When I was a PhD student in particle physics, my world was filled with uncertainty. Not the everyday kind — the kind you shrug off when you miss a bus or wait for a delayed email. This was deeper: entire experiments, entire careers, were built on the chance that something invisible might leave the faintest trace in a detector.
We didn’t see the Higgs boson directly. We didn’t see dark matter at all. What we saw were hints, distortions, the shadows of something hidden. And our task was to weigh those shadows — to ask, not “what’s the answer?” but “how likely is this to be true?”
That was Bayesian reasoning. And once you start thinking that way, you never really stop.
Bayesian physics wasn’t about certainty. It was about living in likelihoods. You imagine possible worlds. You let the data speak. And then you update your beliefs. Not once, but again and again, with each new clue.
It’s a strange kind of humility. You stop thinking in absolutes, and start thinking in probabilities. You stop asking for “proof” and instead ask for “weight of evidence.”
And yet — that humility was also power. It kept us honest. It kept us resilient. It kept us from being seduced by coincidence.
Years later, when I began working in finance, I felt that same atmosphere of uncertainty. Data everywhere. Metrics. Indices. ESG scores. Analysts chasing patterns.
But something was missing. Too often, the conversation stopped at correlation. If two things moved together, we acted as if one explained the other.
In physics, we knew better. A bump in a histogram might look like discovery, but until you asked what hidden parameters could have generated it, you had nothing. Correlation without causality is just noise with good marketing.
Bayesian thinking was my first lesson in causality, even if I didn’t call it that at the time.
Because causality isn’t magic. It isn’t a separate discipline. It’s the natural extension of the Bayesian worldview. First you accept uncertainty. Then you ask what hidden structure might explain it. And finally, you test whether that structure holds up as new evidence comes in.
It’s the same game, whether you’re chasing invisible particles or hidden financial drivers.
I’ve come to see Bayesian thinking not just as a method, but as a way of life.
We all live with incomplete information. About our work. About our relationships. About our future. The temptation is to demand certainty — to grab onto correlations, to believe that this event must explain that outcome.
But life, like physics, doesn’t offer us that comfort. What it offers is likelihoods. Weights. Signals hidden in the noise.
And if we can learn to live inside that uncertainty — to embrace it, to update our beliefs gracefully — then we can see further than we imagined.
So today, when I look at finance, I don’t see rows of numbers. I see shadows. Hints of hidden drivers shaping visible outcomes. I see the fingerprints of causality.
And I remember the lesson from physics: the most important forces are often invisible. But if you learn to ask the right questions, if you learn to weigh likelihoods instead of chasing certainties, you can make the invisible visible.
That’s the quiet gift Bayesian physics left me. Not just a set of equations. But a mindset. A way to carry uncertainty with humility. A way to search for causes, not just patterns.
And that mindset, more than any specific tool, is what guides me now.
