Dylan Evans: Are Some People Just Better At Taking Risks?

Nov 8, 2019
Originally published on November 8, 2019 11:51 am

Part 2 of the TED Radio Hour episode Risk.

About Dylan Evan's TED Talk

Dylan Evans spent years studying pro gamblers and found that the savviest risk-takers are better at calculating probability. He says we can make better decisions by learning from expert gamblers.

About Dylan Evans

Dylan Evans is a risk intelligence expert who serves as the CEO and the founder of Projection Point, a company that creates risk intelligence training programs for clients.

He has written several books, including Risk Intelligence: How to Live with Uncertainty and Emotion: The Science of Sentiment.

Evans received a PhD in philosophy from the London School of Economics and has held academic appointments at King's College London, the University of Bath, the University of the West of England, and University College Cork.

Copyright 2020 NPR. To see more, visit https://www.npr.org.


It's the TED Radio Hour from NPR. I'm Guy Raz. And on the show today, ideas about risk.

I tend to be risk-averse. Is that just a fixed, like, thing - that some people are and some people aren't?

DYLAN EVANS: So I think whether you're risk-loving or risk-averse - that seems to be fairly hard-wired in the sense that it's - there's a heritable element. Whereas risk intelligence is all about how good people are at estimating probabilities.

RAZ: This is Dylan Evans. He's a behavioral scientist, and he studies why some people are better when it comes to taking risks.

EVANS: Someone who's risk intelligent is capable of gathering a wide range of information - not approaching the information with fixed theories but willing to constantly update and modify their beliefs. This is one thing, interestingly, about professional gamblers - is that they make quite good decisions.

RAZ: Dylan Evans picks up his idea from the TED stage.


EVANS: I've spent the last few years interviewing some of the most successful gamblers in the world, and it's not just because I like spending my time at casinos. I've got a hunch that there's a special kind of intelligence for thinking about risk and uncertainty.

So if psychologists study gamblers at all, they tend to focus exclusively on problem gamblers. But that's like studying obesity and neglecting haute cuisine or studying alcoholism and neglecting fine wines. Not all gamblers are fools. Some of them are experts. And I think we can all learn a lot about how to make better decisions by studying the way these expert gamblers think.

RAZ: So what is the - in your view, like, the biggest misconception that most of us have about gamblers? Because I think most of us just assume that gambling is just about chance and luck.

EVANS: Yes. Well, a lot of gambling is just about luck.


EVANS: But I think the professional gamblers that I've talked to - it's not the individual wins that thrill them. It's a much more cognitive pleasure of figuring out the rules of this particular challenge, gathering the data, processing it almost like a scientist would. And there are still a few games where there is still room for skill - blackjack, poker, sports betting. And by sports betting, I mean horse racing, betting on football or soccer any of the kind of traditional racing games.

But I think the area where we see risk intelligence at its most developed is the sports betting because there, we're talking about a system which is much more open-ended. We're talking about real people or animals, which is much more complicated than the rules of a game in a casino.


EVANS: And what got me interested in this group was a fascinating paper I read some time ago by two American psychologists called Steven Searcy (ph) and Jeffrey Liker (ph). Back in the 1980s, fresh out of grad school, they visited a horse racing venue in Delaware, and they interviewed 30 avid racetrack patrons.

Men - they were all men - they went to the racetrack every day of their working lives. They didn't make their money from gambling. It was just fun. But they were absolutely obsessed with it. And 14 of them were just incredibly good at forecasting the odds, the final odds that would be offered on the horses just prior to the race, which is a pretty good guide to the probability of the horse actually winning.

So over the course of four years, Searcy and Liker figured out that the best way to model what the expert gamblers were doing when they were thinking about predicting the horse's chance of winning the race was to model it as a complex linear equation consisting of seven variables, each of which related to the horse's performance in previous races, like its average speed, its average finishing position and so on. But, of course, these expert gamblers - they didn't know anything about linear regression. They'd never heard of statistical modelling. They were doing it all unconsciously.

RAZ: So take me to a racetrack, right? You have a pretty good ability to assess risk, and you've got a hundred dollars to spend. What are you going to do?

EVANS: Well, first of all, I would have done my homework before I even set foot into the racetrack. I would have studied the form of the horses...


EVANS: ...How many races they'd won, the races that they'd came second and third in. Also, the horse's preference for different kinds of ground - hard ground or soft ground. You're looking at the jockey, the accumulated winnings of a particular horse. On the basis of that, you want to come out with some initial idea of what the horse's chances are of winning a race. But that's not enough because it's not necessarily looking for the horse that's going to win the race. You're looking for the horse that the other players have underestimated.


UNIDENTIFIED PERSON #1: And Sir Winston has won the Belmont Stakes.

RAZ: Why do you think that risk intelligence - you know, it's easy for some people, and some people just ignore it or don't even consider it?

EVANS: Well, above all, it's about being fiercely honest with yourself. The really amazing risk-takers - they understand their own weaknesses. They keep very detailed records of their own performance, and they look at where they've got things wrong. And they learn from those mistakes...

RAZ: Wow.

EVANS: ...Not just when they're assessing their own performance but when they're dealing with other people as well.

RAZ: So is there a way to kind of increase our confidence when we're taking risks? I mean, is it about - is it as simple as gathering information?

EVANS: It's about gathering as wide a range of information as we can. That's why people have - when they get good at risk intelligence, it tends to be in a very particular domain. So you might be very good at assessing political events, but you might be very bad at medical risks because you need to spend a long time - years, perhaps decades - immersing yourself in a certain domain in order to gather the domain expertise and build this model inside your head.

RAZ: Do you think that people take risks based on how much they're willing to lose?

EVANS: So one of the key elements that you're thinking about when you make a decision about risk is your possible gains and the likelihood of those gains. You have to take into account your possible losses and the likelihood of those losses. Someone who's risk-intelligent weighs those things out equally. They don't have a particular aversion to loss - but with the important proviso that they never risk so much that they can't bet again because then you're really out of the game.


EVANS: So I want to finish off by just describing simple ways in which you can use probabilities when making your decisions. The first way is simply to set a threshold. So, for example, you might say, well, I'm only going to ask this person out on a date if I think that there's a 60% chance that they will say yes. And the great thing about this method is, of course, that everybody can set their thresholds differently, at different levels depending on how desperate you are to go on a date...


EVANS: ...And how much you hate being rejected. The second is betting an amount that is proportional to your confidence. So, for example, you could decide to buy more or less shares in various companies depending on how strongly you believe that each company was going to do well. Overconfidence is going to be fatal here.

Now, of course, luck can always go against you. There's no guarantee that even if you use these decision-making methods that you'll always win. But if that thought upsets you, make your peace with chance. Just accept that you can't control her. Accept that she doesn't care about you, and she never will. But don't despair. That's the way to bet. Thank you.


RAZ: That's Dylan Evans. He's a behavioral psychologist and the author of the book "Risk Intelligence." You can see his entire talk at ted.npr.org.

By the way, are you a gambler at all?

EVANS: I have tried gambling in casinos when I was doing the research for my book. But I think I could really do with some risk training myself. Transcript provided by NPR, Copyright NPR.