Thor Henrykson: Thanks for agreeing to spend a little time chatting with us. How is your backgammon game these days? Are you still discovering and learning new aspects of the game?
Peter Hallberg: To start with, I will give you a short introduction to my backgammon career. I became a semipro backgammon player in 2002. Much of my time was spent studying the game in order to advance from being a good player to a strong player. Eventually, I was successful in reaching a higher skill level. "Unfortunately," my small computer company received a contract, thereby ending my short professional career. From that point on, I played backgammon only as a hobby. Suddenly, backgammon got a lot more interesting, as I began digging into various backgammon problems because I wanted to, not because I had to. I enjoyed learning new (and, of course, old) stuff again.
One of the things about backgammon that I find fascinating is that no matter how hard you study, there will always be concepts that you can study in more depth. Everybody can quite easily learn to master the basics. The hard part is to pinpoint when the basic strategy does not apply. When you are rusty, you tend to diverge from the basics either too often or too seldom.
For example, it is common knowledge that you have to fight vigorously for your 5-point in the opening of a game. This is true only to a certain extent. The problem is to know when to continue the battle for the 5-point or accept that you came out second. If you don't realize that the battle is over, you probably will get into a world of hurt.
Lately I have taken a break from backgammon, but 2007 offers a lot of great opportunities to play in some of the major tournaments. Attending big tournaments calls for some weeks of intensive training and preparation. To prepare, I use a mix of playing and analyzing.
First, it is important to have a natural flow in your game. This is also one of the first things that disappear when you don't play on a regular basis. Most of the time, the correct play is the most natural play that you would make by heart when you feel the flow.
Second, you have to learn from your mistakes, and everybody makes mistakes. I split my mistakes into two categories based on whether it was a game-plan error or just an inaccuracy based on the right idea. I don't take it that seriously if my mistakes fall into the second category. On a good day when I'm focused, I probably won't make those kinds of errors. The game-plan errors, I study in depth, to understand why I got things wrong. Mentally, I try to get the concept right and forget about the actual position in which I made an error.
These days, my goal is to reach a high level before a tournament. The goal is not to reach the highest level, which is possible only by playing professionally every day, which I don't.
Still, I add new things to my game and I try to do my best every time I play in a tournament. I usually don't make huge adjustments to my game based on my knowledge of my opponent, because it often ends up with me outsmarting myself. This I've seen done by strong players many times.
TH: How important do you consider the skillful use of the doubling cube to becoming a great backgammon player?
PH: There is no doubt that the doubling cube is the hardest thing to master in backgammon. Deciding on a cube action involves perfect information, as in chess, where you have all the parameters available to make the correct decision. To have great cube action, you need the skill to both evaluate all available parameters separately and put them together to the final result. In other words, you need to know the exact value of your position.
Some of the parameters that you have to take into account when playing in a tournament are:
• Pip count: What is the race? Do I have a lead? Is it important in this type of position?
• Reference position: Do I know a reference position close to the one at hand? Am I doing better or worse than in the reference position?
• Game plan: What is the game plan? Does the game develop favorably or not?
• Structure: Are my checkers somewhat connected? Flexibility?
• Match score: Which adjustments do I have to make according to the match score?
• Threats: Does my opponent have some immediate threats I have to act upon?
And, of course, there's much more.
The reason why it is harder to master cube action compared to checker play is that it's two very different ways of decision-making.
The first part of deciding upon a cube action is to estimate the value of the position. For example, how many points would I be ahead/behind if I played this position 100 times. If I would be ahead 32 points after 100 games, the value (gammon lingo: equity) 0.32 "points/game" is implied. You use some or all of the factors mentioned before to estimate an equity. When you have the equity, the rest is math.
To make a checker play, you compare the possible moves to each other, caring only about their relative equity and not their actual equity.
It is much easier to compare two alternatives than estimating an actual equity. Let's say that you evaluate a position all wrong. Your cube action would be all wrong due to your bad estimate of the equity. If you have two alternatives for a move, it doesn't matter if you believe both alternatives are good or poor, as long as you can tell which alternative is better.
