Managing Partner at Dragonfly Capital. Effective Altruist. Airbnb, Earn.com (acquired by Coinbase) alum. Instructor @ Bradfield. Writer. Former poker pro. Donate 33% of my income to charity.
What is learning? How does it work? How can we learn better?
These questions might seem naïve on first glance. Learning is fundamental to how humans negotiate the world, and yet it feels strange and reductive to take it aside and interrogate it. But if you are a poker player, learning is an essential part of your vocation. To really improve at poker, it is imperative to optimize the process of learning.
It is possible to learn how to learn. This article will consider the different types of learning involved in poker, the science of that learning, the three most important techniques to optimize the way you learn, and, finally, the common mistakes and pitfalls that face every learner.
As a poker player, you have all sorts of resources readily available to learn from—videos, forums, books, and so on. Yet no amount of watching videos or reading forums will, on their own, make you good at playing poker. Why is this? Isn’t poker, after all, a purely mental game? It’s not like football or tennis; in poker, all you have to do is know what the right play is, and then do it. Yet why doesn’t it feel that way?
No amount of study will make you good at poker. This is because there are fundamentally different modes of learning involved in playing and in watching a video or reading a forum, some of which we established in Chapter 5. To return to some of those concepts, the three types of learning are as follows. First, there’s poker theory or knowledge—what we might call propositional learning. This is the kind of thing you generally absorb from watching a video—it is your ability to talk about poker, all of your ideas about exploitation, combinatorics, game theory and so on. Then there’s another type of learning, your know-how or performative learning—in other words, the ability to play unconsciously. This is your ability to click buttons for two hours, instantly to think of the right play in the moment—your raw poker intuition. Third, there’s emotional learning—often called “the mental game.” This is your ability to take losses, to play consistently, to follow your own rules and monitor your mental states. Although all three of these kinds of learning are essential and mutually reinforcing, they are profoundly distinct. If you want to develop your poker know-how, it is unlikely that engaging in an activity that develops your theory or emotional skills will be much help (although they may still bolster your overall game).
It is helpful to imagine each of these as essentially being composed of separate networks in our brains. While each of these networks contributes to the overall effect of playing good poker, these networks are distinct; each is strengthened and grown in a different way, by different stimuli, and are practiced differently. To draw an analogy, if you’re a boxer, being good at your sport involves developing your stamina, your punching strength and your footwork, and so on. Although all of these skills together might be called “boxing,” all of these are distinct abilities, developed differently, and none of them individually could be called “boxing.” Yet they are each essential to the final boxer.
So, as poker players, we have three distinct skillsets that we have to develop: our poker theory, our poker intuition, and our emotional skills. How do we develop each of these individual skills? What’s the best way to learn them?
In order to answer these questions, we will have to delve into a deeper understanding of the process of learning itself. This leads us to what I call the three-fold path—the three most important techniques that are essential to optimizing learning.
In Chapter 4 we discussed how the brain is composed of neuronal networks—networks of neurons. The interconnections that form among these neurons are strengthened by repeated firing, and are inhibited by inactivity. Through this continual firing and non-firing, networks are gradually borne out. These networks are our mental building blocks—it is in the shaping of these networks that all of our knowledge and capabilities are ultimately stored.
We can define learning as simply any change to the connections in our neuronal networks. Whenever a connection is strengthened or weakened—in any way at all—we will call that learning (barring changes caused by injury or malnutrition). You might protest that this seems overly inclusive. After all, sometimes your neuronal networks will be conditioned to make you play worse. If your brain convinces you that J4o is a lucky hand and you always win when you 4-bet it—that’s a change to how your neurons interpret J4o, but do we really want to call that learning?
Absolutely. It is imperative that we don’t prejudge learning, or take learning only to mean productive learning. Learning involves a lot of trial and error (emphasis on error), so learning incorrect plays is an inevitable and important part of the learning process, especially early on in the development of your mental network. For every bad habit you learn, however, the goal is to simultaneously learn two good ones.
