Learning to think probabilistically through poker – with Maria Konnikova

You may remember the episode of Friends where Phoebe starts teaching Joey how to play the guitar – so one of the special skills listed on his resumé is actually true – but refuses to let him so much as touch the instrument. Pretty soon Joey seeks out a new teacher and yet, if the experience of our latest guest on The Value Perspective podcast is anything to go by, he might have been a bit hasty.

Writer and psychologist Maria Konnikova joins us to talk through some of the ideas in her latest book, The Biggest Bluff, in which she charts her journey from poker ingenue to international champion with more than $300,000 in tournament earnings. And the Joey connection? She spent months studying poker textbooks and absorbing advice from her mentor, Poker Hall of Famer Erik Seidel, before she played a single hand of cards.

An aspect of Konnikova’s training that really caught our eye when we read her book, here on The Value Perspective, is probabilistic thinking – not surprisingly, perhaps, given our previous discussions on the subject, including with another very successful poker player, Annie Duke. Did the practice come naturally to her or was it something she had to learn it – and, if so, how?

“No, I was not good at thinking in probabilities,” she replies, laughing at the question. “I am not good at math in general. The last time I took a math class was in high school – many, many years ago – but I did get a PhD in psychology where I studied decision-making. One of the things we covered was how bad people are at thinking probabilistically – and, every single time I saw that, I thought, yup, that’s me.

“When you read the studies or take them yourself, you can train yourself within the context of an experiment to know what the answer is. But then it is such a big leap from knowing how to answer an experimental question in the laboratory correctly to being able to act on probabilistic knowledge in your day-to-day life and in your day-to-day decisions.”

The ‘description-experience gap’

Konnikova freely admits to being guilty of many of the biases she learned about and researched but suspects she is by no means alone in this. “It is common even among people who are statisticians or who study probabilities for a living,” she argues. “Studying these things is not the same thing as living them – in fact, one of the things you learn in psychology is the idea of the ‘description-experience gap’.”

There are two ways of learning, Konnikova explains – one of which is through description. “I could tell you there is a 1% chance of one thing happening, a 10% chance of another thing happening and a 70% chance of something else happening and you take that on board,” she says. “Or else we learn from experience – something happens to me, it happens to someone else I know and I internalise that.

“Overwhelmingly the way our brain learns is through experience – that is what sticks – but, unfortunately, probabilities are not evenly distributed in life. We do not live in a perfect bell-curve distribution and so, if something is supposed to happen 25% of the time, it does not automatically follow that, if it did not happen the first three times, it is going to happen the fourth time.

“That is not how probabilities work – and yet we continue to learn from what happens to us. So we will overestimate the probability of something when we know someone to whom it has happened or when it has happened to us. In contrast, we will understate the probability of something when we do not have any personal experience of it. It is always skewed.”

What Konnikova had not expected when she set out to learn poker, however, was that the game was going to be such a good way of learning to think probabilistically through experience. “Poker is this phenomenal tool because, while you are playing, you are actually sampling – but you are sampling correctly because you are playing thousands and thousands of hands,” she explains.

A feel for probabilities

“So, if you study the game seriously and know what the probabilities are, you begin to learn, oh, this is what 1% feels like and this is what 5% feels like. There is a huge difference between 2% and 3% and, if I can have a 1% edge, then I am printing money. That is what taught me to think probabilistically and to be able to take that knowledge away from the poker table and apply it to the decisions I make in my everyday life.”

As an example, Konnikova points to an experience she had at the very start of the pandemic, exactly this time last year. “Very early on, I realised Covid-19 was probably going to be bad because, even though the numbers at that point seemed very small, one of the things I had taken from poker was small numbers can get bad very quickly,” she says. “There is exponential growth and the virus has ended up showing what this can look like.”

That knowledge led Konnikova to make “a very personal decision”. “Last February, when the pandemic was just starting – it was in the US but we did not really know about it at the time – I was due to fly from a conference in New Orleans straight to Los Angeles to play in one of the biggest poker tournaments of the year,” she remembers. “Then, the night before, the media started covering the first cases arriving in LA.

“It was only a few – a handful – but I actually sat down and looked at the data and decided the last place I needed to be at that time was in a casino. So I cancelled the trip and went back to New York, which went into lockdown, I think, three days after I landed. Prior to poker, I really do not think I would have been able to make that kind of decision.

“You learn by doing – and while I am not saying you have to become a professional poker player, if you actually play a game like that and take it seriously as a way to enhance your ability to think in percentages, it is such a huge edge in life. It makes you much better able to understand risk, to manage risk and to make decisions that are going to profit you – in both a professional and a personal sense – going forward.”