This sounds very mathematical, but over the board, it is impossible to do all of these calculations, and you must follow your gut feeling, as you would in a poker game and in all other aspects of life.
There is as much math involved in estimating an equity for a backgammon player as there is ballistics involved in kicking a ball for a soccer player. It's a question of using your intuition based on your experience and the knowledge you acquired from playing and analyzing. You have to make hundreds of mistakes to get it right. Then, you can make a qualified guess of what is right or wrong.
When playing chess, a winning strategy is playing the objectively best move every time. Backgammon tends to be the same way.
It should be clear by now that it's not possible to become a great backgammon player if you don't handle cube action well.
TH: Backgammon players, like Gus Hansen and Paul Magriel, are now a regular fixture on the tournament poker circuits. Having achieved the coveted title of world champion in 2004, do you now have your sights set on conquering the world of poker?
PH: When you talk backgammon, these two guys are among the heavyweights. They have amazing talents toward games involving imperfect information. It is no surprise to me that they also do well in poker.
I started playing poker at the end of 2004. I did not know the game, and I knew only a few people who did. It was difficult to start learning a new game, and as probably most people do, I started playing at a level where I did not belong. Also, I wanted to learn everything at once, which certainly did not help me improve my game.
After some introspection, I took a month off and started building up my bankroll and my routine from the bottom. This was a very difficult thing to accept, six months after I won the world championship of backgammon, even though there are no direct connections except my ego.
My poker library has since become quite extensive. My favorite books are
Harrington on Hold'em: Part 1 and
Part 2, and
No Limit Hold'em by Sklansky and Miller. I like to learn and understand ideas and concepts, as opposed to concrete examples. Examples can teach you only so much. In tournaments and cash games, I try to incorporate an idea at a time.
I hope to take on the poker world, as a lot of other backgammon players have done. It is difficult to get experience at a high level when you are on the come due to the buy-in level of the big tournaments. As of now, I hope to find a sponsor who is willing to take a chance and put me in some of the big tournaments. I am sure I will prevail sooner or later.
TH: Between poker and backgammon, what percentage of your gaming time do you devote to each game?
PH: For the time being, I'm investing a lot of time in poker. I am at a point where I am taking huge steps in the development of my game, and I want to take advantage of this as much as possible. At a later stage, I will divide my time more in the direction of 70-30 in favor of poker.
TH: What is your favorite poker game? Do you play online or live poker? Which do you prefer, live or online? What is your best tournament poker result to date?
PH: I like hold'em a lot. It looks simple, but it's very hard. Mostly, I play cash games, because I like playing after the flop. It is difficult to find deep-stack tournaments at a reasonable buy-in.
Fixed-limit hold'em, $15-$30, was the first level I tried playing online. Obviously, I got outplayed in every aspect of the game, losing more than $9,000 the first month. I really did learn a lot about the game during that period. It became crystal clear to me that knowing things like bankroll management, table selection, tilt control, and, of course, the game are necessities to become a winner.
Over the next year, I lowered the stakes and focused more on shorthanded games. But, like most people, I tried to move up when I could beat the rake at a certain level. This resulted in never making a sound profit from the levels I could beat. Eventually, I got tired of playing fixed limit and started playing no-limit. It was almost impossible for me to play low-stakes no-limit after making it in $20-$40 fixed limit.
Yet again, I played too high for both my bankroll and skill level. After a tough year, I was breaking about even in $5-$10 no-limit. Over time, I realized that learning poker is like taking an education. If you take a short education, you have less potential to make money than if you study longer. But the key point is that you build up potential only from studying. You have to start working by using what you learned to convert your potential to money.
Naturally, I then decided to stop learning by always striving to move up in stakes. I took some steps back and started grinding it out in $1-$2 no-limit. I also started playing six-plus tables at the same time and became a consistent winner.