Now, let’s posit that there is some ideal neuronal structure, with a very specific shape. We can call this shape the “optimal network.” That network is essentially the brain of a perfect poker player—it is the structure of the perfect poker game. Your own network has its own shape, probably simpler than the optimal shape. If you superimposed your structure onto the optimal structure, like putting one transparency over another, you would be able to see all of the inconsistencies—all of the differences between your game and the perfect game would be apparent. You want to train your neuronal networks toward hammering out those inconsistencies.
This training happens naturally. Because the optimal network is the shape that makes the most money in the long run, you will automatically evolve toward this network. That is, simply by a process akin to natural selection, your game will gravitate toward the shape of the optimal poker game, just by getting conditioned by the never-ending feedback of making winning and losing plays. However, it is unlikely ever to get there on its own (if it did, just everybody who has played millions of hands would be a great player, and of course that’s not the case). Why does it get blocked in its movement toward that optimal shape? There are many reasons—lack of selection pressures (not playing in tough enough games), persistent cognitive biases and misperceptions, lack of critical awareness, being motivated and reinforced by factors other than money (such as excitement, etc.), and, sometimes, hurdles of complexity (like having to evolve an organ that’s ineffective until it’s fully evolved). There are many obstacles in the way of the undirected learner.
Entertain a metaphor for a moment—imagine that learning poker is like a game of Minesweeper. When you start playing, your map of the terrain of poker is very naïve—you have no idea what’s good or bad, everything just looks like an open field. The underlying reality (that is, the optimal, real map of poker) is very complex, but you’re oblivious to all of that. You just see the blank Minesweeper game. The only way to get more information and fill in your map of poker is by starting to take risks. You must uncover new blocks in the map, and with every click you begin to locate mines (bad play) and open space (good play). However, unlike in Minesweeper, you do not lose the game when you make a mistake. Instead, you might lose a pot, but you are allowed to continue. At worst, you take a slap on the wrist. In this way, poker encourages us: take risks! Make mistakes! Spread out as far and wide as you can, to figure out what works and what doesn’t. The more data points you have, the better and faster you will learn the terrain. The ideal approach to learning is one that uncovers the underlying field as quickly and diligently as possible.
Learning poker is hard. If it weren’t, it wouldn’t be lucrative. But this understanding of what learning poker is (that is, tempering the shape of our networks toward the optimal network), gives us valuable insight. The process is one of trial and error, continual feedback and calibration. Our goal in learning, then, is to speed up and optimize that process as much as possible. We want to maximize the data points, increase the rates at which our neuronal networks are altered, and we want to increase the retention of those alterations (i.e. make the learning stick). This is true in almost any skill—if you are not continually testing your limits and challenging your boundaries, you are not growing as a learner. In fact, one study of Olympic ice skaters showed that the ice skaters who achieved the highest skill level were the ones who took the most falls during practice. Rather than taking a mistake to be a bad sign, take it as a sign that you’re pushing the threshold of your skill level—which is exactly where you should be!
So the first key is that we want to have a strategy that takes a lot of risks. You’re playing a game of Minesweeper, except you effectively have innumerable lives. Click around. Make mistakes. In the end, every mistake is an opportunity to gain new knowledge about the terrain, and to use it to your advantage in the future.
Of course, there are many differences between poker and a game of Minesweeper. One of these differences is that as poker players, we are not forced to start with a blank slate. We are given access to other people’s maps of the minefield called poker.
Articles, videos, books, through interacting with poker friends or even just watching hands play out—they all give us valuable insight into the way that other poker minds are structured. Although looking at such things from the outside does not make us capable of emulating their poker mind (no number of hours watching Phil Galfond videos will make you play like Phil Galfond), it does allow you to see the overall structure and organization of their game. It gives you an idea of what a good map looks like, what it’s composed of, and how different elements relate to one another. Instead of simply throwing darts on a blank board like a game of Minesweeper, it allows you to have some idea of approximately where and how things should be allocated, the way mines are usually spread out, what a final map probably looks like.
