Trimming up Your Poker Strategy with Occam’s Razor
You Could Learn A Lot From A 14th Century Franciscan Friar …… And not just about theology, life before the printing press or how to pull off those haircuts where the crown is shaved, but a ring of curls remains along the edges. You could, in fact, learn a decent amount about poker, at least from the particular friar I have in mind: William of Ockham. He’s the logician credited with what we today refer to as Occam’s Razor, a principle that can be loosely summarized as “all things being equal, the simplest explanation is usually the best.” He’s also the reason why IT nerds and philosophy wonks own shirts that say “I shave with occam’s razor.”
Ever think so much in a given situation that you find yourself on the brink of mental paralysis? Let’s say you’re facing a massive bet on the river, which looks to you to be an absolute brick of a card, from an opponent who has been passively calling the whole hand. The pot is huge. You hold something marginal but defendable, say split two pair, you have position, and you‘ve been betting the whole way. Suddenly this opponent, who up until now had seemed clearly predictable, appears to you as virtually indecipherable, and your brain is just a bit less overloaded than an Amish kid standing in the center of a Best Buy. Other players stare at you with thinly veiled disdain tinged with impatience (of course the answer is obvious to them - they’re not in the hand), you riffle your chips unsteadily and it dawns on you that you really can’t make a decision. You will make one, of course, but you’re starting to think it’s going to be a bit of a random act, not the output of the cool and calculating poker master you imagine yourself to be. In short, it’s a decision that will probably sound a lot like: “Screw it, I … [insert whatever comes out of your mouth when you open it], because all of your options seem equally reasonable.
In this spot, you’re simply lacking a tool to help you choose between all the competing options [and the logic behind those options]. This makes it almost impossible for your brain to reach an informed conclusion - so it just draws any old conclusion to get the problem out the door. This is clearly a terrible way to make decisions. Sometimes you’ll be right, sometimes you’ll be wrong, but the outcome is really pretty irrelevant when the process is so deeply flawed. Some people might call this type of decision making “going with their gut”, but the truth is that intuition plays a very small role in impulsive decisions and confusing the two - impulse and intuition - is an easy way to become a pretty bad poker player. While that’s an important idea, it’s not the focus of this article, so let’s bring it back around to William of Ockham.
William would not be surprised that you’re having a problem making your decision. He might wonder why you’re not speaking Latin, but once that got all worked out, he’d appreciate your dilemma and identify the root culprit as we already have - a failure to recognize which bits of your analysis are relevant and which are not. He’d invite you in for some mead or ale or whatever they drank back then and ask you to outline some competing explanations for your opponent’s behavior.
Here’s the first place Occam’s Razor can help out your game, although it is admittedly drawing more on the general spirit of the idea - simplicity is good - than the principle itself. Rather than immediately spouting out theory after theory for why your opponent bet, remember that there really can be only three reasons - your opponent wants you to call, wants you to fold, or wants you to raise. What you need to do now is assemble a quick logic for each possible reason behind your opponents action. Let’s say, for the sake of simplicity, that you had pegged this opponent as a fairly straightforward player. With that in mind, here are some very generalized logics for their action.
1) My opponent wants me to fold: First, we’d have to assume that your opponent is acting a bit of character by bluffing. Next, we’d have to assume that your opponent has picked up some information about you that has made them think you can lay down a hand. Also, we’d have to assume that there was something on the flop and turn they thought was worth calling for that did not improve on the river.
2) My opponent wants me to call: Here we’d have to assume that your opponent had already made a hand before the river hit, and quite likely before the turn hit. We’d also have to assume that they believe there is a good chance you won’t bet again, but at least some chance you might call.
3) My opponent wants me to raise: Here we would have to assume everything from “to call” above and also that your opponent has made several assumptions about you: first, that you are capable of viewing them as a very tricky player who can lay down after making a massive bet; second, that you are capable of interpreting a large river bet as a sign of weakness; third, that you have a hand strong enough to raise with (if you’re not bluffing on the re-raise).
Over-simplified? Sure, but the point remains. The best decision in our hypothetical is clearly coming down to a race between options one and two. Remember, what this razor wants to do is cut out the unnecessary, to reduce the number of variables and assumptions involved in reaching a conclusion. Option three simply asks us to undertake too many assumptions to be attractive in the face of two equally predictive, simpler logics.