I play both online and live poker. Most of the time, I play cash games online, but once or twice a week, I go to a live game. The live games are more fun and I use them as "a day out" or training for future live tournaments.
To this date, I have played in only about 20 live tournaments. It is illegal to arrange poker tournaments in Denmark outside of the casinos, of which we have only five in the whole country.
My best finishes in the 20 tournaments was fourth place in a field of 120, and a 13th-place finish in a field of 720. Online, I placed ninth in a $300 tournament with 200-plus participants, and in the top 60 a couple of times in the 5000-plus participants Sunday tournaments at PokerStars.
TH: How have your skills of being a successful backgammon player helped you in your transition to playing poker?
PH: The most important thing I have used in the transition from backgammon to poker is to accept the role of luck involved in both games. I do not know why, but a higher percentage of backgammon players can handle the luck involved compared to poker players. It probably has something to do with backgammon players being used to having the possibility to get the correct answer to most problems from a computer analysis. It's easier to accept what a computer says than it is listening to the fish telling you what you did wrong.
Backgammon and poker have a lot of game properties in common. Here are a few:
• Starting hands can be compared to opening rolls.
• Bets or raises can be compared to cube actions.
• Stack sizes and pot odds can be compared to match scores.
Starting hands can be compared to opening rolls
In poker (no-limit hold'em), you have a few starting hands that you would expect to make a profit from playing. Then you have a lot of hands with which you would expect to break even, whether you played them or not. When you play hands from the second group, you often do it for a reason. You might want to get a loose image even though you play only break-even hands or better. The point is, you have a choice of strategy without giving up equity.
In backgammon, you have a few opening rolls that you always play the same way. For example, if you roll 3-1 as your opening roll, you make the 5-point no matter what. Any other opening roll than 3-1, 4-2, 5-3, 6-1, and 6-5 has at least one good alternative. Again, you can choose to play the same opening roll different ways, depending on your playing style, without giving up equity.
In both cases, the choices you make early have great impact later on in the game.
Bets or raises can be compared to cube actions
If you accept (take) a double, you play for double stakes. If you decline (pass) a double, you lose one point times the values of the doubling cube before being offered. Let's say you were doubled to 2. If you take, you play for 2 points. If you pass, you lose one point and a new game starts. To accept the double cube in backgammon, you have to win approximately 25 percent or more of the games. In backgammon, you wait for a situation in which the winning probability is as close to 75-25 to make a perfect double. In fixed-limit hold'em, you are limited to bet or raise a predefined value, just as you are limited to double the stakes or play on in backgammon.
Stack sizes and pot odds can be compared to match scores
When playing matches (tournament backgammon), the score has immense influence on the cube action, and checker play, as well. The first game in a 17-point match is pretty much played as if it were a cash game. The same goes for the early stages of poker tournaments.
Poker example
You are sitting in a poker tournament's middle stage. You have a well above-average stack. The flop and turn come K
10
4
8
. There is 26,000 in the pot. The under-the-gun player moves all in with his last 10,000. You are pretty sure he has A-K by the way he has been playing the hand. You have 25,000 chips left and the A
10
, and are last to act. What should you do?
You figure that you have nine cards to the flush and two tens. That's 11 of 46 cards left in the deck to give you a win. That's approximately a 24 percent chance to win. The pot offers you 26,000 + 10,000 + 10,000 = 46,000 for a price of 10,000, which is just below 22 percent. You call.
Backgammon example
You are sitting in a backgammon tournament. You are playing a match to 5-points ad, leading 3-1. The following position comes up. You have six checkers left on the low points and your opponent has six checkers left. (As stated earlier, you can accept the doubling cube if you have more than a 25 percent chance to win.) In this case, you have a little more than an 18 percent chance. But it surprisingly is a take, due to the match score.
