$800 -> $85k in 72 Hours: Reflections on Luck and Skill from the Part Time Poker Grind
In the early hours of May 13th, 2024 and after about sixteen hours of play over two days, I achieved one of my poker dreams and outlasted 773 entrants in the RunGood Poker Series $800 No Limit Hold’em Main Event hosted at Graton Casino and Resort to claim the outright win and top prize of $85,780.
“I’m a little dead inside” are probably not among the first words you would expect out of my mouth after winning a tournament. However, I’ve been at this for a while and as anyone who regularly plays poker tournaments will tell you, it is first and foremost a grind. Even if you assume a player has a significant skill edge, the odds of winning a tournament with hundreds of players is very small. Sometimes it can take a while before even good players score a victory – I am very lucky that it came to me now instead of later.
My poker adventure started after my brother got me into the game and I recorded my first live tournament cash back in 2019. I spent my early poker career getting practice by playing online tournaments. Since then, I have chosen to focus on playing live tournaments as the stakes are higher and the competition weaker. I am thankful to be able to do this on a part-time basis while my day job is spent on Uber’s Policy Research team. As much as I like playing poker, the variance can be brutal and it is a blessing to be able to not rely on it as my primary source of income and treat it as a hobby.
At the time of writing, it has been a few months since the win and I am currently winding down from a summer World Series of Poker stint in Vegas. I’ve found this to be a good point to reflect a little bit on my poker journey as a whole and use this as a starting point to write more about my hobbies. For this one, we will focus on poker and how the hundreds of thousands of hands that I’ve played over the years has given me a few perspectives on the role luck plays in our lives – especially in domains where luck can ultimately play a larger role than skill in outcomes.
Results are for Viewers, Process is for Players
I’ll be the first to admit that the title of this article is purposefully clickbaity. People are conditioned to respond to results as the primary piece of evidence for whether or not someone is qualified to opine about a topic. It turns out that this is generally a well calibrated belief! In many of life’s activities the people we view as experts are the ones who have a track record of success in that field. Being results oriented can be a reasonable way to model the world when results are a good signal for skill.
The results for most competitive disciplines that we encounter in life are primarily skill driven. Take the following examples:
A professional athlete will win the vast majority of the time against an amateur athlete in most sports.
A chess.com player who has an ELO 200 points higher than a competitor will have a ~75% chance of winning.
Someone who studies more for an exam is likely to score higher than someone who is less studied.
However, this world view breaks down catastrophically when results are no longer a reliable signal for skill. There are a number of activities where luck can play a larger role than skill in determining outcomes. For example:
Founding and/or joining an early stage company / startup
Trading individual stocks / options / crypto
Poker
A person’s results in these activities are a much weaker signal for their skill level. You have probably seen or are at least aware of self promotion grifters who attempt to sell courses or a personal brand based on their short term results in high variance disciplines like options trading, crypto “investing”, or building a social media following. In fact, I would not fault you for levying the same criticism at me! I hope the difference between me – a self admitted clickbait artist – and the average grifter is I can at least offer an important piece of advice when assessing claims made by people who work in luck dominated activities.
When results are no longer a reliable signal, the only thing left is the process.
Take the poker tournament I won as an example. As a baseline, let us assume that poker is a game that is completely luck based. For this particular tournament there were 773 entrants which means my baseline likelihood of winning is roughly 1/773 or just about 0.13% if we assume there is no skill involved. Now let us be more generous to me – a player who has spent a non-negligible amount of time studying poker with the aid of solvers and a study group of strong players – and assume that I have a significant edge (to be clear this is a pretty strong assumption and you would be right to be skeptical if some random person on the internet claimed this). If we were to quantify that edge as being three times more likely to win that puts us at… just about a 0.4% chance of winning which is still extremely unlikely. We would also have a higher likelihood of making it further into the tournament with that edge. If you do the math on the expected value of playing the tournament given our edge assumptions and the payouts table, having an edge equivalent to “being three times more likely to win” results in a very positive expected value decision to play the tournament. However, the actual odds of winning are very, very slim.
