The Poker Player’s Manifesto — Part XIV: Variance II

Now that we’re friends with variance after our last discussion on the topic in Part 13 of The Poker Player’s Manifesto last issue, enough time has passed that we’re pretty sick of our new friend. Completely disloyal, we often think that even though our friend variance enables our ability to win at poker in the long term, the jerk also enables us to lose in the long term. He’s like that train wreck of a friend that you’ve known since you were a kid and love to death, but he certainly makes you dislike him in moments. After years of understanding my friend and those moments, I understand that being a train wreck is simply part of his personality and I can choose to be his friend and accept him for who he is or stop being friends with him. I still like my friend, so I will take the good with the bad. I still like winning at poker, so I have spent most of my 12-year professional career accepting the train wrecks that come with natural variance in poker.

A dozen years ago, I thought a couple hundred big bets was sufficient to four-table limit hold’em online. I thought I was right for a while as I kept winning. Eventually I hit a downswing and lost that bankroll. Wondering what went wrong, if maybe I was just lucky for those first few months, I fired my last $55 in a sit-n-go and won it.

Fast forward three years to the World Series of Poker. I decided to give tournament poker a go because I wanted to land one of those lucrative endorsement deals back when money was growing on trees. I finished second in my first event, the $500 buy-in Casino Employees event. That payday earned me $66,000, instantly boosting my bankroll 1,100 percent. By the coming summer I had spent or lost most of that money, then I cashed for another $140,000 taking second in the $1,500 Omaha eight-or-better event. I didn’t cash again that summer, and I went into 2008 with two career WSOP cashes in about 30 tries, both for second place.

Come March of 2008 I was broken down to peanuts of a bankroll once again. I was backed at the time and traveled to Reno to play in the $7,500 World Poker Tour World Poker Challenge at the Grand Sierra Resort. I lost most of my cash playing Chinese Poker with Shaun Deeb, then I lost the rest at dinner playing credit card roulette, and then I went on to make the money in the tournament. The first hand I play with 18 players remaining gets me involved in a pot for the chip lead. I have a set of tens on 10-9-7 rainbow and am all-in against a 7-6 offsuit that covers me and a 5-5 that I cover. The turn is an eight, and I retired in that moment. I knew I was good enough at poker but it wasn’t working out, now I’m broke and deep in makeup, and summer’s coming up so I’ll go back to rafting. Then the river was another eight. I un-retired, went on to take second for $271,000 and closed out 2008 with my first million-dollar year.

It was around that time that I started doing some actual work into investigating the scientific nature of variance and how it applies to poker. The most shocking study I did was simulating 1,000 identical years in the life of a professional poker player. Each player was assumed to play 500,000 hands of $5-$10 limit hold’em with an earn rate of two big bets/100 hands played. If a player is four-tabling, this accounts to 1,250 hours of work in a year, a very attainable number. This gives each player an annual expectation of $100,000. What happened frightened me.

A pretty looking bell curve formed off of a high point at $100,000 won after 500,000 hands. One lucky guy won $400,000 in a year while another guy lost almost $100,000 on the year. Those two guys were identical poker players playing identical volume in identical games. Most alarming to me were the swings involved. Nearly 20 percent of those players experienced a downswing that year of over 700 big bets, while almost 30 percent of them lost at least 500 big bets over the 500,000 hands.

So, now I cannot use my results to prove that I am a good poker player. My million winning hands online is a nice piece of evidence, but it alone does not make me good at poker. My live tournament results are a joke due to the statistical insignificance of things. Live tournament poker is extremely high variance and I firmly believe that there is no such thing as a “long term” that can be realized in one’s lifetime. My live cash game results can similarly be argued due to the nature of how few hands one gets dealt in live poker.

At this point in my poker career, I can take solace in the fact that I have stood the test of time. I have withstood four major downswings, the most recent of which lasted 15 months. I am a long-term winner, but even still my results do not make me good at poker. What makes me good is the way I think about the game, and the way I approach difficult problems both on and off the felt. We know that good decisions lead to good results, and we know that we cannot control what happens with the cards after we make a decision. Therefore, I continually focus on making the best decision I can and trust that the results will follow.

The other big thing that data pointed out to me was that my concept of bankroll management was silly. If 30 percent of the years I play I am going to lose 500 big bets in one downswing, then a 500 big bet bankroll is not nearly enough. I believe that I have stood the test of time in poker because of my exceptional bankroll management skills rather than because of my poker ability. Although they are certainly complementary traits, I have seen many poker players better than me come and go because they couldn’t make it through one of Variance’s train wrecks. In the next issue I will discuss optimizing bankroll management in the context of variance. See you then. ♠

Bryan Devonshire has been a professional poker player for nearly a decade and has more than $2 million in tournament earnings. Follow him on Twitter @devopoker.