 My name's Corey Chamberlain. I'm a software engineer at PagerDuty. Today I'm going to talk to you about poker, postmortems, and life. How many of you here have ever played poker before? Okay, most of you have played. Hopefully this talk is helpful for you and it's helpful for the people who've never played before. I'm not a professional poker player. I work on software. But I've taken a few courses and I've read some books and I've played a lot of poker. It's a serious hobby for me. I am, however, an expert on postmortems, having attended and caused quite a number of them in my travels. And my conclusion is that postmortems would be a lot better if they were run and attended by people who understood poker. So today I'm going to make you into those people. I submit that most postmortems as they exist today make us risk averse and that learning about poker can help us to mitigate. So first we're going to talk about poker. We're going to talk about the rules of the game and basic strategy. That's why most of you are probably here. Next we're going to talk about postmortems and how we can mess them up in unintuitive ways because we're not thinking like a poker player might. And finally we're going to talk about some other areas of life where thinking like a poker player can help you out. At the end of this talk you'll hopefully have a few ideas that you can take with you to your next poker game and maybe you'll gain some insights that you can take with you to your next postmortem. So poker. Poker is a game of numerous variants but it's typically played with the standard 552 card deck. There are usually some number of hidden cards, some number of rounds of placing bets on which player has the best hand, calling those bets or leaving the hand which we call folding. When the betting is all done the player that remains in the hand with the strongest five card hand wins all the money in the pot. The best hand. These are the hand ranks in a game of poker from the weakest hand to the best possible. It's not critical for this talk that you learn them right now or anything. It is pretty important if you want to play poker that you know which ones are better than which. Most are self-explanatory. A straight though is five in a row. A flush is five cards of the same suit. A full house is a pair and a three of a kind and they're like different. So maybe three fours and two fives. A straight flush is both a straight and a flush. The best possible hand in all of poker is 10 jack, queen, king, ace suited. And that's a special straight flush called the royal flush. And you're guaranteed to win if you have that hand. Within the framework of these rules many different variants of poker are played. But for the purpose of this talk we're going to focus on just one. No limit hold them is the most popular variant of poker today. It's called the Cadillac poker. If you buy a beginner book on poker or if you get into a course or you watch poker on TV, odds are that it's going to be no limit hold them. You really kind of have to go out of your way to avoid it. It's such a big part of poker. So let's talk about high level overview how no limit is played. Very first there's a round of blind bets. Two players called the big blind and the small blind put in a bet dictated by the table before they receive cards. These bets are made before each hand to encourage action at the table. So not everybody waits for the very best cards to play. They're usually used to set the buy in for the table and often players consider their opening bets in terms of how many big blinds they're putting in. The small blind is usually about half the big blind. Different tables have different blind levels so online you can get into a micro stakes game with big blinds of like two cents and a small blind like a penny. And then they go way, way up to like a couple hundred dollars or something. Live play usually starts at more like a dollar for the small blind and two dollars for the big blind. The blinds along with the whole betting order rotate around the table each hand. Next, everybody gets two cards. These are called the whole cards. Then starting to the left of the big blind and then working clockwise players either call the big blind, raise the bet or they fold. The blinds then get another opportunity to bet so if you're the big blind and everyone calls your bet all the way back around to you, you get another opportunity to raise. Besides that though, once a bet makes it all the way around to a player who made that bet without being raised, the next stage of the hand begins. And the next stage here is called the flop. Three community cards, they're dealt face up. Eventually we're going to end up with five community cards are considered to be part of everybody's hand. So how it works is you take your whole cards plus the five community cards and you figure out the best possible five card hand that you could make and that's your hand. Anyway, after the flop, another round of betting occurs. It's important to note that from here on out, the player to the left of the button will start each betting round. So that means the button bets last any time after the flop and this is a really big advantage that we're going to talk about in just a minute. Then comes the turn. So after the turn, during the turn we get one more community card and another round of betting and then the final card is dealt on what's called the river and a final betting round occurs. If two or more players are left in the hand, we get a showdown where each player in turn reveals their hand and the best hand wins. The vast majority of hands do not get to this point. Usually one player is able to convince the other that he or she has a better hand and to fold. Most hands even in before the flop. Poker, it's a very simple game in a lot of ways. That's pretty much all there is to it. You can