It should be readily apparent that the luck factor dominates skill in a poker tournament. Even if you gave someone an insane edge equivalent to a 10x chance of winning this tournament compared to baseline they would still “only” win 1.3% of the time. Therefore you should heavily discount this result as evidence of my skill level in playing poker. A common fallacy when assessing skill in luck dominated disciplines is thinking “well if the odds were so against them then they must have been skilled to achieve a successful result!” This misses the fact that, statistically, over a large enough sample someone is almost guaranteed to get very lucky. Mechanically, someone has to win the poker tournament. If you flip a coin enough times, you will eventually flip heads ten times in a row. Is this evidence that you’re good at flipping heads? Probably not!
For these high variance, high luck factor activities, the process itself becomes the primary signal for whether or not a player is skilled. Anyone can win a hand of blackjack, but not very many people can describe the ins and outs of a correct card counting strategy. Anyone can win a poker tournament, but only a skilled player can describe a comprehensive strategy for a particular scenario. Once you have identified an activity as being more luck than skill driven, a person’s process for the activity starts to become a much stronger signal for whether or not they are actually any good.
Unfortunately, identifying processes that are a signal for success can, in general, be difficult. The most common process assessment strategies are by seeing if that process has worked for others individually and in aggregate as well as looking to domain experts for answers. In domains where luck plays a larger and larger role in outcomes, these assessment strategies can fail. There are plenty of scam artist gurus who run social media ads claiming they have a system for crypto / day trading / sports betting / real estate that you can only learn if you buy their course. They try to exploit our process assessment strategies by cherry picking or straight up fabricating “positive results” for them and their students as well as self proclaiming that they are an expert on the topic based on those results. It can be hard to figure out who is right and wrong when we are not already experts in that domain ourselves.
Outside of “common sense” advice (e.g. “don’t fall for get rich quick schemes”), I don’t have a cure for how to separate bad processes from good processes in luck dominated activities. Even for poker, which is a structured game that has well defined rules and outcomes, it can be challenging to separate poor thought processes and good thought processes that generate a particular strategy. I have certainly received my fair share of bad poker advice and the only way that I have been able to see through the garbage is to build a strong foundation for understanding how no limit Texas Hold’em works and to have a trusted poker study group that is passionate about understanding the game.
This is probably the single most important and generalizable observation that I have had from participating in the poker ecosystem. If a party claims to be skilled at an activity because they have had good results in that activity the first questions you should ask yourself are:
How much does luck factor into the results? Is this an activity where you can be wrong and still succeed?
What is their process, can the process plausibly explain their positive results, and is it repeatable?
The more luck there is, the more process matters. For domains where luck dominates skill: the results are for the viewers, the process is for the players.
Volume, Volume, Volume
As some of you may have already rightfully identified that there is, perhaps, one method of identifying skilled from lucky participants in games of skill which have a large luck factor. That method is to see if their process works across a long period of time or if they are able to reproduce positive results using that process over a very large sample.
In theory, over a long enough time period with enough trials, we could start to figure out who is actually skilled, but for many of the luck dominated activities I’ve described before you wouldn’t be able to get to a high enough sample even in a lifetime. Evaluating someone’s skill in making career decisions is pretty much impossible – it’s not like we can simulate hundreds or thousands of decisions at every point in someone’s career and then compare outcomes to see if they have an edge against their cohort. Most folks are probably not making more than a few dozen choices on whether or not to accept a new job or to stay at their current gig over a lifetime. Founding a startup or company is in a similar bucket – how many companies can one person found in a lifetime anyway? Certainly not enough to definitively prove that they are skilled when by far the most common result is failure to return anything to investors.
What makes poker an interesting point of comparison is that, while very complex for a game, it is far less complicated than modeling someone’s career or starting a business. The payouts for poker tournaments are known quantities and with some simplifying assumptions we can model distributions of outcomes for tournament players. In simulations using Primedope’s poker tournament variance calculator, even with a decent edge, after playing 1,000 poker tournaments (this is 10 years of play assuming 100 tournament buy-ins per year) with an average field of 500 players there’s still a ~33% chance that you lose money. Clearly the long run sample requirements in tournament poker likely exceeds the total number of tournaments that a casual player might play over a lifetime!