This is the second key: to use blueprints to accelerate the shaping of our games. With blueprints, we are able to learn leaps and bounds faster, and things like videos, articles, hand histories, and our relationships with other poker players allow us easy access to very powerful blueprints.
So how do we best take advantage of blueprints?
Consider the education of apes. In certain species of apes, when a young ape is trying to learn how to properly crack a nut, the mother will put her hands around the hands of the infant and manipulate the infant’s hands to make it crack the nut correctly. This is called shaping. By shaping, rather than simply demonstrating visually by cracking her own nut, the mother is able to directly imprint on the infant’s mind the proper technique, muscle memory, and timing of cracking a nut. It primes the infant to learn the correct technique on its own. Shaping in this way is much more powerful than mere demonstration; it gets the technique “into the learner’s body,” so to speak.
So how can we take advantage of shaping in poker? The most prominent example is getting sweated. Being sweated by a stronger player is one of the most powerful methods of learning in poker, for the simple reason that it instills the learning into your body. If you are sweat and your sweat-er makes you do a check-raise you’d never ordinarily do, you will be much more able to perform the action in the future compared to if he merely pointed out the check-raise on paper. The simple act of clicking the button (and getting all of the table feedback of your opponent thinking, calling time, eventually folding his hand, you winning the pot) primes your poker-playing-mind to be able to make such a move again in the future. Of course, a sweat is more than one check-raise—in an extended sweat, you will assume a completely different player’s style, logic, and attitude toward poker. Inhabiting the space of another person’s mind will allow you viscerally to explore their blueprint of poker, and let its possibilities imprint upon your own mind.
Unfortunately, while sweating may be ideal for learning, real-time sweats may not always be possible. So when we can’t do a real-time sweat, what can we do? Well, we can instead simulate sweating. We will talk about this in more detail when we discuss elaborative rehearsal, as the ideas go hand-in-hand.
Language is another invaluable aspect to incorporating a poker blueprint. The way that someone talks about poker reveals a great deal about the way that his blueprint is organized. As an experiment, try observing the differences between the ways Phil Galfond, Sauce and Jungleman talk about poker hands in their videos—what are the first things they talk about? What concepts do they continually invoke? The things they all tend to talk about, and in what order do they mention them? What do they not talk about? By analyzing these things, you will reveal a lot about the way they think about poker, and to what concepts they give the most priority. Although these three players would reach many of the same conclusions about certain hands, they have very different blueprints of how poker should be played and thought about.
The point is this—start speaking with the right kind of language, and the structuring of your poker thinking will follow. If you start speaking about poker the way Phil Galfond does, chances are, your play will start to gravitate toward his. So language is important. But how exactly does one learn to speak “poker’?
Of course, apes are not the only creatures that shape their young—we humans do it too. When young children try to tell their parents stories, parents will instinctively guide their child toward the right way to tell a story. This is because children tend to rattle off facts—they have not yet learned how to create coherent stories. So a parent will lead them: “why did Billy do that?” “And then what happened to Billy?” “What did you learn from all that?” And so on. This is called scaffolding—the teacher provides a skeletal framework which leads the student towards the right kind of linguistic structure. Ideally, a competent poker coach will engage in the same sort of scaffolding. “Why are you making this bet?” “What do you think he perceives your range to be?” “What inference should we make from the hand he just showed down?” and so on. These kinds of questions lead the student toward having the right structure of thoughts, and prime him to follow a pattern (and perhaps ask those questions to himself). The scaffolding doesn’t “create the story,” so to speak. It merely guides provides skeletal frameworks to which stories can be grafted.
Ultimately, blueprints must be utilized in conjunction with risk-taking and, as always, acquiring lots and lots of data. No matter how well you’re following a blueprint, your poker game must be filled in with experience, and its blank spots must be filled in with raw data. The path to perfection is, as always, paved by playing hands, hands, and more hands.
