How Prediction Market Odds Work

In a prediction market, the price of a contract IS the probability. A contract trading at 65¢ is the market's collective estimate that the underlying event has a 65% chance of occurring.

This is fundamentally different from traditional betting odds. A bookmaker might offer you 2/1 (decimal 3.0) on an event, which implies a 33% probability — but the bookmaker also takes a margin, so the true market price might be 40%. Prediction markets strip out that margin (almost entirely in the most liquid markets), giving you a cleaner probability signal.

The formula is simple: implied probability = price in ¢ / 100. A 72¢ YES = 72% implied probability. A 28¢ NO = 28% implied probability. YES + NO should sum to 100¢ (they sum to slightly less than 100¢ due to transaction fees).

Expected value (EV) is the average outcome of a bet if it were repeated infinitely. A positive EV bet is profitable in the long run; a negative EV bet loses money in the long run.

Formula: EV = (probability of winning × profit per win) - (probability of losing × loss per loss)

This is a positive EV bet: for every $1 you put in, you expect to make 10¢ in profit on average. The market is pricing the candidate at 60% when you believe the true probability is 70% — a 10-point mispricing.

The challenge: your probability estimate must be more accurate than the market's. If you're wrong about the 70% and the true probability is actually 55%, the bet is negative EV.

Prediction markets are highly efficient in some areas and systematically inefficient in others.

Every time you trade, you pay a cost in the form of the bid-ask spread. On a liquid market, this might be 1-2¢. On an illiquid market, it could be 5-10¢ or more.

If you're buying YES at 65¢ and the mid-price (midpoint of bid and ask) is 63¢, you're already starting 2¢ below fair value. To break even, the true probability needs to be at least 65%, not just 63%.

For this reason, small-edge trades are often not worth making in illiquid markets. Your edge needs to exceed the transaction cost. In practice, target at least a 5-point edge (you believe true probability is 70%+, market prices at 65% or below) in illiquid markets, and 3-point edge minimum in highly liquid markets.

A calibrated forecaster means: when you say something is 70% likely, it happens about 70% of the time. Most people are systematically miscalibrated — usually overconfident (70% confident estimates that turn out right only 55% of the time).

You can measure your calibration by tracking your prediction market outcomes over time. If you buy contracts at 70¢ and they resolve YES less than 70% of the time, you're overconfident in your edge. If they resolve YES more than 70% of the time, you're either underestimating your ability or cherry-picking high-conviction trades.

PaperPoly's analytics dashboard shows your Brier score — a calibration metric that measures how well your entry prices predicted actual outcomes. A lower Brier score is better. Improving calibration is the highest-leverage skill improvement available to any prediction market trader.

Before every trade, write down three things: your probability estimate, the market's implied probability, and your reasoning for the difference.

If you can't articulate why your estimate differs from the market — if you're just going with a gut feeling — don't trade. The market has aggregated a lot of information. Without a specific reason why you're right and the market is wrong, you have no edge.

Start with markets in domains where you have genuine expertise. A doctor has an edge on healthcare regulatory markets. A securities lawyer has an edge on SEC enforcement markets. An avid sports bettor has an edge on their sport. Use the knowledge advantage you already have.