Implied probability explained

What implied probability actually means

Every odds price is a probability in disguise. When a book posts -200 on a favorite, it is saying "we treat this outcome as about 66.7% likely." Implied probability is just that number pulled back out of the odds.

The key word is *implied*. It is the book's price, not the true probability of the event. The book shifts the number to protect its margin and to balance the money coming in on each side. Your job as a bettor is to compare the implied probability against your own estimate of the real chance. If you think a team wins 55% of the time and the price implies only 48%, that gap is where value lives.

Implied probabilities across a single market always add up to more than 100%. That extra slice is the sportsbook's margin (also called the vig, juice or overround). On a coin-flip market priced -110 / -110, both sides imply 52.4%, totaling 104.8% — the 4.8% over 100 is the house edge.

The formulas: American, decimal and fractional odds

US books quote American odds by default, but you'll see decimal and fractional formats too. Each converts to a percentage with a simple formula.

**American odds (minus / favorite):** Implied probability = negative odds ÷ (negative odds + 100) Example: -150 → 150 ÷ (150 + 100) = 150 ÷ 250 = 0.60 = 60%

**American odds (plus / underdog):** Implied probability = 100 ÷ (positive odds + 100) Example: +130 → 100 ÷ (130 + 100) = 100 ÷ 230 = 0.435 = 43.5%

**Decimal odds:** Implied probability = 1 ÷ decimal odds Example: 2.50 → 1 ÷ 2.50 = 0.40 = 40%

**Fractional odds:** Implied probability = denominator ÷ (numerator + denominator) Example: 6/4 → 4 ÷ (6 + 4) = 4 ÷ 10 = 0.40 = 40%

Quick reference points worth memorizing: -110 = 52.4%, +100 (even money) = 50%, -200 = 66.7%, +200 = 33.3%, +150 = 40%.

Worked example: an NFL point-spread market

Say you're looking at an NFL spread and both sides are priced at -110, the standard juice on point spreads and totals.

Favorite -110 → 110 ÷ 210 = 52.38% Underdog -110 → 110 ÷ 210 = 52.38% Market total = 52.38% + 52.38% = 104.76%

That 4.76% over 100% is the sportsbook's margin on this market. To find the fair, vig-free probability of each side, divide each implied probability by the market total:

Fair favorite chance = 52.38% ÷ 104.76% = 50.0% Fair underdog chance = 52.38% ÷ 104.76% = 50.0%

Now shift the price. Suppose the favorite is -130 and the underdog +110: Favorite -130 → 130 ÷ 230 = 56.52% Underdog +110 → 100 ÷ 210 = 47.62% Total = 104.14% Fair favorite chance = 56.52% ÷ 104.14% = 54.3% Fair underdog chance = 47.62% ÷ 104.14% = 45.7%

If your own model says the underdog wins 50% of the time, the +110 price (implied 47.6%, fair 45.7%) is offering more than you think is warranted — a positive-expectation spot. That single subtraction is the whole point of learning implied probability.

Turning implied probability into expected value

Once you have both your estimate and the book's implied probability, you can size up expected value (EV) on a stake.

EV = (your probability × profit if you win) − (chance you lose × stake)

Work a $100 bet at +110 (profit $110) where you rate the true chance at 50%: EV = (0.50 × $110) − (0.50 × $100) = $55 − $50 = +$5

A positive EV of $5 per $100 staked means the price is in your favor over the long run. Flip it: if the true chance were only 45%, EV = (0.45 × $110) − (0.55 × $100) = $49.50 − $55 = −$5.50, and you should pass.

The break-even point is the implied probability. At +110 you need to win 47.6% of the time just to break even. Any genuine edge over that number is profit; anything below it bleeds money regardless of how the individual bet lands.

How sportsbook margin changes your break-even number

Tighter margins mean the implied probabilities across a market sit closer to the true 100%, so you keep more of your edge. We track average margins on a sample of US sportsbook prices, and the spread between books is real.

Using our sampled data, the tightest average margins belong to Fanatics Sportsbook (4.25%, 466 markets sampled), DraftKings (4.25%, 1,068 sampled) and FanDuel (4.46%, 596 sampled). Middle of the pack are Caesars Sportsbook (5.27%, 906 sampled) and BetMGM (5.21%, 438 sampled). The widest in our sample is BetRivers at 6.35% (480 sampled).

What that means in implied-probability terms: on a two-way market a 4.25% margin loads roughly 2.1% of overround onto each side, while a 6.35% margin loads about 3.2%. That difference compounds across every bet you place. On the same coin-flip market, a tight book might quote -104 / -104 while a wider book posts -110 / -110 — and your break-even climbs from 51% to 52.4%.

DraftKings also holds the highest best-price share in our sample at 54.0%, meaning it posted the top available price more often than any other book we compared. FanDuel (36.1%), Caesars (36.2%) and Fanatics (41.8%) trail. Best-price share and average margin are the two numbers to watch if you line-shop seriously.

Which sportsbook fits a probability-focused bettor

Implied probability only pays off if you shop the number across several licensed books. All operators below are licensed by state gaming regulators (NJ DGE, NYSGC, PGCB, MGCB and others); the minimum age is 21, and a licence in one state does not make a book legal in another.

• **Choose DraftKings** if line-shopping is your core habit — the highest best-price share (54.0%) and a joint-tightest 4.25% average margin in our sample make it the reference price for many markets. Rated 7.9 for odds, 9.2 app.
• **Choose FanDuel** if you want tight margins (4.46%) plus a top-tier app (9.3) and Same Game Parlay depth. Best-price share of 36.1% is lower, so still compare.
• **Choose Fanatics Sportsbook** for a 4.25% average margin and a strong 41.8% best-price share on our sample — a value-friendly newer entrant (odds 7.4, app 8.7).
• **Choose bet365** for the deepest market coverage (9.6) and live capability (9.4) if you want the most prices to convert and compare, though we hold no margin sample for it.
• **Avoid leaning on BetRivers for closing value** if you're margin-sensitive: its 6.35% sampled average is the widest here, meaning higher break-even points per bet, even though withdrawals rate a solid 8.6.

Having two or three of these open lets you convert each price to a percentage and take the side offering the lowest implied probability for the outcome you want — the practical payoff of everything above.