This brings us to the third key of learning: rehearsal; specifically, elaborative rehearsal. Elaborative rehearsal is the concept that when you are trying to learn something new, you should try to engage that learning in as many ways and modalities as possible, and relate it to other things you know. This maximizes the retention of new learning.
Say, for example, that you’re trying to memorize this list of words: “cat / farm / blue / chip / pave / tree.” If you try to memorize this list using a shallow rehearsal then you’ll simply repeat the list of words to yourself until you can recite them from memory. It will probably take you a while, and isn’t very efficient. If instead you learn this list of words by creating a story and relating the pieces of information to one another—by *elaboratively engaging *all of the information and activating structures of meaning—you will learn the information faster, and retain it for longer. So instead of just a list of disconnected words, perhaps you try to create an image: “There was a cat farm littered with blue chips and paved with trees.” Or perhaps instead of simply reciting the words out loud, you practice writing them down, or creating a melody to accompany the list.
By engaging with the information presented to you in different modalities, you create stronger interconnections within or between the networks that already exist in your brain. Instead of having to forge a completely new network of knowledge (an auditory list of disconnected words), you take advantage of all of the other pre-existing networks that you can rely on to reinforce this knowledge and make its retention easier. You want your rehearsals (that is, your practice) to be as elaborative**and engaging as possible, and the same principle is fundamental to learning good poker.
Say that you are reviewing some of your hands in PokerTracker or HoldemManager. Your review method is to find your biggest losing hands, read them over, decide mentally if you made a mistake or not, and then move on. This is a very shallow rehearsal—there is a limited modality of engagement here, which is reading a hand history. The rehearsal becomes more elaborative if you do things like write down what you think your mistake was, or replay the hand in the hand replayer and imagine yourself playing it again. Simply visualizing the hand again will engage many more mental networks—the parts that are activated while actually playing—and make the learning more likely to stick. Anything you can do to engage the stimuli in more modalities makes you more likely to learn and retain the new information.
I should also mention, at the risk of stating the obvious, that when it comes to rehearsal, reviewing past learning is essential to retention. Once in a while, go back to your marked hands from last month or two months ago, and read some of them over again to remind yourself of what you’ve learned. Once in a while, read over all of your notes on your table nemesis, even if you feel like you know him like the back of your hand. Review is essential—so don’t short-change yourself, even if you feel confident in your reads.
There is a second principle of elaborative rehearsal, which is practice like performance. According to this rule, to optimize learning we should make rehearsals as similar as possible to performance. If, for example, you are going to be tested on your ability to write down the aforementioned list of words on a test, what do you suppose is the best way to practice for that? By the principle of practice like performance, you shouldn’t review the list again and again until you remember it, or try to memorize reciting the list verbally. The best way to improve your test performance is by testing—that is, by simulating the conditions under which you’ll have to perform your knowledge. Studies have repeatedly shown that if you practice by repeatedly testing yourself using a quiz similar to the way you will be quizzed, this vastly improves testing scores over any other method of vocabulary retention. This insight has many implications for us as poker players.
After all, in poker, what are you studying for? What’s the final performance? The performance is actually playing**poker. This means two things: one, we should deliberately practice our skills while we are playing poker, and, second, we should make our external practice as close to the reality of playing poker as possible.
Consider how most people watch videos. They sit idly, listen to the video-maker speak and almost zone out until they see/hear something really cool, or something they disagree with. Then they are engaged—but their engagement is primarily with the theory part of their mind. This is why watching a video is primarily good at developing your theoretical muscles, but not as good at engaging the play or performative skillset.
