Taking Risks in Backgammon

Look at the backgammon position below. It is a money game and you are White and have a 5-2 to play. Should you hit or not?

Many people do not want to hit here because of the fear of getting hit back and getting gammoned. There is a saying that I really like: “Sometimes the greatest risk is to take no risk at all.”

Backgammon is all about taking risks, but they must be calculated risks. When you hit, it’s important to look at the odds. First, you only get hit back if your opponent rolls a 5, 3-2, or 4-1. That’s 15 out of the 36 possible rolls. That means that 21 times you don’t get hit, and when you don’t, you have an excellent chance to win the game. If you don’t hit, your chances of winning are very slim. Further, if it is gammons you are trying to prevent by not hitting, you might get gammoned anyway as you cannot safety the checker on your 12 point and that might get hit anyway.

Let’s take a look again at what happens if you get hit. First of all, you might roll a 6 immediately and still have a chance to win the game and still have a good chance to get off the gammon.

Secondly, some of Black’s hitting numbers cause him to break his inner board making it even easier for you to come in and even giving you a shot at another checker (5-3 and 5-4). If you take those 4 rolls away from his 15 hitting rolls, that means Black has only 11 good rolls and 25 bad ones. So the odds are better than 2 to 1 that after hitting he will roll badly.

So that is the reasoning behind the hit. Over the board, you have to look at these odds…look at the “most likely scenarios” that will probably happen, and decide how they will affect you.

But there’s one other thing experienced players do. We put the position into Snowie or GNUBG, the backgammon computer programs, and see what they say and see what numbers the come up with. In this case, the results, as shown below, clearly agree with my analysis.

What the numbers show is that it is a 19.5% error NOT to hit…a major blunder. By hitting you have a chance to win about 47% of the time. If you don’t hit, you will only win this game about 33% of the time. So your wins go up by a whopping 14% on just this one play of the game.

If you don’t hit you certainly don’t get gammoned very often—only about 8% of the time. If you hit, you do get gammoned a lot more--almost 24%. So why trade 14% more wins for 24% more gammons?

The answer is in the math—risk reward. If you lose a gammon it costs you 2 more points (when the cube is at 2), but when you win a game, you actually gain 4 points! You gain the 2 points you win PLUS the 2 points you didn’t win. So a win, after the cube is turned, is worth twice as much as a gammon.

You also have to add in the number of gammons you win with each play, and in this case you don’t win many gammons either way, but as the analysis below shows, you win about 5% gammons by hitting as opposed to 2% if you don’t.

The lesson here is that coming up with the right play takes a combination of reasoning, math, and experience. I “think” hitting is right because my reasoning over the board about what I think is most likely to happen favors hitting. I am able to come to this reasoning because of many years of backgammon experience, but also because I have testing my reasoning by putting many positions like this into Snowie and GNUBG to verify my thinking. Luckily, I have a good enough memory to remember many of these positions and apply the same logic over the board.

The BIG LESSON HERE is that if you want to improve your backgammon game, you learn very little by just playing and seeing what happens—you must test your thinking with backgammon programs or a teacher who will help you learn and understand the reasoning and match behind the plays. For example, if you had played this game, not hit, and happened to win, that might have reinforced your thinking that not hitting is right and you would make plays like that in similar positions in the future. And if you did that, the odds would certainly catch up with you and you would find yourself losing a lot more than you should. And like most who haven’t studied the game, you would probably blame it all on bad luck and the dice, not having any idea it was your own poor plays that cost you the money or the match.

How to Calculate your Backgammon Risks

What Snowie Says?

What about Gammons?

This Week's Backgammon Lesson: