With a standard deck of cards, there are around 2.5 million potential hands at the start of each game of poker. The key to success is in the decision process—determining how to play a hand, and how much to bet, despite not knowing which cards are in another player’s hand. Like in poker, every investment decision includes at least some degree of uncertainty. What lessons can poker teach investors about sizing their bets, managing risk, and making better decisions for their portfolio? 

This episode of The Outthinking Investor delivers insights on the benefits of probabilistic thinking, the best methods for analyzing portfolio decisions, and how investors could develop a sound decision-making process for targeting returns - even when facing uncertainty and market volatility.  

Our guests are Annie Duke, a decision scientist, former world-class professional poker player, and the author of books including “Thinking in Bets” and “Quit”; Tina Lindstrom, Head of Oil Derivatives Trading for North America at Marex; and Adam Papallo, Head of Implementation Research at PGIM Quantitative Solutions. 

Do you have any comments, suggestions, or topics you would like us to cover? Email us at thought.leadership@pgim.com, or fill out our survey at PGIM.com/podcast/outthinking-investor. 

>> The world's most successful poker players are keen observers. They can spot an opponent's tell in the blink of an eye, yet they have self-awareness to check their own emotions. They can adapt their playing style to take advantage of a table. Most of all they understand the odds of winning for each hand they play. That's because for the standard deck of 52 playing cards, there are around 2.5 million potential hands at the start of each game. So while the ability to bluff or read another player's bluff is useful, the real skill is in the decision process. To be successful, players need to make good decisions on how to play each hand and how much to bet. And they need to do it fast, over and over again. The most successful players are also able to synthesize information over time, improving their process of decision making. All of this happens amid great uncertainty. Like in poker, investment decisions are made without complete information. We can't control the markets but we can improve the way we evaluate investment opportunities, manage risk, and make decisions that increase the odds of a winning investment. What can the classic game of poker teach us about investment decision making? And how can investors use these principles to improve their own risk management and decision making?

[ Music ]

To understand today's investment landscape, it's important to know how we got here. This is The Outthinking Investor, a podcast from PGIM that examines the past, the present-day opportunities, and the future possibilities across global capital markets. Annie Duke is a decision scientist and former world-class professional poker player. Her recent books include Thinking in Bets: Making Smarter Decisions When You Don't Have All the Facts and Quit: The Power of Knowing When to Walk Away. Tina Lindstrom is head of oil derivatives trading for North America at Marex. She began her career as an intern and assistant trader at Susquehanna International Group. And Adam Papallo is head of implementation research at PGIM Quantitative Solutions. Poker and investing both rely on probabilities, risk management, and decision analytics. For Annie Duke, the similarities are obvious. Poker started out as a way to make money during her time in the PhD program in cognitive science at the University of Pennsylvania. But it's the problem-solving aspect of poker that she fell in love with.

>> Here's the thing that I think is really interesting about decision making under uncertainty. There's kind of two issues. One is how do you actually make a decision when you're not 100% sure that it's right or that it's going to work out well? And I think this is really hard. It's what I call the I'm not omniscient and I don't have a time machine problem, which are the two issues that really plague our decision making. I'm not omniscient means that really almost any decision that we've ever made in our life has been made under conditions of uncertainty, meaning I just don't have all the facts. Some of the beliefs that I have are inaccurate and I just don't know it. There's a whole bunch of stuff that I don't know or I've never encountered. And then there's the second form of uncertainty which is the time machine problem, which is that there's just plain luck. I can make a great decision and have a bad result. So I can kind of think about poker and the fact that you can't see the other players' cards and you don't know what the next card that the dealer is going to deal is. So I can be in a situation where I could be 80% to win a hand against you. And by definition, I'm going to observe a bad result 20% of the time and I don't have any control over when that 20% is going to occur. And I think that that makes the starting really hard for us because we're trying to protect against those bad outcomes that we don't really have control over. What we have control over is making decisions that are going to reduce the chances that we're going to observe a bad outcome. But you really can't bring a bad outcome down to zero but we sort of try. So that makes starting hard.

