Prediction markets tell you what will happen. They don't tell you what to do.

Prediction markets are having their moment. Election odds, rate decisions, sports, geopolitics. Billions in volume, and a growing belief that the probability is the answer.

It isn’t. A probability is an input. The decision is a separate problem, and it’s the harder one.

Two different skills

Forecasting asks: what’s the chance this happens? Strategy asks: given that chance, what should I do? These feel like the same discipline. They’re not, and mixing them up is expensive.

A prediction market can tell you there’s a 70% chance a rate cut lands in March. It can’t tell you whether your company should refinance now or wait, because that depends on your balance sheet, your risk tolerance, your alternatives, and what it costs you to be wrong in each direction. The market prices the event. It doesn’t price your exposure to it.

Poker players learn this distinction on day one, because the game forces it. Knowing you’ll win a hand 70% of the time settles nothing by itself. Whether to bet, how much, whether to fold anyway: that depends on stack sizes, position, what your opponent does with the rest of their range, and what happens to your tournament life if the 30% shows up. Every serious player has folded hands they knew were probably winning. The probability said one thing. The decision said another. Both were correct.

Why the confusion matters

The failure mode I keep seeing, in crypto and in business, is people treating a forecast as a strategy.

They read 70% and act as if it’s 100%, sizing their whole position on the modal outcome. Then the 30% arrives, as it does three times in ten, and they’re wiped out of a game they were playing well. A poker player would call this playing without bankroll management. It’s not a prediction error. The forecast was fine. The decision layered on top of it was reckless.

The reverse failure exists too: people who forecast beautifully and never act, because no probability ever feels safe enough. Analysts, not operators. Poker has a word for them as well. They’re the players who fold their way to the bubble and out.

What deciding well actually takes

Sitting between the forecast and the action is a stack of skills that prediction markets don’t provide:

• Sizing. How much of your bankroll does this decision get, given that you’ll be wrong often?
• Asymmetry. What does each way of being wrong cost? A 70% bet with ruinous downside is worse than a 40% bet with a capped one.
• Range thinking. Not “what will happen” but “what’s the full distribution of what could happen, and am I okay across all of it?”
• Reversibility. Can I change course when new information lands, and what does keeping that option cost me?

Poker trains every one of these, thousands of hands at a time, with real money enforcing honesty. That’s why I keep calling it the purest training ground for decisions under uncertainty. Not because the game matters, but because it’s the only place where forecasting and deciding are drilled together, at volume, with immediate feedback.

Prediction markets are genuinely useful. I want them to exist, and I use them. They’re the best probability engine we’ve built. But a probability engine without a decision framework is a speedometer without a steering wheel.

The markets will keep telling us what’s likely to happen. The edge belongs to the people who know what to do about it.