The flat worlds of poker.

The Wrong Paradigm

Understanding our environment has almost always been a major concern for humans. If you don’t understand, you can’t control, and if you don’t control, almost anything can wipe you “out of competition” overnight, so to speak. This is a great source of anxiety, and naturally, or could we say hopefully, the stronger it is, the more we’ll be inclined to do whatever we can to reduce it.

That’s why theories are elaborated.

Theories & Models

Building models is something we do since we are born. Our vision, for instance, is inherently dealing with two-dimension representations of the world; but our brain does the job of converting the 2D model into a three-dimension universe. Sometimes, it can be tricked with optical illusions, but overall it works quite well.

Naturally, some problems are much harder to model than others. The formalism of things as central in our world as light in physics, for instance, isn’t something we established a long time ago. Even our understanding of how our galaxy is organized is relatively recent in Human History.

Before then, we had simpler models, that appear hopelessly flawed by today’s standards, but that worked well enough back then. That’s how models evolve: we make simple, intuitive ones, we check if they represent our world well enough, and if they do we keep them till they comes apart at the seams, when applied in a wider context. The flat Earth model was fine when people didn’t travel very far, but it couldn’t resist any longer when our ancestors started exploring the world on a larger scale.

But, as a corollary, this also means that many models and theories of ours, as obvious as they may seem, can be plain wrong — even if they appear to hold true under certain circumstances.

But We Can See It’s Flat!

Logically enough, many wrong models are used because they are intuitive. That is, they are consistent with all our other models. If you take an orange, you can see that it’s round. Even if you imagine that you were small enough to be on its surface, you would still be able to see the curves. In addition, it seems obvious that if you put something anywhere but on the top, it is inevitably going to fall. In other words, a round Earth contradicts common sense, or, it is counter-intuitive.

But what does make a model intuitive in the first place?

Well, we could say it is intuitive if we can easily guess how it works, based on our current knowledge. This is related to a process called assimilation by cognitive psychologists. In a nutshell, we apply, or we could say extend, existing concepts and knowledge from one field to another. For instance, if you know what a boss is, you can easily understand the idea of captain, general or bishop. Likewise, If you understand what a doctor is, you can figure out what a veterinarian does, or even infer what the job of a mechanic is. Or, as a less trivial example, a specialist of the submarine fauna and animals like sharks and mantas can figure out why many aircrafts are shaped as they are.

This is a very effective way of apprehending the world, since we can generalize pieces of knowledge rather than having to deal with a completely new set of concepts for every single domain. Unfortunately, as for every generalization, there are pitfalls, since it can be tempting to consider two things similar because they look similar, while they are in fact quite different.

As a rule, if two things are different, the model won’t match. You could say that a teacher with his students is like a general and his army, but it won’t take long to show that these are two different things. For the wrong model to live, first it must seem to be a good fit to some depth, and it must also “work” in practice in enough cases so that it looks quite reasonable.

Note that “work” here means be consistent with personal perception and understanding, and this includes beliefs. That is, the model can influence your view on a question, if you find it convincing enough, or conversely you can choose a model precisely because it matches your current view. But in both cases, there is no guarantee that your view be correct!

In fact, some wrong models live because we want them to live. They are intellectually appealing, they can be elegant, or compelling. But they can also conveniently bear out our vision of the problem. The worst enemy of the wrong model is negative feedback, ie. cases when it fails. In a strict scientific context, you can’t push them under the carpet to hide them; but in everyday’s life, there are often many reasons to account for the several mismatches between the model and reality. Even extremely serious professional can get it wrong: ask Argentina what they think of the economic policy advocated by the International Monetary Fund.

The Weird Physics of Poker

As we know, poker is a strange kind of planet, with many physic laws that are quite difficult to apprehend, since they depend on so many factors. It is as though gravity didn’t always work downwards, depending on the weather. Under such conditions, it is difficult to come up with accurate models, but this somewhat paradoxically makes them even more necessary. Thus, most players approach the game with the usual process of assimilation, and compare the game to some domain they know, so as to try to understand it better — or strengthen their position.

For instance, risks are pervasive in poker, and everybody has a different way of dealing with them. In other words, people have different risk profiles. Players who invest money for their family and have a porfolio, mostly consisting of bonds, are generally quite risk averse (unless poker is an outlet for their frustation of not gambling enough!). Likewise, engineers working in a nuclear plant or air traffic control operators generally have a strong risk aversion; one could say they don’t have to, but you generally choose a career in accordance with what you like to do. If you are thinking ten times a day “godd.mn those unnecessary security measures”, these jobs are going to drive you crazy.