This poses a somewhat uncomfortable question for the participants in these activities. How can we know if we are skilled if we effectively can never get to the long run? How do we mitigate the risk that even if we have a true edge we end up in a world where we end up losing? What we would like to do for these high variance, high luck factor activities that we believe we have an edge in is to increase the volume of attempts we get to smooth over when we get unlucky. The more chances we get, the less that luck will have on the overall outcome.
Poker players have a few solutions to solve this problem of getting in more volume. The first solution is somewhat brute force: getting in more volume by playing online. Playing online offers a number of advantages for the winning poker player. For one, you don’t have to leave your home and incur any travel costs of going to a physical location to play. The second is that you can get in a ton more volume by simultaneously playing multiple tournaments and hands at the same time. While the live poker player is constrained to playing one tournament at a time, the professional online poker tournament player can churn through dozens of buy-ins a day looking to exercise their edge over a large sample. This order of magnitude difference in volume allows online professionals a good shot at realizing their long run edge.
The second solution that poker players use to increase their volume is to buy pieces of other poker players’ action in what is commonly described as staking. Staking allows one player to assume a share of another trusted player’s wins and losses without having to play themselves. You would be right in thinking that this is a form of diversification - by betting on other players whose play is largely uncorrelated from their own, the staker is effectively increasing their volume of play. In fact, there are some poker players who no longer play professionally themselves and instead manage a “stable” of players that they coach and own pieces of.
Each of these solutions generalize in some way to the other examples of career and startup founding that we’ve been using. Finding ways to increase your volume in your career can be as simple as choosing to be more deliberate about interviewing and giving yourself more choices as well as seeing what the market is like. Personally I try to do a recruitment push once a year just to see what is out there even if I am content with my current role. For startups, getting more volume may mean iterating faster or being more willing to pivot as a market is explored.
If the staking solution sounds awfully like diversification in investing it is because the fundamental principles are the same. In our careers diversification may mean learning a new skill or even managing your finances in a way to make it uncorrelated with your job. This is also roughly the model of venture capital funds in the startup world: instead of founding startups themselves, they try to get to a long run level of volume by making bets across dozens of companies.
It is somewhat sobering to know that my live tournament poker career is unlikely to get to the volume needed to get to the long run even across a lifetime of playing. That’s what makes the wins special - being rewarded with a very unlikely outcome for the hours of work put in to build an edge in this game. It also speaks to my motivation for playing poker in the first place. I genuinely enjoy studying and playing the game and I think that can ring true for others that participate in skill based activities which require a degree of luck for success. We love the process and we love trying to put in the volume to demonstrate our edge.
Games have always been a source of immense joy and fascination for me. They can serve as an arena where many of life’s nuances are abstracted away and we get to compete or cooperate in a controlled environment. It is not a surprise that games and sports are so often used as metaphors and points of comparison for other more complicated areas of life. Poker, with all its complexity and variance, is in my mind the ultimate game to express the role that luck plays in the competitive aspects of our lives. It is rare to have a game where one can be punished so dearly for playing perfectly or rewarded so handsomely for playing poorly. It has taught me a lot about how we perceive people who are wildly successful, how we are meant to harness the advantages we are given in life, and to be grateful for the privilege of being able to play this game for more than lunch money.
Thanks for reading! If you made it this far, I hope you have taken away something useful or found something interesting about this write up. Poker can be a very solitary game – shout out to the folks I study with who have made this journey less lonely and have time and time again made me a better player – so it is nice to have the opportunity to put some thoughts that I have had bottled up into the world. If you are a poker player or participate in a domain that has a high degree of luck involved, I would love to hear what you’ve learned in your experiences - especially on how you’ve built an edge and how you’ve increased your volume to weather out the swings.
I have a few ideas percolating on what to write about next - it will likely be some more poker related content, exploring my sports betting journey, or thoughts on what it takes to achieve Partner status as a streamer on Twitch. Definitely not a coincidence that all of these have an element of luck involved in finding success. Until then, take care folks!
Special thanks to the friends who reviewed and provided invaluable feedback on initial drafts. Shoutout to Nick (@dollarsandata) who has been steadily encouraging me to get these thoughts down over the years until I finally did it.
Also, had to really make sure everyone knows you're not a pro LOL
Nice post! This should be automatically sent to anyone who has something good or bad to say about poker results