This article originally appeared in the July 1999 issue of GammOnLine.|
Thank you to Kit Woolsey for his kind permission to reproduce it here.
There is an old saying, which is: "When in Rome, do as the Romans do". What
this means is that you should do what is commonly done depending on the
environment, rather than take actions which are contrary to your situation.
The same concept applies to backgammon. It is important to choose a play which is consistent with your position. A game plan may be great for one position, but change that position and it becomes the wrong game plan.
Backgammon is, in essence, a race. Despite all the subtleties involved, the game comes down to moving your checkers around the board and taking them off before your opponent does. Your overall plan is to take the lead in the race, bring your checkers around, and take them off. If you are behind in the race you must do something about it, such as containing one of your opponent's checkers or forcing him to leave you a shot which you hit.
It follows that our choice of strategy should be quite dependent upon the status of the race. To put it simply:
When ahead in the race, race.
What could be simpler and more obvious? Yet I have countless times seen good players make horrible blunders because they didn't follow this advice, choosing a strategy which wasn't consistent with their racing status.
Let's start with a simple example which everybody would get right:
Blue has a substantial lead in the race. Therefore, his game plan should be to get into a race. The 2-2 roll gives him that opportunity to carry out that game plan. It is clear for him to play 18/14(2). This almost assures that he will be able to break contact completely without leaving a shot, so the game will turn into a straight race which is what he wants.
Contrast the above with a similar-looking but quite different position:
This time Blue is well behind in the race. His game plan should not be to race, but to try to get a shot. Playing 18/14(2) would be a terrible blunder. He should simply play 9/3, 5/3, and wait.
Another variation on this theme:
Blue is still behind in the race, so racing is not the right game plan. However he is only behind by a little bit, so preparing to race is a possibility. If Blue hangs back on the 18 point the timing is likely to go against him, and if nobody rolls doubles he will eventually be forced to crunch his board or leave the first shot. Best is the compromise play of 18/16(2), 6/2. This retains plenty of contact, since White can get by only with doubles or 6-5. However those extra four pips Blue plays with the back checkers help his timing. Now he is in position to make a run for it himself if he rolls 6-5 and takes the racing lead, and his board won't crunch quite as fast as if he stays back on the 18 point.
When considering whether to put it all on the line with one roll, the race is usually the biggest consideration. If you are well ahead in the race you don't want to take any unnecessary chances, but if you are behind or if the race is close it may be correct to make the all or nothing play.
With the White checkers moved in 9 pips, it is another story. Now the race is very close. Rather than get involved in a close race, Blue is better off playing 8/2, 6/4* and putting it all on the line now.
The choice of which anchor to play from is often a function of the race. If you are ahead in the race or if the race is close, you usually want to play from the most advanced anchor available. However if you are behind in the race, that is a losing proposition. You want to maximize your chances of hitting a shot, which means playing from a deeper anchor if that is feasible.
Now the race is fairly close. If Blue hangs back on the 22 point he will soon run out of time and be forced to crunch his board. Instead, he is better off advancing to the 20 point with 22/20, 6/2 and playing from there. He will have decent racing chances, as well as some shot-hitting potential. The plan will be to advance the other back checker up with the next two, and then spring it safely with a large number. It is no coincidence that the race status and the timing of the position go hand in hand. In the first example when Blue was well behind in the race, there figured to be enough time to play a decent three-point game and hold the board together. In the second example when the race was close, Blue's timing for the three-point game didn't figure to hold up. This is a very common phenomenon. The farther behind you are in the race, the better your timing for a holding game, back game, or priming battle will be.
Anchor breaking decisions are often very race dependent. While there are plenty of other factors such as relative inner board strength, danger of being attacked, comfort of the play if you hold the anchor, and general timing of the position, the race plays a very important part. Often the race status will go along with some of the other relevant factors.
Blue is ahead in the race, so he should race. The correct play is 21/13. Blue's plan will be to jump the other back checker out as soon as possible, and if he gets away with this plan White will be left with an inferior holding game. It is inferior because White will be behind in the race. Note how the race status affects some of the other factors which are important when determining whether or not to break an anchor. White is short on attack material -- a direct function of White's being behind in the race. On the other hand White figures to have the better timing if Blue stays on the anchor -- again, a function of Blue's racing lead.
This time, the race is very close. Blue should hold the fort and simply play 13/8, 6/3. If Blue runs and gets away with it (i.e. manages to safely escape the other back checker), he still won't have all that great an advantage since the race is so close. On the other hand, running with 21/13 risks having the other back checker attacked, and White has more ammunition with which to attack than in the previous position. Also, the timing could go either way with the race so close. All the arguments which pointed to running when Blue was well ahead in the race now point to holding the anchor with the race even.
Timing is the big thing in priming battles, but timing is closely associated with the race. The decision of whether to concentrate on escaping the enemy prime or build up your own is a common one, and the race is a key factor. For example:
Blue can create an symmetrical position with 13/7, 8/7, or he can make a run for it with 23/16. The problem with running is that it is a disaster if the shot is hit, while if Blue gets away with it he still has plenty of work to do because the race is so close. White has an escape hatch via Blue's bar point, so White is one roll away from getting away and getting into an even race. Blue does not have the racing lead, so racing is not called for. He is better off making the bar point and taking his chances in the priming battle.
This is another story. Now Blue does have the racing lead, so if he plays 23/16 and is not hit he will have a significant advantage. Also because of his racing lead his position is more advanced than White's. This means that he figures to have the worst of a priming battle. All of these considerations point to 23/16 as the best play.
Offense vs. defense is often a dilemma. Is it better to concentrate on building a new offensive point, or is grabbing a defensive anchor more important. While several factors may be involved, it is not surprising that the race is often a major consideration.
Blue has a substantial racing lead, so he should race. The correct play is making the advanced anchor with 24/20, 23/20. If Blue makes the offensive five point he could find himself on the wrong end of a priming battle despite his better board and White having three men back.
Now Blue only has a small racing lead, which translates to better timing for him in the priming battle. This time his best play is 9/5, 8/5. If he can extend his prime to a 5-prime, White will be hard-pressed to stay in the game. Blue would still be in good shape if he makes the defensive anchor, but this gives White more of a chance to fight for Blue's five point. If White can win that fight he will be right in the game, since White is competitive in the race. Even though Blue has a small racing lead, his offensive potential is so strong that he is better off building up his board.
Slotting plays also are often race dependent. The key is that if the slotted checker is hit you lose a lot of ground in the race. If the race is close or you are ahead in the race this is very costly, arguing against slotting. On the other hand if you are already behind in the race the cost of being hit isn't so great.
Blue has a moderate racing lead, so getting hit is costly. Since he is ahead in the race, he should race. The way to do that is to split the back checker with 24/23, 13/7.
This time the race is close. Blue gains a lot if the slot works, since it develops his position very well. Also the split is more dangerous, since White has two extra checkers in the attack zone. Blue should play 13/7, 6/5. Note how the change in the race made splitting more dangerous, since White's gain in the race translated to more ammunition up front.
These examples illustrate the importance of taking the race into account when choosing your strategy. While the race isn't everything, it is very important. A proper analysis of a position is not possible without a feel of where you stand in the race. Once again:
When ahead in the race, race.