You might have two questions at this point - where’s the value in all this work, and how do I use this principle in more complex, game-time settings? Let’s take them in reverse, starting with the application issue. Obviously, the example in this article is tailor-made to articulate Occam’s principle, and not all of your real-life situations will be so clear cut [that’s not to say that a lot of them won’t be - poker is arguably a pretty simple game most of the time]. So, a few things to keep in mind when you’re working this principle into your reasoning:
Occam’s Razor is not a corrective tool for bad assumptions. Look, all of the fancy-sch mancy logician’s tools in the world won’t save you if you’re working from a bad root - if you think a player is a donkey and they’re really a pro, game over. Any correct decisions you make will be in line, statistically, with the number a monkey would make. The only difference will be less feces flinging on your part [hopefully]. So, if you’re making decisions with this tool and you’re coming up on the dumb end more often than not, don’t blame the razor - blame the data you’re feeding in. As my computer programmer friends say, ‘garbage in, garbage out.’ Oddly, my garbage collector friends do not say this.
Occam’s Razor is not meant to be a simple scale for weighing the sheer number of assumptions on each side. You can’t cheat the razor simply by piling up extra, superfluous assumptions on the side of the decision that you want to avoid. This is a tool for weighing useful, substantial assumptions and it only works when that’s what you give it. For example:
Relevant assumption: This player would have to have a reason to view me as weak in order to make this play as a bluff
Irrelevant assumption: This player probably watches a lot of History Channel.
Last, and probably most importantly, Occam’s Razor is just one more tool in your decision-making arsenal. It should probably be thought of more as a compliment to other tools, a tiebreaker, than a stand alone solution for your toughest poker decisions. Remember, it does not generate logical explanations for you - it only helps you to choose between competing decisions of roughly equal plausibility. Say that last bit again - of roughly equal plausibility. This tool, then, comes into play only when you feel that you’ve reached a bit of an impasse, when you’ve already done the work required to generate two [or more] competing explanations for the situation at hand.
………
So now that we’ve covered what it can’t do, on to what it can. Where’s the value in this tool? First off, Occam’s Razor can function as a handy red flag that alerts you when you are attempting to rationalize what you know is a bad play. Everyone, at least on occasion, gets stuck to hands they know should’ve been released a round ago. If you’re anything like me, you’ll often find yourself attempting to convince yourself that continuing with the hand is a sound, rational decision [given the size of the pot of course, and the fact that you haven’t won a hand in an two hours, so you’re due, damnit] even though you know it pretty much isn’t. Sometimes that voice sounds pretty convincing though, doesn’t it? It’s a trick. Usually what that voice is doing is just constructing a possible, though not often plausible, scenario in which you’re going to win the hand. Luckily for you, possible although not plausible scenario construction requires a greater number of assumptions than plain old plausible scenario construction, so when you hear thoughts running through your head at a critical decision point that sound like this:
“Now I’ve never seen this guy bluff before but he did make a comment about 30 minutes ago about being bored and I think he said something about 3 hours ago when I mucked my cards after making a big raise on the river, plus he seems like he wants to shed his tight image that he’s had all day so I can see him calling for the runner flush draw on the flop… etc”
When you hear those thoughts, the razor should remind you that any explanation which requires an inordinate amount of assumptions is, on face, suspicious. Why should you be suspicious of your own logical prowess? Never forget that most poker players have to believe they can win more than their fair share of hands in order to be good players and, when you get right down to it, most of us are kinda greedy to boot. These are both forces that can skew your thought process and actually focus a large part of your mental abilities on rationalizing amazingly poor decisions, but a little poke with the razor can often help to keep them in line.
I think there’s additional value in the fact that Occam’s Razor gives you a reason to make a decision. I know that might sound a bit odd at first, so let me try to clarify. Think about the situation described way back at the start of the article. It’s not an uncommon one, every day I know that I make several decisions just to make them, just to have the tension of the situation or problem resolved. Again, there are a ton of flaws with this decision-making model, but the relevant one here is a lack of replication. Since you made the decision randomly, you can’t be sure that you’ll make the same random decision the next time. If you can’t repeat the process, it’s awful hard to troubleshoot it when you find your decisions going consistently wrong. If you have a reason for making decisions, then you at least have a starting point for diagnosing the problem. So, if you always use the razor as your tiebreaker in those close call situations and you’re wrong more often than not, you can start to address the problem by making sure you’re applying the concept itself correctly. It might not sound like much, but I’m sure there’s some cliché about the first step in solving a problem and how it is the most difficult step - and you can be sure that if I knew that cliché or if I had access to the internet while I was writing this, you would have gotten the cliché instead of all this. Wait, is it “Problems seem impossible to solve until you start”? No, that doesn’t sound right. Well, fill it in yourself then.
With all of the hoopla surrounding poker right now, I’m sure it will only be a matter of time until some fancy philosophy professor / mildly successful poker player decides to cash in and write a Tao of Poker-esque book in which he or she applies the concept of Occam’s Razor to poker in a far more elegant and insightful way. When that happens, I will remind people that my article came first, was funnier, and a good deal shorter. Also, I made a good faith effort to keep the shaving puns to a minimum. And hopefully, I got across the point that, for all the potential complexity involved, poker still remains a game where the shortest distance between two points is very often the straightest and simplest line.