>> For investors, this is the ex ante or forward-looking analysis before an investment decision is made. It is no coincidence that the Latin word, ante, also refers to the bet players make before a hand is dealt. But uncertainty doesn't end here. There's also uncertainty in the ex post analysis, which is backward-looking or after the fact.

>> On the back end, I think there's two really big problems that come from this situation that we all find ourselves in, which is I don't know much and I can't control a lot of things about the way that things turn out. And the first is, under those conditions, how do you close feedback loops well? If I have a good outcome, how do I figure out why? Was it because I made a great choice or was it because I got good luck? So you can see that so clearly in poker, right? Because, you know, in a situation where I'm 80% to win and you're 20% to win and you win, that doesn't mean that your choice was good. And very often, we kind of don't find out why because I don't get to see your cards in the end. A lot of poker hands don't end with the cards actually being revealed. And even so, there's just so much volatility in the system. And we can see that this is a huge problem for investors. You make an investment and you lose on it. Why did you lose on it? Was your thesis wrong? Did you make a bad choice? Was your bet sizing wrong? Or was it just that the world happened? The second problem for the back end is that when we do make these decisions under these conditions of uncertainty, what that means is that definitionally after we've made a choice, after we, for example, made an investment, we're going to find out new information after the fact. So sometimes it may not be that you learn new facts that you wish you kind of had known before. So as we learn those new facts, we now get into this new problem which is do I want to stop the thing that I started? But that is also the decision that's made under the same types of uncertainty. And we have the exact same difficulties with that decision.

>> Trying to untangle that mess to better manage risk and improve decision making is the crux of the investment dilemma. So where does that leave investors who will always need to make decisions in the face of uncertainty?

>> The more uncertainty that there is, the more I think that people are willing to say things like, "Well, this is sort of a different world and you just sort of have to go with your gut," because, you know, I can't know anything until 10 years from now. And also there's power law, maybe if you're in early stage investing. And so what they go with is that idea of that sort of special nose for value. And one of the things that I always say about that is, OK, well, imagine that you're talking to an options trader and you say to them, "Well, what's your investment process?" And their answer is, "I just have a really good nose for value so I go with my gut." Would you invest in them? And of course everybody says, "No, I would never do that," right? And I think that the tightness of those feedback loops when you're trading options makes it so that you can't get away with that rationalization. Everybody recognizes that you have to have quant work behind that because we're going to find out pretty quickly what's going on.

>> Tina Lindstrom knows this to be true. She's an energy options trader and started her career at Susquehanna, one of the first investment firms to incorporate poker into their internship program to help new traders develop decision-making skills.

>> That taught me a lot because the concept of winner's curse doesn't come up that often in modern trading. People aren't really talking about winner's curse, how people bet, the information that you're gleaning from, the steps that market participants are doing that you're not picking up on. Susquehanna's poker teachings and arbitrage, mathematical, positive expectancy bets have taught me as a trader now to think probabilistically and also to constantly question my thoughts, constantly reassess, given what other opponents in the market are doing. A lot of people, when they put on a trade, they give up too much in execution costs and that can be a very, very big drag on returns because if you give up a lot of edge, negative slippage going into the trade, going out, and you repeat this over time, you're giving up a lot of your returns. You could be giving up 10%, 20% of your returns through bad execution. Market makers see you coming. Market makers can see the pattern, the times that you put an order through. They can see imbalances where everybody else is putting on the same trade, if you're putting a trade on market on open, market on close, certain patterns.

>> Traders can adopt different tactics to increase returns, depending on the market environment. Trading options is more like solving a puzzle, always being on the lookout for those small mispricings.

