The Poker Delusion

The Poker Delusion

On the Irrationality of Poker Players and Mark Up

Andrew Slack

Take a second the next time you are in a poker tournament and ask yourself how many people in the room actually have a positive expectation to make money.  Now ask yourself how many players think they have a positive expectation to make money.  Finally ask yourself, “Do I have a positive expectation to make money in this tournament?”  

We already know the answer to the third question, of course you think you have a positive expectation (to forthwith be referred to as being +ev).  You ponied up the money to enter the tournament and it’s doubtful you did it out of the goodness of your heart to pad the bankrolls of those players that are better than you.  Following this line of logic we have the answer to our second question, nearly every single person in the room with you thinks they are truly +ev (with rare exception for those that admit they are not, and just want the experience or to learn from other players).  The answer to the first question is actually the most obvious, but disregarded by nearly every poker player I have ever met: if the tournament is paying out 10% of its participants (a fairly standard payout rate throughout tournament history) than 10% of the field actually has a positive expectation, and 90% do not (they are, in fact, -ev).

So why do we have a room of 1,000 people who all think they have a rightful claim to the prize money, when in fact only 10% of them do?  Psychology tells us that learning at its most basic level requires two things: frequent practice and immediate feedback.  Frequent practice is not a problem in poker, even with the de facto closing of online poker for Americans most can still find places to play with relative frequency.  The main issue is that poker, by nature, masks feedback to the point that it is almost useless to amateur players.  Making a patently incorrect play is often rewarded and the opposite holds true; time and time again you will make the correct play and not only find a lack of reward, but an immediate punishment.  Take the following hand as an extremely basic example.  You are dealt an ace and a king (AK) and have the opportunity to put your entire stack on the line against a player who accidently went all in then exposed their cards as have a jack and a queen (QJ).  In theory you have 64% equity in the pot (your opponent has 36%) and you would correctly be happy to find yourself in this position, quickly calling the all-in wager.  But 36% of the time you will lose or tie (usually losing) and the feedback your brain will receive is negative, in spite of the fact you made an obviously +ev move.  This example is extraordinarily basic, most of poker is far more nuanced and receiving “incorrect” feedback on decisions that are not obvious makes learning to play poker correctly difficult.  

Showing up to the casino for your first ever session of poker, against a table in which every player is better than you, and still walking away a winner is quite common.  You play how you think you should (but is theoretically incorrect in many ways) and are rewarded.  It is now easy to see why so many young men and women truly believe that they are +ev when many are in a habit of making plays that are laughably -ev to experts.  

A second feedback problem that poker offers is an immediate out to disregard all the correct information you receive by blaming it on luck.  When a player makes obviously incorrect choices and loses a pot they still have the opportunity to blame it on bad luck.  This is especially true for recreational players who don’t have the capability yet to understand that the plays they are making are -ev and often think that losing plays are actually winning plays, attributing their loss to unlucky cards.

The third feedback issue is that of sample size.  In poker it takes an incredibly large amount of hands played to come to any accurate assessment of your actual skill level.  Playing 5,000 hands of poker sounds like a large amount to anyone outside of the poker world, but you will quickly be laughed out of the room if you cite your last 5,000 hands as proof that you are a huge winner with a high winrate that you can “prove”.  All players, even good ones, underestimate the role that variance plays in poker results by a large degree and are quick to come to the conclusion the are +ev when they are not (or at least do not have nearly enough played hands to prove it).

So this brings us back to the main point, when you are sitting in a tournament room the level of delusion in the crowd is immeasurable; comical to those with a dark sense of humor.  Thousands of hours and dollars are being wasted around you by those under the false pretense that they have a reasonable expectation in the tournament when in fact only one out of ten does.  

This on its own is distressing, but a required part of poker.  Losing players are needed for there to be winning players, and most losing players do not understand the extent of their situation (in fact, as previously discussed, they don’t know they are losing players at all).  It’s less distressing when you understand the fact that everyone has the opportunity to become better.  Poker books still line the shelves of your local Barne’s and Noble and poker tutorial websites are easy to find with a quick google search (in fact, the best information available is often free, if you frequent the correct poker forums and take the time to find out who you should actually listen to).  So all hope is not lost, and I don’t think there is anything immoral about playing poker as a winning player, but it is a disturbing look into the human condition no matter what way you cut it.

Here is a quick heuristic to use: if you are not the best player at your 9 or 10 handed table, you probably are not +ev in the tournament ( not always true by a large stretch due to the variance of table draws, but a decent starting point).  

So what can you do to make sure you are not one of the 90%?   At the risk of sounding cliche, the first step is to admit you have a problem.  The vast majority of players will never admit that they are probably -ev, a side effect of conflated ego’s brought on by pots won that should have been lost.  If you sit to yourself and think with honesty “I may not be +ev after all” the truth is you probably aren’t.  In fact if you answer that question with “I’m certainly +ev” you probably still are not, it’s hard to overcome the base rate of 90% of players being losers  To be in the top 10% you need to work harder and play better than the other 90% (and a high level of intelligence and self control is probably required as well).   A chinese proverb goes as such “The best time to plant a tree was 30 years ago, the second best time is today.”  If you have been playing for a few years, assuming you were +ev, but not doing much studying, you need to start today (or quit poker as a source of income; playing for entertainment with the understanding you may be a losing player is fine).