Remember when we talked about the value of sweating? Well, videos provide an excellent platform to mentally simulate sweating. If you watch a video while imagining that you’re really playing the hands, this will activate a totally different network in your brain from the video-watching network (that is, if you visualize it correctly). You might then imagine that the video-maker is directly speaking to you and telling you what to do. There is a wealth of difference between sitting and thinking about the theoretical implications of checking your set to him on the river and actually visualizing that you’re doing it. I would even recommend clicking the actual button on the screen, as though you were actually playing the table! This sounds silly because we are married to the idea that poker is a “mental game”; that it is one undifferentiated skillset. But in theory, every video is an opportunity wholly to inhabit the mind another poker player, and allows you to test your own physical and mental boundaries in a safe environment. All it takes is visualization and commitment.
Remember, again, the importance of fear and mental boundaries. Respect the difficulty it takes to click a button. You might think it’s something trivial, but the difference between making a bluff that popped into your head and not making it is often simply having experienced having clicked that button before. If you’ve seen a certain bluff made in many videos, yet have never made it yourself, it is simply the fear of hitting that button that holds you back. Respect the physicality of poker! Something as simple as rehearsing going all-in or making a ballsy bluff by clicking a button on a frozen video can make a big difference.
Practice like performance. Anything you can do to make your practicing, studying, and preparation as close as possible to actual poker will maximize your learning gains. Your brain is not as smart as you—it can be tricked easily. Take advantage of visualization, multiple modalities, and elaborative rehearsal.
Finally, let’s look at isolation drills. Earlier we said that if we want to make practice as close as possible to performance, then we should deliberately practice our skills while we are playing poker. Isolation drills provide us a way to do this.
The first and most obvious isolation drill is to simply isolating attention. If you want to practice 3-barreling, make working on this the focus of your session, and pay extra attention every time you barrel someone and there’s a chance for a third barrel. In short, make your session unofficially “about” doing third barrels, and put everything else on the backburner. This might seem simple, but it’s an invaluable tool in any player’s skillset.
Remember the three types of learning? By isolating your attention, you can foreground your theory network or emotional network while playing. Let’s say you tend to expect to lose every coinflip (emotional defeatism). You might play an isolation session, in which you practice mentally imagining that you’ll win every coinflip. If you know you are weak in a certain situation and you know how to isolate and counteract your natural reaction, an isolation session can powerfully enhance your learning and development.
The second type of isolation drill is an artificial experiment. An artificial experiment is one in which you create artificial rules for yourself during the experiment, designed to test or practice a very specific skill. For example, say you have a lot of trouble fighting 3-bettors out of position. You might design an artificial experiment by moving down maybe one or two stakes below your normal stakes, and playing a session where you call every single 3-bet out of position and try to play well postflop. By isolating and continually exposing yourself to the situation, you will quickly increase your sensitivity and experience, as well as force yourself to challenge your preconceptions about what it’s like to deal with such spots. Artificial experiments are not just valuable for gaining experience and improving your understanding of the underlying map of possibilities, but also for challenging and breaking down recurrent negative cognitions, such as “it’s impossible to play well when you call a 3-bet out of position.”
To conclude—if you have a good strategy coach or mental coach who is guiding you in your learning process, ideally you should have a learning regimen that incorporates all of these aspects in tackling your poker weaknesses. But even without a coach, as long as you have the knowledge (and a little creativity), you can engineer a plan to attack your weaknesses yourself. By incorporating ample risk-taking, proper blueprinting, and effective rehearsal, anyone can optimize and speed up their growth as a learner of poker.
There is a vague and shadowy mythology surrounding the Feel Player. He is a genius of mysterious origins. No one can explain how or why he is good at poker—not even the Feel Player himself. He is like a mathematical prodigy hailing from a tiny, illiterate village in India, or a low-born squire who pulls a mythical sword from its pedestal. He is a fluke, an exception. He is not...
I don’t tell people these days that I used to be a professional poker player. It seems like a weighty and cumbersome admission, one that always leads to the same tired conversation. The last time I played a hand of poker was well over a year ago. That life feels like an episode long behind me—an identity I’ve left behind. And yet sometimes, in my dreams, I find myself playing...