>> But options, they expire at some point. It's either in the money or it's out of the money. Either you made money on it or you didn't make money on it. So that part is super interesting. And even equity derivatives versus commodity derivatives, that's all completely different too because for example, if you're trading agricultural derivatives they are off of different months, different crops. The fear, the skew is completely different. The fear is to the upside. Something scary that happens with commodities is when there's not enough. When you cut off my access to oil, if let's say the Strait of Hormuz is closed. So usually the fear is to the upside, unless, of course, it's during COVID when the fear was to the downside because the WTI delivers to Cushing, Oklahoma, landlocked tanks. Then we were at the risk of tank tops which is what caused the negative oil prices during COVID.

>> Good data is key to a good decision. But most decisions also have some qualitative component. And research shows that most people are overconfident in their ability to make good decisions.

>> People have a very bad idea of whether their gut is good or not. And part of it is because in these situations where there's more volatility, it's easier to attribute good things to good gut decision making and bad things to I just got unlucky. It's actually the uncertainty that allows that to happen. Because if I add more certainty into the equation, this idea that your gut is fine and you don't need any process and this problem where you sort of have these open feedback loops that allow all this bias to creep in kind of goes away.

>> It's not just the decision of whether to invest but also how much. And that hinges on the skill that an investor brings to the decision or what Annie Duke refers to as expectancy or expected value.

>> How do you figure out what your expectancy is? This is the question that keeps me up at night. Because risk and bet sizing can be handled by different approaches. I can go do a calculation that's going to allow me to know how much should I bet on any given time. The problem is that that calculation depends on an estimate of my expected value. In most of the types of decisions that we're making, we don't actually know what our expected value is. And it's very easy to fool ourselves into thinking that it's higher than it actually is. Very famously, we're very bad at, for example, assessing tail risk. Really understanding how often the tails are going to come in, just things like that. But mostly, if I'm in a system that's pretty volatile, so there's a lot of variance, a lot of luck in the outcome, and there's lots of hidden information that I don't know and I'm trying to assess what my skill level is which is going to get you to what your edge is, I think that we just make a lot of mistakes with edge.

>> Part of the challenge is in differentiating information from noise when the variables are virtually endless and the environment is constantly changing.

>> Every trader has their information sources, their news sources, and network to talk to you about participants, headlines, et cetera, right? And often there is noise, especially for the correlation trader or the relative value trader. Sometimes Asset A is a good indicator for what's going to happen in Asset B, unless something very specific is happening to Asset A that does not apply to Asset B. And there is a lot of noise in the market. But I think, one, you should always strive to be an objective decision maker. It's very hard to check yourself but I think that part is important. The second part is you're constantly looking at your independent variables and your dependent variables. And so you have to tweak it. I think it's an art, it's not a science, weeding out the noise, but very active attention can mitigate some of those risks.

>> One especially noteworthy trade Tina made was in January of 2020. It was Martin Luther King Jr. Day and she was online watching energy prices. At first, heating oil was outperforming crude oil then heating oil took a sudden U-turn and sold off dramatically. It continued to underperform without any clear reason. Then information came from an unlikely source. Tina's mother had been reading Chinese newspapers and learned that large numbers of people in China were dying. The world was still unaware of a looming pandemic.

>> It occurred to me that this was a big, big, big deal. This was COVID. But the Chinese stock market was closed for Chinese New Year. And so people kind of heard about it in the West and in Europe but didn't really pay attention to it. And the more I read about this, I was more alarmed than ever. And I remember doing a bunch of ratio put spreads. It seemed like there was a possibility of this really bad thing happening that you could hedge by buying put options in oil, but yet the prices were not reflective of that at all. So I was able to put on all that I wanted to without really paying for that much at all. I wasn't aggressive. I was really pretty patient. And then I ended up to be right a few weeks later and certainly benefited when oil prices went negative. A lot of people didn't change their models over until it got too dicey and they just shut off their models for the rest of the year. And even though I was a prop trader, I became a market maker for the rest of the year because the market makers shut down their trading. So, more systematic trading, more reliance on electronic markets, it didn't work in this instance because there was nobody looking at the models and saying, hey, these deltas don't make sense for each of these options. There was nobody to question it. It was sort of just from a black box which then by the time they realized was too late and they were so scared, they were not able to readjust.