sit down at a poker table and you can pretty much play in about ten minutes. The rules aren't very complicated. The cards will come however they will and the only actual choices that you really get in the game of poker are to call to raise or to fold when it's your turn. However, those choices are enough to make poker a game that's dominated by skill over a long period of time. So what's the skill of poker? The skill of poker is making good bets. A player who makes better bets than their opponents will, given enough time, take all their opponent's money. So what are some ways that we can improve the quality of our bets? Here's a few tactical things. First of all, most beginning players play far, far too many hands. The goal in poker is not to win the most hands. It's to win the most money. You should play hands which play well post-flop. A tight player is going to fold 80% of the time before they ever commit money to the pot. As you improve, you can learn to outplay your opponents with a looser but still aggressive style of play and you'll still probably be playing far fewer hands than most beginning players play. At the end, I'll mention a few courses and books and they'll have recommended hand ranges for opening. But the general idea is it's a lot fewer than you would think. And the ones that are good are not necessarily intuitive. Somewhat counter-intuitively, most beginning players like to call too much. So they like to take weak hands and then call with them. If you have a narrow range of quality hands, you more often than not want to raise your opponent's wider range of weak hands because, for one, they're likely to call because beginning players like to call. And two, your smaller range of hands is likely to beat their hand. For example, many beginning players like to slow play a very strong hand pre-flop like ace, ace, or king, king. But that's the opposite of what you should do. You want to keep raising until you have exactly one opponent left since your hand has great odds against individual hands. But if you have a bunch of opponents, your odds go way down of winning with that hand. Finally, understanding position. Remember when we talked about how the betting order rotates around the table? It turns out that this is very, very, very important. The players who bet last have the most information. Therefore, you should be playing fewer hands in earlier betting positions. And you should be a lot more aggressive in later positions. It's hard to overstate how powerful it is that you get to decide what to do after your opponents have already committed their money to the pot. Let's say, for example, you have a hand like 8-7 suited. This particular hand performs very well in a pot with a ton of collars before you. But it's a really bad hand against a single opponent, since they're unlikely to give you enough money to be worth the slim odds that you hit either a straight or a flush. Suppose that you're the very first player to act before the flop. That's the player to the left of the big blind. We call them under the gun. So how does this play out? Well, you don't know what anybody else is going to do, but you think, this hand is OK. I've got a medium suited connector. That means I can get a straight or a flush from this hand. So you go ahead and you call the big blind. And then the hand folds around to the player before the button. We call that player the cutoff. And that player raises the pot six big blinds. And naturally, everyone, after that player, folds because that player has represented a very strong hand. And so now it comes back to you. And your options are to either call that raise or to fold or, I guess, to raise yourself. You don't really have a choice but to fold, because it would cost you so much to continue to play, and you think that this hand is unlikely to beat a very strong hand that your opponent's representing. In this case, it costs you a big blind to learn the strength of your opponent's hand. Imagine, though, if you'd been on the button, everybody's hands at the table are the same, but you're over here. Everyone folds around to the cutoff, and they open at five or six big blinds, because nobody called before them. And you would know that you don't want any piece of that player. So you can fold right now, and you get to save your money. This little edge of knowing what most of the other players at the table are going to do, played out over many, many hands, ends up being really, really big. If you give an amateur player permanent position over a seasoned pro, the amateur is most likely gonna have a better day than the pro. The state of playing later than your opponent's is called being in position. Playing earlier than your opponent's is called being out of position. So folding enough, raising enough, position play, these are all things, sort of small ideas that help you make better bets. And the way they do that is they improve the average amount of value that you expect to receive from making a bet. That's called expected value. It's how you know whether a bet is any good. It's the sum of all possible outcomes of a decision times their individual likelihoods. Let's do an example. Suppose we make up a game, not poker, it's a simpler game. It costs a dollar to play, and we flip a fair coin. If the coin comes up heads, you win $5. And if the coin comes up tails, you lose the game. Should you play this game? Well, half the time, you're gonna win nothing. So half of zero is zero. Half the time, you're gonna win five bucks. So on average, that's $2.50. If you add it up, you get $2.50 of expected value each time you play this game. Is $2.50 more than the dollar it costs to play? This is the ruby part of the talk. Yeah, $2.50 is more than a dollar. So you should play this game as many times as you're able, assuming that a dollar is not a huge amount of money for you, the more