This is why. If he gives up the game, you're ahead 4-1 to 5, which is an 82 percent favorite to win the match. If he takes and loses the game, he has a 0 percent chance to win the match. In a backgammon match, it only matters that you won, not by how much you won. This is why you redouble at your first opportunity to play for the whole match if you take the double. That means that you've got a 100 percent chance to win the match if you win the game. So, you risk losing 18 percent match-winning chances to play for 100 percent match-winning chances. This means that you can accept the double if you have an 18 percent chance to win the game.
Common ground
This may sound very mathematical, but it's really not that hard. In both examples, I used a well-known formula stating the relation between how much risk I can take for a certain gain:
Break-even point = Risk ÷ (Risk+Gain)
Poker example: 10,000 ÷ (10,000 + 36,000) = 0.217
Backgammon example: 0.18 ÷ (0.18 + 0.82) = 0.18
The ways of thinking are much alike. I believe that's why most backgammon players tend to do better in poker tournaments when they cross over. We are used to thinking in terms of risk/gain scenarios, because it's essential to playing the game just fairly well.
TH: Which backgammon skills do you utilize the most in playing poker?
PH: In backgammon, it is of great importance to know what is probably going to happen and why. It is of much importance to learn to understand the continuity of the game. You apply this skill when you have cube action. You want to double at the right time - not too early (easy take) and not too late (easy pass). The same goes for poker. You don't want to bet either too much or too little. Every early decision affects the later parts of a game (or hand) so much that you can't rise above intermediate playing level without understanding the later implications of your current decisions.
When you accept a double, you "own" the cube and have exclusive access to further redoubling. The value of owning the cube depends heavily on the type of position. If you have great potential to make a good redouble later, the value of owning the cube is high, thus you can take more. On the other hand, if you'll probably never reach positions where you'll have a good redouble, the value is low, thus you can take less.
In poker, you can often pay more than your immediate odds warrant if you have a hand that will likely be paid off if you hit, or you have other reasonable ways of winning the hand without showdown. In general, I believe that most poker players pay too little or too much attention to the implied concept.
Of course, I use the mathematical toolbox from backgammon quite a bit when I play poker, as explained earlier.
TH: Can a backgammon robot regularly beat a human, or does the skillful use of the doubling cube negate the advantage that such a robot may have?
PH: The best backgammon programs beat any human over a long period. The bots have excellent checker play and cube action. The great thing for everybody is that there is much room for improvement, even for the bots. The general level of play in the championship divisions of the big tournaments has risen a great deal over the last 10 years, and I think it is going to get even tougher.
TH: In your opinion, do the backgammon bots threaten the rising popularity of backgammon?
PH: I believe we have learned more from the bots the last decade than we learned the last century. The development of new ideas and the possibility to verify them through deep analysis aided by computers are essential for getting closer to understanding the beauty of the game. Bots are a tool to guide you in the right direction. They can't tell you why a certain move is the best. All serious chess players have a program as their sparing partner. It was long thought of as a threat, but recent years have shown it to be a valuable tool, and I believe the same goes for backgammon-playing bots.
Some believe the bots have taken some of the mystery out of the game, and that's true, but I believe that the bots have opened more doors than they have closed.
TH: What insight can you provide for someone who is making the transition from poker to backgammon?
PH: When learning poker, you have to play a lot of games to get an idea of what is going on. The dynamics of poker are hard to explain in a book. In backgammon, you do not need quite as much experience to grasp the ideas you get from books. Most basic ideas are easy to grasp and use immediately. I believe that you can improve much faster by reading books on backgammon than on poker. Generally, the books written on backgammon are of much higher quality than most of the poker literature on the market today.
I can recommend two books for the new backgammon player:
Backgammon by Paul Magriel and
Boot Camp by Walter Trice.
You can play backgammon with lots of people online. There are online backgammon rooms like the ones in poker. There you have the opportunity to play for fun or real money. I recommend that you start out by playing for fun, until you are confident in your skills. Backgammon can seem to be a game of luck only, but don't let that fool you.
If you want to know more about me or backgammon in general, you can visit my website at www.PeterHallberg.com. I play a lot of promotional online matches for fun or low stakes. Check the website for information about where and when I play.