Consequently, these type of players are likely to develop the same risk averse strategy in poker. They will play very conservatively, avoid most difficult situations, and won’t go all-in without the nuts. To them, going all-in is like pushing the nuke button. It says it all, doesn’t it?
Of course, they shake their head when they see bold bluffs, whether they fail or succeed in fact; the fools will have to pay sooner or later anyway.

Is it really a bad approach? Well, yes and no. Many times, it’s going to serve them well, especially against inexperienced opponents. This will reinforce their position of course. The thing is, this is not a bad approach in itself, but not one you can apply come hell or high water. Some players are going to give you a rough time if you stick to this very conservative style; you need to be able to adapt to them. Sometimes, going all-in with nothing on the river can be the best play by far. And sometimes, you’ll need to call your opponent down with a light hand, if he forces you to do so.
Peddling the nuts is a luxury you can’t always afford if you want to be successful at poker.

Another disputable approach is that of players who want to control the table almost whatever the cost. Like a manager, they think they need to show who the boss is, so that the other players don’t start being annoying, or even worse. Again, this can be a very nice achievement when they succeed; poker is a game of people, and forcing people to bend can’t be wrong in poker.

Yet, you can be sure many good players won’t comply. And if you persist, you may well end up with much less chips than you started with. If you want to bully your way, the big bets are going to work only so long.

Beyond Simple Models

The problem with these relatively simple models, is that they can work locally, like the flat Earth model, but can fail miserably if applied indiscriminately. This is like the many heuristics we tend to use, like, more expensive means better quality. This is generally true, but a shrewd salesman can exploit your heuristics and rip you off painlessly. Similarly, a good poker player can exploit your approach of the game.

Thus, it is important not to have unfailing faith in these heuristics. Unfortunately, for several reasons, we are often prone to put too much confidence in them. First, they sometimes look so right; this is the same feeling when you work on something difficult to understand, then suddenly everything clicks, and all the chaos appears clear and ordered. Our solution then looks as right as it can be.

However, we can be convinced and still be wrong. A good example is the Monty Hall problem. If you don’t know it, it’s was TV show where you had to choose a box between three, and only one contains the prize. You pick one, then the presenter discards an empty box among the two remaining ones. You then can keep your pick, or change your box for the other one. The question being, should you keep your box or swap, and does it matter?

We’ll let you figure it out and search the web for solutions, but the point is that most people are absolutely certain of their solution, and it doesn’t take long for boisterous arguments or even flames to happen. Well, almost everyone falls for this one, even reputed math whizzes have been reported to be wrong (!) and take part in the argument. Anyway, this perfectly illustrates that you can be dead certain about something, and nonetheless be wrong. This is something we should always keep in mind.

Second, once we find an explanation, we don’t let it go easily. This is a process called anchoring. Studies show that we often fail to question previous hypotheses or preliminary conclusions, and on the contrary we have a bias to support them too long, even in face of elements that cast serious doubts on them. This bias is not difficult to understand, to some extent: when we are strongly convinced of something, we don’t change our mind easily, and if the previous model have been in use for some time, agreeing it was wrong would amount to acknowledge that you’ve somewhat been fooled all that time. This is related to cognitive dissonance. Besides, breaking our old houses cause anxiety, and as we saw we avoid it as much as possible.

As a last example, consider the classic “tight is right” common phrase. People playing in Limit Holdem full ring games have had many occasions to prove it right. Even in NLHE tournaments, playing tight is often quite reasonable, and on their lucky days they can won it all. Well, even if most of the time they bust out near the top third, after surviving for some time. Anyway, they do get positive feedback on several occasions. Why would they change their play, especially if they’ve played that way for years?

With the Internet poker boom, there’s now a wealth of information and discussions about poker, so there’s much more material for those who want to improve their game. Consequently, it has become clear that many quite successful players are by no means playing tight. They play tournaments, but favor short-handed ring games. Some of their moves look very weird, to say the least, to the tight-is-right player. This creates a dissonance, which leads the tight player to find a way to reduce it. He can study what the other player is doing and how it can work — or he can qualify these plays as unnecessary gambles, or he can even figure the successful player is some kind of genius and mere mortals generally can’t win playing that way.

And that’s how wrong paradigms live.

Bottom Line

Realizing the limitations of simple models is a necessary step towards a better understanding of the game, since it shades light on the various combinations of factors that had been neglected so far and that make the simple model work or fail.

(1) comments

Related Items:

• Cognitive Dissonance