you play this game, the richer you will get, on average. So a poker example. Suppose your heads up versus a single opponent. You hold the nine aid of diamonds. Based on your opponent's betting decisions, you believe it's reasonably likely that she opened with a small pair and she hit three of a kind on the flop. That makes you think she has three sixes. She might also have a really strong queen hand, like king, queen, or queen jack, or maybe even queen 10. Right now you have nothing, but what you do have isn't opened in a straight draw. If you draw either a seven or a jack, you complete your straight and you're gonna win. The opponent has bet $3 into a $7 pot, should you call? Well, if we call we're betting $3, that we can win $13. That's the pot of $7, the opponent's bet of $3 and our bet of $3. So what are the odds that we draw the winning card? Well, there are 52 cards in a deck. We have two known cards in our hand and there are four known cards on the table. We treat the opponent's cards as unknown because we don't really know what they have. So there are 46 unknown cards total. Of those 46, exactly eight will give us victory. That's the four sevens and the four jacks. We call those outs. Therefore our odds of winning are eight over 46 or 17.9%. If you multiply this by our theoretical winnings, you get $2.33 of expected value. Since losing pays out zero dollars, we know this is all we can expect to get on average from making this bet, $2.33. Since it costs us $3 to make the bet, we should fold in this spot. This simple calculation in poker is called pot odds. What are the odds that the pot is giving you to call? There are other ways to extract equity out of a hand. There's implied odds. When you estimate how much additional money you can get out of your opponent if you hit your hand. In this case, maybe not so much since your straight is sort of obvious and the way you're playing is sort of obvious. There's fold equity, which is the percentage of time that you can get an opponent to fold a better hand if you raise. But if you add up all these values together, you get the expected value. One of the big things in poker is that if you suspect your opponent is drawing to a straight or drawing to a flush, you can manipulate the pot odds and induce them to make a mistake by your bet sizing. So you can create a situation where it is incorrect for them to call and they do so anyway. You'll notice in both of these scenarios, the coin game and the poker hand, there's no conclusion to the story. We don't know in the coin game whether you got heads or tails or whether you would have completed your straight in the poker hand. It's for a very important reason. The outcome is immaterial to the strength of the decision. If you give yourself enough runway and if you run this bet a thousand, a million times, over time, that's all that matters. It only matters that you repeatedly make good bets. If someone is offering you four to one money at three to one odds, you should take the bet, assuming that it's a sustainable amount of money for you to bet. You don't bet the house, but you take the bet. You're gonna lose that bet 75% of the time, but the 25% that you do win it's more than gonna make up for the losses. Should you be upset about losing that bet? No way. You expected to lose 75% of the time, but you still made a good decision by taking the bet and you should do that every time somebody gives you the opportunity. Conversely, if someone offers you three to one money at four to one odds, you should not take the bet. And even if you win that bet, you're still a sucker even when you win. One really important way to improve at poker is to run postmortems on your poker sessions. It's really simple to do in poker because it's so numeric. Most online clients record your hand history, including the replay. So if you run into a tough spot and you're not really sure what the right play is, you can mark the hand and you can review it later. And then you can come back after you're done with your session. You can use free tools like Equalab or PokerStove and you can calculate your exact equity based on how your opponent plays and how they represent their hand. And you can figure out what the right play would have been. It makes it really, really easy to see when you made the right call even if it went badly or if you made the wrong call even if it went well. So that's a super high level view of poker and some ideas for poker in about 15 minutes. So what can we learn from this with respect to postmortems? According to PagerDuty's incident response documentation, a postmortem is a blame-free detailed description of exactly what went wrong in order to cause an incident. Usually it involves things like having an investigation into the root cause of an incident, creating follow-up tickets based on the result of the findings, scheduling a meeting with internal stakeholders and developing communications for our customers to let them know what went wrong and what we're gonna do about it. I've read a lot of postmortems working on this talk and it seemed to me that postmortems are really tied up in this question. They take it on faith that if we'd just prepared harder or if we'd investigated more correctly or if we'd done the right things up front that we could have prevented all of this and they take it on faith that anything that we could have done to prevent this in the past is now a critical action item. If we analyzed poker play in this way, we would not be playing poker for very long. It's a little easier to get away with in software because we have venture capital or we have things like stickiness or customer inertia or we're only like a really small part of a big machine or maybe our competitors also haven't caught on to this edge yet and so we're all sort of equally terrible at it. Why do I think this is so bad? There's no