But things get even worse for the poker community.  In recent years the practice of charging markup to sell action has become commonplace.  Markup in and of itself is not a problem, it only becomes problematic when investors are paying a huge premium to invest in something that is -ev.  Markup works like such: a player wants to sell pieces of their action in a tournament that they believe they have a substantial edge in.  They deem their return on investment (ROI) to be +30% so they ask that investors pay them an amount above $1 for every dollar in action the investor receives.  If their true ROI is 30% then an investor paying $1.30 for $1 in action is making a breakeven proposition.  If they are able to pay only $1.20 (or any number under $1.30) they are making a +ev bet, and anything above $1.30 is -ev.  If you could accurately ascertain a winning players ROI in a tournament, this would be simple math and investors would not have many problems.  This is simply not the case, sample sizes are way too small, and the overall skill level of the average tournament player is rising too quickly, for us to get anything close to accurate predictors of true ROI for most players.  Yet players, many of them big names and relatively smart guys, solicit exorbitant markup and (usually unknowingly) shaft their friends and investors.

These two problems conflate each other very quickly, we have a massive amount of people asking for (and receiving) inappropriate markup that are not even +ev in the tournament to begin with.  So it is possible to find a player who is being invested in for 40% markup who is actually in the bottom 90% of the field, a double knock out punch to investors as a whole.

The defenders of markup as a commonplace practice cite that we are simply in a marketplace, in which the price of markup can just be decided by the market and selling at that number is ok.  The efficient market hypothesis stands on two legs: “The Price is Right” and “No Free Lunch”.  “The Price is Right” dictates that the price of a good is dictated by all information available and ends up with a correct price for that product.  Watch news break about any publicly traded company or commodity and the stock approaches a rational price almost instantaneously (with today’s technology, many reach this price in milliseconds).  The “No Free Lunch” portion dictates that noone can beat the market over the long run (mostly because of correct pricing).  For our discussion of markup in the poker marketplace we will focus on “The Price is Right”.  You can find countless examples of irrational prices if you look for them.  While browsing twitter today I read two tweets offering the sale of action within one week of each other.  The first tweet was asking for 3 to 1 on a tournament, and then a few days later the same player was soliciting sales at 4.4 to 1.  Both of these prices are laughable, investors would have to be sure he has a 300% and 440% ROI respectively to make it break even, but that’s not the point.  The point is that he was selling (and supposedly received action) on two different prices with very little change.  The tournaments had similar structures and were the same game format, I have a hard time believing this player added 110% to his true ROI in a few days.  This same player sold action to the same tournament at both 2.5 to 1 and 1.5 to 1.  Both of those prices cannot be simultaneously correct.

It would be nice if we had anything near an efficient market, but we don’t.  Most investors have nowhere near all the information available to them to come close to an appropriate price and are in general paying way too much.  There are many other issues with markup as a concept, namely how a player who has sold a lot of action at a steep markup has a different set of incentives on how to play in certain stages of a tournament that are directly opposite to their investors well being.  We also have an idea from economic theory that as stakes rise people should act more rationally, selling action inherently lowers the stakes for those selling and would theoretically dictate less rational play.  I won’t get into all the issues with markup here, but ask professional poker player and economics student Andrew Barber about it and he will probably be more than happy to share.

I will offer two caveats to this issue that I think are important:

1. Charging markup in and of itself is not immoral, and those that offer incorrect prices are usually not doing it maliciously.  Ignorance isn’t a good excuse, and it should be shameful to sell at exorbitant prices regardless of the reason, but it’s important to distinguish malicious behavior (of which there is little) from ignorance (of which there is much).
2. Buying pieces at markup, just as playing at a table in which you know you think you may be -ev, is fine if you are ok with paying a premium for the entertainment value and are cognizant of what you are getting yourself into.  The point of this paper is simply to make people aware.

I buy small pieces of freinds frequently and at times for small amounts of mark up.  While I would never buy at 300% markup, I have and will continue to buy pieces at rates that I feel are at least in the ballpark of reality (for microscopic portions of my bankroll).  It is a huge mistake to pay 300% to a guy whose true ROI is 20% and only a small mistake to pay 20% to a guy whose ROI is 10%. The small amount of ev I may pay is made up in life value.  So buying action is ok, but buyer beware.  

If there are two things to take away from reading this they are as follows: don’t overvalue yourself (everyone else is doing it) and put yourself into -ev spots.  And don’t fall for the trap of others overvaluing themselves, before you buy action for any significant portion of your bankroll you need to do serious due diligence and be aware of the situation in every knowable way.