play when ahead and time is waning but you can choose the type of game you play.