Implied probability turns a betting price into a percentage chance. Take the decimal odds, divide 1 by that number, and multiply by 100 — that’s the probability the odds are telling you the outcome carries. A price of 2.00 implies exactly 50%; a price of 1.33 implies 75%; a price of 10.00 implies 10%. It’s the fastest way to read what a bookmaker actually thinks about a result, stripped of the marketing language around “favorites” and “long shots.”

The catch is that bookmaker odds are never a pure reflection of true chance. Every market carries a built-in margin — often called the overround or vig — which means the implied probabilities of all outcomes in a market add up to more than 100%. That extra percentage is the bookmaker’s edge, baked into the price before you place a single dollar. If you only look at implied probability from one side of a market, you’re seeing the bookmaker’s estimate plus their cut, not a clean number.

This is why implied probability matters more as a comparison tool than as a standalone fact. Serious bettors calculate it constantly, not to marvel at how likely something is, but to check it against their own estimate of the true probability. If your model says a result is more likely than the odds imply, the bet has positive expected value — at least according to your model. If your number is lower, the price isn’t worth backing, even if the pick “feels” right.

Example

Suppose a Champions League quarter-final between Inter Milan and Borussia Dortmund is priced by a sportsbook as: Inter to win at 1.95, Draw at 3.60, Dortmund to win at 4.00.

Converting each to implied probability:

  • Inter: 1 ÷ 1.95 = 0.5128 → 51.28%
  • Draw: 1 ÷ 3.60 = 0.2778 → 27.78%
  • Dortmund: 1 ÷ 4.00 = 0.2500 → 25.00%

Add them up: 51.28 + 27.78 + 25.00 = 104.06%. That extra 4.06 points is the overround — the book’s margin for offering this market. To get the “fair” probabilities the market is really pricing in, divide each figure by 1.0406:

  • Inter: 51.28 ÷ 1.0406 = 49.28%
  • Draw: 27.78 ÷ 1.0406 = 26.69%
  • Dortmund: 25.00 ÷ 1.0406 = 24.02%

Those three now sum to almost exactly 100%, and that’s the bookmaker’s genuine view of the match once their commission is stripped out.

Now say you’ve built a model — using recent form, injuries, and home advantage — that puts Inter’s true winning chance at 55%. The market’s fair-adjusted figure is 49.28%. Your estimate is higher than the market’s, which means the 1.95 price is offering more return than Inter’s actual risk of losing justifies, in your model’s view. A $50 stake at 1.95 returns $97.50 if Inter wins. At your assessed 55% chance, the expected value of that bet is (0.55 × $97.50) − $50 = $3.63 positive per $50 staked — a small but real edge, assuming your 55% figure holds up. If your model instead agreed with the market’s 49.28%, there’d be no edge here at all, regardless of how confident you personally feel about Inter.

Key Points

  • Calculate it as 1 ÷ decimal odds: A price of 2.50 implies 40% (1 ÷ 2.50 = 0.40); a price of 1.20 implies 83.3%. Always work from decimal odds first — converting from fractional or American prices to a percentage in your head invites errors.
  • The market always overstates 100%: If you add up every outcome’s implied probability in a two-way or three-way market and it comes to more than 100%, that gap is the bookmaker’s margin, not extra risk in the event itself.
  • Strip the margin before comparing to your own view: Normalize each outcome’s implied probability by dividing by the market total (as in the Inter–Dortmund example) before judging whether a price is generous or mean — comparing your model to the raw, un-normalized figure understates the market’s real assessment.
  • Shop for the lowest implied probability on your side: Two bookmakers might price the same tennis Grand Slam final at 1.80 and 1.91 on the same player. The 1.91 price implies 52.4% versus 55.6% — a meaningfully softer number for backing that player, even though the outcome hasn’t changed.
  • Track how it moves, not just where it sits: If a horse’s implied probability climbs from 20% to 28% in the hour before a race with no obvious news, that’s money talking — informed action shifting the price — and it’s worth knowing before you bet into it.
  • Don’t confuse implied probability with your win rate: Winning a bet priced at 25% implied probability one time in four is expected, not lucky, and losing it three times in four at that price is not proof the bet was wrong — judge the process over a large enough sample, not any single result.