room in this model to accept a painful reality not every outage is the result of a bad decision. In fact, sometimes outages will be caused directly by making good choices and some outages will only be preventable by making bad choices. For example, remember the coin flip game? Remember how we did math and we figured out that you make on average 250 every time you play the game and it's right to play. So you go home from this talk and you explain this game to your buddy Ben and he doesn't really buy into the whole math thing and he's bad at money so he offers to be the house and he's willing to let you play this game so you take your shot and then you lose because you're gonna lose half the time and Ben cackles maniacally and he walks away with your dollar and it really hurts and you think to yourself what could I have done differently and how could I have prevented this and suddenly it comes to you there it is playing his day you don't know why you didn't see this before. If these are the rhetorical devices that you use for postmortems the only answer that you can possibly get is that you lost a dollar and you decided to play the game. Everything beyond that decision was out of your control. The only time, this is the only time in all of this when you have an actual decision that can affect the outcome of the game and the lesson that you'll get is that you should never play this game again. This is very wrong. We did math, we did Ruby and we proved that it's right to play. The game has risk but the rewards of the game exceed the risk on average just because you experience the downside of a risk doesn't mean that the risk wasn't worth taking just because a bad thing happened doesn't necessarily mean that a bad choice was made. The term results oriented thinking comes from poker. Well, it sounds awesome on a resume it's not something you wanna be it means that you're measuring the soundness of a decision based on its outcomes instead of based on the information available when the decision was made. It stems from an illusion of control. It's the idea that the choices that we make have a one to one mapping to consequences instead of a probabilistic mapping. It tries to extrapolate lessons from a sample of too few trials. People do this all the time in all kinds of contexts when they say that my grandpa smoked six packs of cigarettes every day of his life and he lived to be 110 or they're disappointed when they've been exercising for weeks and they haven't seen the results that they'd hoped for yet. It's a basic defect in the way that we think. If you're a caveman results oriented risk aversion is probably okay. The world's a dangerous place for you every bet that you make is essentially betting your whole life everything you have. If you decide that a snake is likely to kill you and you shouldn't eat them it's inaccurate, right? But it's reasonable enough given the high stakes involved and the cost of figuring out which snakes are poisonous and which aren't and you got other stuff you can eat, right? However, modern living and modern enterprise require that we take calculated risk. Results oriented thinking will lead to a misallocation of resources. You're gonna try to avoid unlikely or inexpensive failure modes and you're gonna spend a ton of money doing it and we only have so much money we can spend on insurance. This is an existential risk in itself. Here's a good example from the classic television series Star Trek, The Next Generation. In the season two episode, Peak Performance, Data is defeated in the game, Stratagema by third level grandmaster, SIRNA Kaurami. And Data takes it really hard. He removes himself from the bridge. He returns to his quarters and he does the robot version of sulking. And then Deanna Troy comes in and she tries to make Data feel better. She says, look, Data, some of our greatest advances come from analyzing failure. And she says, I know it might hurt your ego and Data's like, hang on a second. I'm an Android, I don't have an ego. Aren't you supposed to be good at this? And then Troy's like, look, you can learn from your mistakes and Data, he says, that's the problem. I didn't make any mistakes. Like I analyzed all my play. I didn't see a single mistake in any of this. And then Troy doesn't have anything to say to this. She just spouts off some platitudes and sort of gives up. And Data thinks that the way he sees the world is fundamentally flawed because he doesn't believe he made any mistakes. So he thinks that the way he analyzed the game itself must be wrong. Pulaski comes in and she tries to give Data tough love. But she also comes off as not really getting it. She's just trying to make him see that it's okay to make errors when no errors were committed. And they go to Picard and they say, look Picard, Data's all messed up and you need to go fix him. And so Picard comes down and tells Data, hey, get back on the bridge. We got a battle simulation real soon and you're the plan maker, you need to go make a plan. And then he tells him this. He says, it's possible that you commit no mistakes and you still lose. And that's not a weakness, that's life. Picard is a good captain and a good poker player. But what he really should have said, if he'd seen this talk, is that Data, you're being really results oriented. So here's a real life example. So sometime ago at PagerDuty, we had an elaborate system for testing our providers. We had a bunch of consumer cell phones mounted on the wall behind me and they were constantly receiving text messages via our providers. The system caused a number of false alarms in its three year lifespan. And as far as I know, it never really alerted us to anything that we didn't detect in other ways. It did, however, wake us up a lot for no reason at all. Eventually, the phones started to degrade from such abuse. The batteries swelled up and the