[ Music ]

>> The COVID-19 pandemic was an extreme macro risk that affected every securities market in the world. That's clear in hindsight. As Adam Papallo knows, it's difficult to identify macro and micro risks and how much they're priced into investments.

>> The key point here, though, should be that they're not always easily separated, macro and micro risks. A micro risk is specific to a company, yes, but a macro risk could come into the market and impact all of these securities but in different ways and in different magnitudes. And I think that's worth thinking about very carefully.

>> A current case in point is the uncertainty of how government policy in the US could change with the new administration and how that could impact global markets.

>> Some of those changes will be policy changes and that will impact potential interest rates, inflation, or even smaller segments of the economy but still broad enough, like electric vehicles. You have these policy changes on whether or not you're going to get a tax credit for your electric vehicle. Well, that impacts all manufacturers of electric vehicles. However, some may be impacted more or less favorably. For instance, if you're a well-run company and your product is in demand, the fact that there's a tax credit may not impact you very much. However, if your product is and is in demand or it's more expensive and you're really getting a benefit from this tax credit, that going away could certainly hurt you much more than another vehicle manufacturer. So while that is a policy risk and we would consider that more of a macro risk, it's impacting these different companies to varying degrees. There's different sensitivity to these. When we talk about measuring how a macro risk will impact your portfolio, we think about, well, how sensitive are certain assets to these types of risks? I'm always more concerned about macro risks because they're harder to predict.

>> The nature of the news cycle also makes it more challenging to maintain objectivity amid uncertainty.

>> We have this bias to focus on what is being presented. So, we get information in the media or information coming to us. We're constantly inundated with information about macro risks such as inflation and how that impacts interest rates, how that impacts consumption or GDP, therefore employment, and how that then trickles down to the mom-and-pop investor. But the institutional investor is taking this all in, attempting to understand, well, these assets respond in this way to changes in inflation. So, how do I best position my portfolio so that I'm not overexposed to, say, inflation risks? How do I deal with that so that I can meet my needs to either funding my liabilities or matching the return that's expected of my portfolio? However, very few companies have existed for over 100 years. I happen to work for one that existed for 150 years but it's a pretty rare event. You don't have very many companies existing centuries. So, knowing what's actually going to happen to that individual company or asset is very difficult to forecast, also in part because there's a lot of specifics on an individual company.

>> At the end of the day, the goal for most investors is to identify skillful managers who are able to provide positive risk-adjusted returns amid market uncertainty. That requires a deeper understanding of the manager's source of returns. Is it mostly luck or is it skill? Luck isn't very reliable and it probably won't lead to long-term consistency.

>> If a manager has phenomenal returns in aggregate, the first thing to do is to dissect, well, how consistent are those returns and are those returns attributable to a measurable philosophy? If you buy a value manager, their value's going to be out of favor for a while potentially and it's going to come back, right? But you can understand that. If a manager came in and they correctly forecasted the response of the market to the COVID pandemic, that's a one-time event. Why were they right? You have to understand their philosophy going into it. Well, did they see the news from China? Did they get data and that fed into their process that allowed them to allocate in a certain way? If so, then that may be a manager worth investing in. If they're just pessimistic and they got that right, that's hardly an investment philosophy that one would want to invest in. Process is repeatable and luck due to getting one event is not. I mean, anyone can go back and write a narrative as to why they did well but understanding that their process has been consistent throughout the time period is what will really be key to picking out managers that are skilled over luck.

>> Having a rigorous investment decision process is key to deciding when it's time to hold a position and when it's time to exit. If it also works to shorten or close feedback loops then it can help make us better investors over time, even amid great uncertainty. Thanks to Annie Duke, Tina Lindstrom, and Adam Papallo for their insights on managing risk and making decisions amid uncertainty. The Outthinking Investor is a podcast from PGIM. Follow, subscribe, and if you like what you hear, go ahead and give us a review. If you enjoyed this episode and want to hear more from PGIM, tune in to our Speaking of Alternatives podcast. See the link in the show notes for more information.