What many people get wrong about Probability & Risk

Probability — the simple concept that many people get wrong, time and time again.

Random Chance Outcomes

If a coin tossed, there are only two outcomes: Heads (H) or Tails (T). The probability of a coin landing on either H or T, when flipped, is exactly 50%, every time. And that doesn’t change.

If there is a streak of seven tosses in a row, and they all land on heads — (H–H–H–H–H–H–H) — the probability of the next toss landing on Tails is not any more likely than it is Heads. It’s still 50%.

That’s where people get things wrong. The probability in games of Random Chance — those where you have no influence on the outcome — never changes.

Skill-Based Outcomes

In games or scenarios where probability changes with every piece of known information, there are a number of factors which contribute to the likelihood of an outcome.

Let’s look at the Hot Hand debate, for instance. Does the Hot Hand exist in sports, in startups, in investing? The answer is, “sort of”. However, it’s deeper than that, since it’s all contingent on Bayes’ Theorem.

What Thomas Bayes proposed, is that the probability of an outcome changes with the prior information you have about the potential outcome.

Let’s use Steph Curry, as an example. In the 2019-2020 Season, Curry’s overall shooting percentage (FG%) was 40.2%. However, that is a blend of his FG% at Home (36.2%), and on the Road (45.7%).

What we don’t know (without digging deeper) is why that difference of 9.5% exists. Correlation does not equal causation, but here are some things that we can guess, which might contribute to the difference in his overall outcome.

Considerations about Steph Curry’s performance

These are just a few factors to consider, and we’ll only know the answers to them by digging deeper into Curry’s routine at home versus on the road. Regardless though, Curry’s performance each game is not based on Random Chance. The probability of his performance will change from game to game, based on information that we know, prior to the game taking place.

Probability & Risk

So what does probability have to do with taking risks?

Betting on games of Random Chance, such as Roulette, takes exactly zero skill. Betting on outcomes where you have asymmetrical information takes skill. And it takes a lot of knowledge and understanding to do it well.

Don’t bet on Random Chance

A person playing Roulette has no chance of controlling the outcome. If the wheel lands on Red, the chance of it being Red or Black with the next spin is exactly 50%.

Don’t take risks and bet on outcomes that are based on Random Chance, unless you understand that you are doing so strictly for fun.

Bet on Bayes

If you had the chance to bet on Steph Curry’s FG% for a game, and you knew that he only slept 6 hours, instead of his regular 8 hours, and he didn’t eat breakfast that morning, the probability of his FG% might drop drastically from 36.2%, assuming it’s a home game.

This is where Bayes comes in. With this prior information, we know that the probability of Curry’s outcome will likely change, so it’s safe to take a risk and bet that he’ll have a lower FG% that game.

How does this apply to startups?

The probability of a startup succeeding isn’t based on Random Chance. We know that a startup is more likely to succeed with certain, prior information. Startups are complex and can fail based on a number of factors — poor leadership, bad product, lack of capital, and the list goes on.

However, startups are skill-based, in the sense that the probability of success is higher if you do the necessary things to put yourself in a position to succeed. Like Curry’s shooting percentage though, startups take a lot of grit and failure, in order to succeed.

For every shot that Curry makes at home, just remember that he misses 63.8% of the time, and keeps on shooting. It takes patience and perseverance to succeed — and Steph Curry is world-class in his domain.

Head of Growth: Levels / Startup Team: SkipTheDishes / Co-founder: Thisten, Top & Derby / Host: Character Podcast / Rotman MBA