I teach entrepreneurs how to be lucky, and if you dive into the Bayes Theorem — which underpins AI models today and was postulated by Reverend Thomas Bayes in the 1700s — you see how the math of probabilities rewards lucky behaviors.
How is Bayes Theorem commonly applied?
Whenever the term Bayesian is used, it typically implies the idea of updating beliefs in the presence of uncertainty and evidence. Today, Bayes Theorem is used in machine learning, AI and complex decision-making models. These applications are referred to as Bayesian Networks, Bayesian Optimization, and Bayesian Machine Learning.
As I’m interested in modeling luck — and the art of being lucky, these 2 Bayesian concepts apply:
Bayesian Inference
Bayesian inference is a method of statistical inference that updates the probability of a hypothesis as more data or evidence is collected. It contrasts with traditional statistics, which do not account for prior beliefs.
In Bayesian inference we consider prior beliefs and updated beliefs, constantly in motion:
Prior Distribution: Your initial beliefs about a parameter before seeing the data.
Likelihood: How likely the observed data is, given different values of the parameter.
Posterior Distribution: The updated beliefs after considering the data.
Bayesian Reasoning
Bayesian reasoning refers to the process of applying Bayes’ Theorem to update beliefs about the world. It’s often used in cognitive science and philosophy to explain how humans and systems might reason and learn from new evidence.
What does the math say about being lucky?
Bayesian mathematics relates to the concepts of luck and serendipity in how it frames our understanding of probability and updates our beliefs in response to unexpected events.
Here’s how these ideas connect:
1 – Updating Beliefs Based on Unexpected Events
In a Bayesian framework, unexpected events, which could be interpreted as luck or serendipity, lead to updates in our beliefs or models. When something fortunate happens (e.g., a chance meeting that leads to a major opportunity), Bayesian reasoning helps quantify how this event changes our expectations or decisions.
For example:
Prior Belief: You may initially believe that the probability of encountering a career-altering opportunity at a conference is low.
New Evidence: You unexpectedly meet someone who offers a life-changing business connection.
Posterior Belief: After incorporating this new, “lucky” evidence, your belief about the value of attending similar conferences in the future may change, increasing the likelihood you assign to similar serendipitous outcomes.
In Bayesian terms, luck can be seen as low-probability events that, once observed, significantly update the posterior probability.
2 – Luck and Subjective Probability
Luck, in a Bayesian context, is a matter of subjective probability. Bayesian mathematics inherently considers subjective beliefs (prior distributions) that get updated as new events occur. What one person sees as luck may be a highly improbable event according to their prior beliefs, but after it happens, their updated beliefs (posterior distribution) change to better account for such possibilities in the future.
3 – Learning from Serendipitous Events
Bayesian approaches align with the idea that we learn from experience, including serendipitous events. When something unexpected but favorable happens, we adjust our expectations. For instance, if you repeatedly experience lucky breaks in a particular context, Bayesian updating would lead you to expect such breaks more often in similar situations.
By continuously updating our beliefs in response to new and unexpected events, we can better navigate situations where randomness plays a significant role, or luck is available. In this sense, Bayesian thinking doesn’t just accommodate the idea of luck — it embraces luck as a meaningful part of how the world unfolds and how we learn from it.
What does all this mean for being lucky?
When I teach entrepreneurs how to be lucky, I start with actions. Actions drive outcomes. When we get delightful outcomes, this changes our beliefs about what is possible. And that is the key. When you change your Prior Belief in a Bayesian model, you update the probability for a favorable outcome. That is, you get luckier.
(If you find the idea of becoming lucky enticing, take a look at my Quantum Surfing courses, here.)
So, what are some actions that typically result in lucky outcomes?
Curiosity and openness: Explore with child-like wonder and discover insights and opportunities that would otherwise be missed.
Resilience: Look for silver linings and opportunities in the midst of disappointments.
Intuition: Trust yourself and make calls from an internal alignment of what is right for you.
Belief: Simply believe you’re lucky. Seriously.
Want to be lucky?
Experiment with updating your Prior Belief by trying on a few lucky actions today.
