Cube Handling

Is It a Take?
Kit Woolsey, 1984

From Backgammon Times, Volume 4, Number 1, Spring 1984.

Many players are very good at determining whether a double should be passed or taken, yet become confused when deciding whether to offer a double. This is understandable, for it is usually more difficult to determine when to double than to determine whether or not to take. The person being doubled simply has to assess the current position and decide if he will win 25% of the time (taking gammons into account, of course). The doubler, on the other hand, in addition to assessing the current position, has to worry about losing his market, playing for the gammon, and even understanding the psychology of his opponent.

A player can use his ability to assess a position as a pass or a take to determine whether or not to double if he takes the right approach. The procedure is as follows: When you are considering whether or not to double, get up, walk around to your opponent's side of the board (do all of this mentally, of course), and see how it looks to him. Then, answer the simple question:  Is It a Take?

It would seem that the question has two possible answers. Actually, there are three possible answers. They are:

  1. I'd snap that cube up so fast your head would spin.
  2. You couldn't pay me enough to take that cube.
  3. Let me think about it for a minute.

In other words, the position is either a clear take, a clear pass, or there is some doubt.

The most important rule about doubling is this:

If your answer to the question "Is it a take?" is
"There is some doubt,"
then it is always correct to double.

(Yes, I know that there are certain match situations where you are well ahead in the match and there would be overage involved in a recube where this might not be true, but for any other match situation and for all money play it is always true.)

There are three reasons for this. They are:

  1. Even if you think it is a take, if you are not sure your assessment might be a bit off, and it might in fact be a pass. If this is the case, not doubling is criminal, for you would be giving a free roll to your opponent when you could have cashed a sure point, or better, had him take a double which he should have passed.

  2. There is always the possibility that your opponent will (incorrectly in your opinion if you think it is a take) drop the double. Since you are not sure, he might well think it is a pass. In this case, not doubling will again give him a free roll to pull out of a position he would have dropped.

  3. Even if your opponent (correctly in you opinion) takes the double, since you thought it was close it can't be too far from a pass, which means that if your position improves on the next roll it will be a pass. In other words, if you are not sure that it is a take now, there is a good chance that you will lose your market if you wait, which is good justification for doubling.

I must emphasize that you must be absolutely sure that it is a take in order to justify not doubling (or, conversely, absolutely sure it is a pass in order to justify playing for the gammon). By this, I mean that if Paul Magriel said to you, "I think it is a pass, and I'll pay a point and the cube for $100 a point," you would gladly say, "Sit down Paul," and have no second thoughts about it. If you would have second thoughts, you are not absolutely sure whether or not it is a take, which means that it is automatically correct to double.

If you are sure that it is a take, it may still be a double because of the market losing potential. In order to determine this, examine a typical favorable exchange of rolls (i.e., good for you, bad for him, on the next turn). Don't just look at your super-joker sequence (if you are even considering doubling you will almost certainly have one), but just an average-plus average-minus exchange. Examine the resulting position and ask the same question: Is It a Take?

  1. If the answer is (c) "not sure," then it is not correct to double now. The reason is that you have just proved you can't lose your market by much, if at all, for the resulting position, if favorable, will be at best a close pass for your opponent.

  2. If the answer is (b) "definitely a pass," then it is probably correct to double. The only reason for not doubling in this case would be if your position were so shaky (such as a blitz with insufficient ammunition) that if things didn't go well you would quickly be losing. Otherwise, it is essential to double to avoid risking your market.

Now let's look at the other side of the coin—you are sure that it is a pass, and are considering playing for a gammon. Once again, you must be absolutely convincd that it is a pass (i.e., eager to give Magriel a point and the cube for $100 a point). If you are not absolutely sure you must double, because:

  1. Your assessment could be wrong, and it might be a take. In this case it would clearly be wrong to "play for the gammon," for by doubling you could play for the same gammon at double the stakes!

  2. Your opponent might think it is a take. Once again, by doubling you will have the opportunity to play for the gammon for double the stakes, after your opponent makes his (incorrect, in your opinion) take.

    Even if it is a correct pass, if it is close enough that you aren't sure it is best to cash the sure point, for your opponent must have substantial winning chances for you not to be sure, then the play for the gammon won't be good enough to justify risking losing.

If you are sure that it is a pass, in order to determine whether or not to play for the gammon it is once again necessary to look a roll down the road. This time, you consider a typical unfavorable exchange (bad for you, good for him) and ask the usual question about the resulting position.

If the asnwer is "not sure," then you should play for the gammon. This means that even if things go badly the best your opponent can wind up with is a marginal take, which is almost as good as winning outright for you (since the stakes will be doubled). If there is any gammon chance at all, it is worth taking a roll.

If the answer is "yes, definitely a take," then you should probably cash your sure point now. What this means is that you have a substantial risk of losing if things go badly. Only excellent gammon chances justify playing on in this situation.

Let's apply this theory to a few positions. In each case, Black is on roll, cube is in the center, early in a long match.

13 14 15 16 17 18 19 20 21 22 23 24
12 11 10 9 8 7 6 5 4 3 2 1
Position 1.
Should Black double?

Position 1 illustrates a typical early game positional advantage. It is a take? I think it is, and would take it at the table. However, I am not sure. Therefore, it is a double—no more questions asked.

13 14 15 16 17 18 19 20 21 22 23 24
12 11 10 9 8 7 6 5 4 3 2 1
Position 2.
Should Black double?

What about Position 2. Is it a take? I know that it is. White has excellent chances to contain Black's last man, a solid 3-point game in reserve, and is in virtually no gammon danger. What about a good-bad sequence? Suppose Black hits and runs or makes his bar point, and White counters with nothing special. Is the resulting position a take? Probably not, but I am not sure. Consequently it is not correct to double now, for even a good-bad sequence gets to a marginal pass at best.

13 14 15 16 17 18 19 20 21 22 23 24
12 11 10 9 8 7 6 5 4 3 2 1
Position 3.
Should Black double?

In Position 3, White has the makings of a perfect bar-point game. He is only 10 pips behind so he can run with large doubles and be right in the race, and if he doesn't roll a set he may well get a shot from the bar point. Even if his blot is hit he can recover if he gets in quickly, so it is an easy take in my book. However, a good-bad sequence (Black hits, White flunks) eliminates White's racing chances and puts him in severe jeopardy if he stays on the bar too long. Consequently the good-bad sequence results in a definite pass, so it is correct for Black to double now.

13 14 15 16 17 18 19 20 21 22 23 24
12 11 10 9 8 7 6 5 4 3 2 1
Position 4.
Should Black double?

What about Position 4. Is it a take? I don't think so. Even though White's position is sound, Black has the edge in every respect, and if he makes the 4-point he will have a big advantage. Should Black play for the gammon? No! I said I didn't think it was a take, not that I was sure it wasn't a take. Automatic double—case closed.

13 14 15 16 17 18 19 20 21 22 23 24
12 11 10 9 8 7 6 5 4 3 2 1
Position 5.
Should Black double?

In Position 5, White has a crystal clear pass—no doubt about this in my mind. Should Black play for the gammon? Suppose something bad happens on the next roll, such as getting a blot hit or White rolling 5-5. Whill White then have a take? Possibly, I'm not sure. Since the best White can get is a marginal take, it is correct to play on for the gammon (at least for this roll), even though the gammon chances aren't particularly great.

A champion bridge and backgammon player, Kit Woolsey has published several widely acclaimed books on bridge and contributes regularly to Backgammon Times.

More articles by Kit Woolsey
More articles on cube handling
Return to: 
Backgammon Galore