How to Read Implied Probability Like a Sharp

Here's the most important thing to understand about prediction markets: the price is the probability.

When you see a YES contract trading at 45¢, the market is saying there's a 45% chance this event happens. That's the implied probability. Everything else in your analysis flows from there.

From Price to Probability

On Kalshi and Polymarket, prices range from 0¢ to 100¢. The math is direct:

42¢ YES = 42% implied probability

If you buy YES at 42¢ and you're right, you collect $1.00. Your net profit is 58¢. If you're wrong, you lose your 42¢ stake. The break-even point — where you're indifferent about making the trade — is exactly 42%.

The Vig Problem

Here's where it gets tricky. The YES price and the NO price don't add up to exactly 100¢. They typically add up to 96¢–98¢ or so. The remaining 2–4¢ is the spread, which benefits whoever is providing liquidity.

If YES is at 42¢ and NO is at 56¢, those add up to 98¢. You need YES to be more likely than 42% just to break even. This is the vig — the house's cut baked into the structure of the market.

The existence of vig means that for a bet to be profitable, your true probability must exceed the implied probability by more than zero. You need positive edge.

Finding Your True Probability

This is the real work. The implied probability is just what the crowd thinks. Your job is to estimate the actual probability better than the crowd.

Tools that help: base rates (how often does this type of event resolve YES historically?), current news and context (what's changed that the market hasn't priced in?), and models (can you build a simple quantitative estimate?).

FadeMe's AI does this for every pick — it estimates the true probability independently of the market price and shows you the gap. That gap is the edge.

What Positive Edge Looks Like

If a market prices YES at 42% and you estimate the true probability at 58%, you have a +16 percentage point edge. That's a strong play.

If a market prices YES at 42% and you estimate true probability at 44%, you have a tiny edge — but the vig may eat it entirely. Not a play worth making.

If a market prices YES at 42% and you estimate true probability at 35%, the market is actually overpriced. The smart move is to fade it — buy NO instead.

The Mental Model

Stop thinking of prediction market prices as just numbers. Every price is someone's embedded prediction, encoded in cents. Your job is to decide whether you agree with that prediction or not — and how confident you are in your disagreement.

Get good at translating prices to probabilities and you'll see mispricings the market misses. That's where the edge lives.