cases bent and they looked like they were gonna explode. The team who owned it considered very briefly, maybe we should just replace those batteries and keep trucking. However, they looked at their considerable experience with the system and they're like, come on. This costs way too much to continue maintain. Let's just decommission this and think about, maybe what could we replace it with in the future? So all is well, we took it down. And then a few months later, no surprise and unrelated outage occurred. And during the post mortem, it was determined that if we'd had this system in place and it was working, that we could have mitigated the outage somewhat. And so the conclusion that the post mortem came up with was that it was a mistake to turn the monitoring off and that in fact, we should double down on it. Not only do we need one wall of exploding cell phones, we need two walls of exploding cell phones. So, was the conclusion accurate? In my opinion, no. It's certainly a risk that there were theoretical outage modes that only this system could detect. However, our approximation of the likelihood of these failures times the cost of such a failure was a lot less than the cost of continuing to maintain the system. And the fact that a failure actually occurred later had no impact on the validity of that decision. It's easy to see though, how results oriented thinking could lead us to believe that taking this thing down was playing fast and loose when in reality we'd identified a small reasonable risk with a disproportionate upside and we'd made a good play. We made a good decision to have a risk, right? So, it's not helpful for me to tell you that post mortems are results oriented unless you just wanna troll people by telling them they're being results oriented. So, I thought of some questions that people should be asking more frequently in post mortems. If they wanna maximize their potential for learnings and minimize the odds that they're learning the wrong things. Question one, should we have prevented this? So, most post mortems assume that all problems are worth preventing. I disagree. I think if you assume this, it will make you risk averse. Not all outages are worth attempting to prevent. If you think that the decisions that led to an outage took into account reasonable risks and took into account all the information available at the time, then the decisions that you made were sound and you should feel good about it. If you time travel and then you fabricate ways in which those decisions were wrong based on hindsight, then it will lead to worse decision-making in the future. Question two, should we try to prevent this in the future? So, this is actually a different question. We know different things now. The stakes have changed. The odds are different and the costs have gone down. Critically, every post mortem involves stakeholders. Sometimes, one's with actual stakes and pitchforks and torches. And if your system experiences a failure, your customers are gonna expect that it makes sense that you're gonna adjust your process so that you'll never experience this particular failure mode again, regardless of how much that costs and how unlikely that is. Therefore, this failure mode will have a greater cost the next time it occurs. This is why a lot of places have a policy. We don't have the same incident twice. In our favor, though, we may understand the failure mode better, and we may have an easier solution for it than if we'd gone out looking for dragons early in the process. It may be that the odds of this actual, real-life proven failure mode are higher than we thought initially and a solution is cheaper for this than for many theoretical failure modes up front. It doesn't mean that the answer is yes always, though. Odd occurrences still happen no matter what. So if an asteroid wipes out USS-1, it's probably still unwise to buy into an anti-astroid missile defense system. It's just important to know that the game has changed a bit and our decision needs to reflect that going forward, but not looking back. Did we understand this risk up front or was it an emergent thing? Did we just come out on the bad side of a calculated risk? If you think, okay, I'm going to do this and 99% of the time it's gonna go well, 1% of the time we're gonna cause an outage, and everybody's okay with that and an outage happens, well, you knew that there was a chance of that. However, sometimes it's an unknown risk. You could have known about it if you'd done reasonable due diligence and that means diligence, which is likely to be worth it. There's often little point in exploring every possible avenue of failure, but if there are some areas where you have poor visibility or where there's a really complicated system, you can focus maybe some smaller effort there and you can get outsized benefits for the future. Finally, how should our decision making change in the future? I think one of the big takeaways from this talk is that sometimes it just shouldn't. The lesson sometimes that you need to learn is that you can do everything right and still lose. Maybe you had reasonable knowledge about the system and you worked with people that did and you mitigated against clear failure cases. You evaluated the risk of this particular scenario and you felt like this isn't worth preparing for or the odds of it were so slim that even your team of experts, it didn't even occur to them that it might happen. Even so, we can be thoughtful at this time and we can think maybe there's some areas where we can make better decisions in the future. Maybe there's some critical dependency that's too poorly understood or maybe we're taking far too many risks in the aggregate. So if it's your 100th postmortem for the year, maybe you have some problems with technical debt that you need to address. It's also possible you're taking too few risks. If