Finding Value: Probability vs. Price

Value, defined

Every price carries a hidden implied probability — the win rate a bet needs just to break even (you met this in Course 01). A bet has value when one number beats another:

• Your estimated true probability — how often you believe the outcome actually happens.
• The price’s implied probability — how often the price says it needs to happen.

If your number is higher than the price’s number, you’re being offered a better payout than the risk deserves. That gap — and only that gap — is the edge. Notice what’s not in this definition: whether the team is good, whether you like them, or whether they’re likely to win. A heavy favorite can still be a value bet, and a long shot can still be a trap. It depends entirely on the price.

Compare against the fair price, not the raw one

There’s a catch that fools beginners constantly. The implied probability you read straight off a price is inflated by the vig — the book’s built-in margin (Course 05). If you compare your estimate against that inflated number, you’ll think you have no edge when you do, or — worse — convince yourself a fair bet is a good one.

The fix is to strip the vig out first and compare against the no-vig fair price. Take both sides of a market, add their raw implied probabilities, and divide each by that total so they sum to a clean 100%.

Example: a market priced −110 / −110. Each side implies 110 ÷ 210 = 52.38%, and together they sum to 104.76%. The fair probability of each side is 52.38 ÷ 104.76 = 50% — exactly what you’d expect from a true coin flip. The extra 4.76% was never real; it was the house margin. Always measure your edge against the fair number, not the raw one.

Expected value: the math of a good bet

“Value” needs to be more than a feeling. Expected value (EV) turns it into a number — the average profit you’d make per dollar if you could place the same bet thousands of times. It’s the whole game in one formula.

The first term is what you win when you’re right, weighted by how often that happens. The second term is the stake you lose when you’re wrong. If the result is positive, the bet is worth making; if it’s negative, you pass — no matter how much you like the side. Convert American odds to decimal first (Course 01) so the formula plugs in cleanly.

A worked example

Suppose you’re looking at an underdog priced +150. That’s decimal 2.50, and its raw implied probability is 100 ÷ 250 = 40%. But you’ve done your homework and you believe this team really wins about 45% of the time. Your number beats the price’s number — so there should be value. Let’s prove it.

Read that result carefully. This bet loses 55% of the time — most individual bets on it are losers. And you should still make it, every single time it’s offered, because across many tries it returns $12.50 on the average $100. That counterintuitive truth is the heart of value betting: you are not betting on this game, you are betting on the price being wrong.

Where your probability comes from

EV is only as good as the p you feed it. Garbage in, garbage out — a made-up probability produces a made-up edge. So where does an honest number come from?

• A model. A repeatable, data-driven estimate built from the factors that actually move outcomes. You’ll build one in Course 11.
• An information edge. Something true that the market hasn’t fully priced yet — a lineup change, an injury, weather, a situational angle — used before the line adjusts.
• A sharper book’s price. The most reliable starting point for most bettors: take the no-vig fair price from a market-setting book and treat that as your estimate of true probability. If a softer book is offering a worse-for-them price than the sharp fair number, that gap is your edge.

Whatever the source, be honest about uncertainty. If you say a team wins 45% but you’d really only defend “somewhere between 40% and 50%,” your edge might be real or might be zero. Padding your estimate to manufacture value is the fastest way to go broke while feeling smart. A small, defensible edge is worth more than a big, imaginary one.

The long run is the only run

A positive-EV bet is a good bet before the outcome is known, and it stays a good bet whether it wins or loses. The edge shows up only over volume. Flip back to that +150 example: it’s +12.5% EV, yet it loses 55% of the time. Over a single weekend you could lose five of them in a row and conclude you’re terrible at this. Over a thousand of them, the math is almost impossible to escape.

This is why sharp bettors judge themselves by process, not last weekend’s scoreboard. Short-term results are noisy — they’re mostly variance, not skill. The signal you actually want is whether you’re consistently beating the closing price, a measure called closing line value (Course 12). If you’re reliably getting better numbers than the market settles on, the profit follows even when this week didn’t. Win or lose, ask the same question: was the price wrong when I bet it?

Bet the price, not the team

Everything in this course collapses into one rule. The quality of a team and the quality of a bet are different things, and confusing them is the most expensive mistake in betting.

• A great team at a bad price is a bad bet — you’re overpaying for an outcome that’s already priced in.
• A mediocre team at a great price is a good bet — you’re being paid more than the risk is worth.

“Who wins?” is the sportsbook’s question, and it has already answered it inside the odds. Your question is different and more profitable: “Is this price wrong, and in my favor?” Get in the habit of looking at the number before you look at the matchup.

Key takeaways

Check yourself