
About us	Contact us	Site map

Position Analysis in Crawford Game

The following backgammon position came up in the Finals of the Peoria ABT Open Division Championships, when one of the top players got this one wrong, and it might as well have cost him the match.

In the actual match, the score is Black 11 and White 12, and the match is to 13, and it is the Crawford game. That means that neither side can double this game. If White wins the game, he wins the match, and if Black wins the game, then it goes one more game to see who wins. However, if Black can win a gammon and 2 points instead of 1, he can win the match on this game.

Black has just rolled 4-4 and has to decide what to play. What’s your play?

Finding the Best Possible Play

There are many possible plays to consider, but let me take you through the thinking process that I use. I first look at every play that I consider to be “reasonable” and begin ruling them out until I am left with one or two. To me, there are only a few “reasonable plays:

moving all four checkers out from white’s 4 point;
moving 2 out from the 4 point and two to black’s 4 point;
moving 2 out from the 4 point and two down from the mid point (13 point);
moving 2 down from the mid point and two to the 4 point.

The first thing I do is look at the pip count. It is important to know who is winning the race, as I know that EVERY GAME OF BACKGAMMON boils down to a race. In this case, black is down 5 pips BEFORE the roll. Since 4-4 is 16 pips, Black is up 11 pips after the roll. Since the average roll is 8, and white would be on roll after the play, which puts Black up only about 3 pips, which is very slight.

Backgammon is Essentially a Race

So I know that the race is close and could go either way, so I know that in that situation I prefer not to leave shots and get hit, as that would put me farther back in the race. And I prefer to play in a way that tends to block my opponent from being able to advance his checkers without leaving blots, because if I hit him, that puts him farther back in the race.

With that thinking, I see that with the 3 and 4 options there is a big disadvantage that I leave a blot on my 13 point and he can hit that with any 1 or an 8, I would lose a lot of pips, and not have any direct return shot if he hits with a 1 (which is far more likely than if he hits with an 8). So those plays have a real negative.

I like the idea of moving 4 checkers out from his 4 point because I am winning the race and that’s the easiest way to race. If I leave two checkers back there, eventually I have to worry about getting them out. The downside of bringing them all out is that it is easier for white to play behind me; I have my checkers stacked up on one point which gives me less flexibility to advance or hit in the future; and I don’t get to make my own 4 point, which would be a great point to have if I do get a shot and hit it later.

I settled on play 2 for all of those reasons and for another reason as well. At this match score, winning a gammon would win me the match, and with the second play I thought my odds of winning gammons were a lot higher because there was likely to be more possibility of contact (shots) as the game progresses. I might leave more shots too, but it doesn’t matter, if I am black, whether I get gammoned or not.

In the backgammon tournament, the player who played black moved all four out, and he lost the game and the match. Because it was the finals, many of us were watching and several of us debated what the right play was and some even made bets. We then went to my computer, pulled up Snowie, put the position in and settled the bet.

Below is Snowie’s evaluation which clearly shows that play no. 1 is to move 2 out and make the 4 point. hat the numbers show is when you compare those two plays, the best play wins the game 51.2 percent of the time, and the other play wins the game 50.8 percent of the time. So they are very close in that regard. But the key thing, as I pointed out earlier, is gammons. With the best play you are likely to win a gammon 10.2% of the time and with the second play, bringing all 4 checkers out from his 4 point, you only win gammons 5.5% of the time.

So when you make your plays, you have to determine not only what gives you the best winning chances, but what also gives you the best gammon chances. And at scores where it does matter if you get gammoned, you have to consider that as well.