it's your first postmortem in several years and maybe you can tolerate taking a few more chances and spending more time on your product and less on your reliability. It depends, but it's possible, right? So that's how poker can help you see postmortems in a different way. In addition to being great at teaching you how to deal with postmortems, poker teaches us in a painful, visceral way a number of wonderful life lessons. Poker teaches you to deal with tilt. So tilt is when you go out of your mind with rage or frustration because you made a bad play or because your opponent got lucky, something bad happened and then you start making really poor, really expensive decisions. Part of being successful at poker is learning when your mind is not under your own control, getting out from the table and recentering yourself. You learn to know yourself. You learn to pay attention to your feelings, to what upsets you and to accept and deal with those things instead of letting them drive you and drive your decisions. It also teaches you to deal with failure in a constructive way. In poker, every failure is an opportunity. You get to review your play. You can find those mistakes and the next time you get to take advantage of opponents who make that mistake. No matter how good you are, poker will eventually find your flaws and make you learn for them or you'll quit. Poker teaches us to think long term. Poker is an unusual game because you take your score, AKA your money, with you from table to table in one long eternal session, right? Every hand is a mere stitch and a tapestry that you're weaving for your whole poker career, critical only in the aggregate. It doesn't matter, like single hands don't matter in poker. It's the way that you approach hands in the aggregate that matters. So I kept saying earlier, things like giving enough runway and in the long run. So how can you be sure to have enough runway? How can you prepare yourself to survive till the long run? Well, it basically boils down to minimizing catastrophic risks. Catastrophic risks are risks that take you out of the game. Why is this important? Well, even if you only make positive EV bets, if you bet everything every time, eventually you'll hit negative variance. And since you bet everything, it will cost you everything. So you need to construct the game so that if you experience a big slump, you get to keep moving based on your positive expectations. In poker we do this with what's called bankroll management. Your bankroll is the money you've allotted for playing poker and when it runs out you don't get to play anymore. And the greatest poker movie of all time, Rounders, it's about how poker prodigy Matt Damon has the worst bankroll management. He takes his entire $30,000 bankroll to the local underground poker room and he loses it all on a single hand heads up versus the movie's villain, John Malkovich. If you haven't seen this movie yet, you really need to. Anyway, a decent rule of thumb is that you don't be like Matt Damon and you get 20 buy-ins in your bankroll at all times. A buy-in is usually 100 big blinds so that means you need about 2,000 big blinds in your bankroll for the stakes you play. That means you have to go quite poorly to go completely broke in poker. Outside of poker we can mitigate risks in other ways. We buy insurance for our property, for incomes and health. It turns out insurance is negative EV. That's how insurance companies make money but we buy it because it reduces our risk of ruin. Insurance is usually a bet we don't wanna win but we do it anyways because it keeps us in the game. Backups, they're an example of something that's probably negative EV, like insurance. Also like insurance, we can adjust our premium versus our deductible so you can have cheap backups that are sort of hard to restore like tape or you can have expensive backups that are easy to restore like active database replicas. It makes sense to identify the areas which if they failed would be catastrophic and find ways to mitigate against them like business ending events even if it's fairly unlikely but keep in mind it's not necessary that you make this kind of disaster painless. It's unlikely after all but if you can take them from literally ruinous to it's gonna be a bad day if that happens that's a really big win. So I hope today that you got something from this talk you can take back with you to your work, to your life, maybe to a poker table. We talked about poker, we had a brief introduction to the game, some high level strategy to get you started on your journey. We talked about postmortems and how they're way too results oriented. So hopefully you got some new tools now to get more out of your postmortems and you can ask more effective questions. You can analyze the decisions that lead to a result rather than the outcome. And we talked about some ways that poker can make you better at living. If you wanna learn more about poker and you're just getting started here's a few recommendations. This guy Alton Hardin runs an online poker school. He has a lot of Udemy courses. I've taken some, they're really good. He has a popular book on poker math. Dave Sklansky wrote two classic books on poker, Hold'em Poker and the Theory of Poker. And Pokerology offers a free poker course which is how I got started back in the day and I think it's a good place to start. I would be remiss if I didn't mention my wonderful employer, PagerDuty. We are currently hiring so go apply. All the slides from this talk as well as all the resources I mentioned are available at this URL, chamberlin.net slash rubyconf18. You can find me on Twitter at Chamberlin or you can read my blog at chamberlin.net where I mostly write about Lisp and programming and stuff. You can also come up to me after and feel free to say hi or ask questions. Thank you all so much for coming to this talk.