BY 
Every betting line is a probability statement with a fee hiding inside it. Once you can see both parts, a lot of confusing things about sportsbooks start making sense, including how they post odds on hundreds of games a night without sweating. Today I'm unpacking that math, with worked examples you can check on a napkin.
Converting American odds to implied probability
American odds come in two flavors, and each one has its own conversion formula.
For negative odds, divide the odds by the odds plus 100, ignoring the minus sign. So -110 becomes 110 / (110 + 100), which is 110 / 210, or 52.38%. For positive odds, divide 100 by the odds plus 100. So +150 becomes 100 / (150 + 100), which is 100 / 250, or exactly 40%.
A few more, because repetition is how this sticks:
- -200: 200 / 300 = 66.67%
- -150: 150 / 250 = 60.00%
- +130: 100 / 230 = 43.48%
- +300: 100 / 400 = 25.00%
That percentage is the win rate you need just to break even at that price. Bet -200 favorites all season and you need two wins out of every three. Bet +300 longshots and one win in four keeps you level.
The vig: why both sides add up to more than 100%
Now run the conversion on a standard point spread priced at -110 on each side. Each side implies 52.38%, and 52.38% plus 52.38% comes to 104.76%. Real probabilities can't sum past 100%, so that extra 4.76% is the overround, the book's margin built directly into the prices. Bettors call it the vig or the juice.
What does it cost you in dollars? Bet 110 on both sides of that spread, a guaranteed coin flip against yourself. One ticket wins 100 and the other loses 110, so you're out 10 on 220 risked. That works out to roughly a 4.5% expected loss on a pure coin flip, before you've made a single actual prediction.
Sit with that number for a second. A bettor with zero skill doesn't break even at -110; they lose around 4.5 cents of every dollar, forever. The whole project of beating a sportsbook is the project of overcoming that toll.
Removing the vig to find the fair line
Since the book's percentages are inflated, you can deflate them back into a fair estimate. Divide each side's implied probability by the combined total. That's the whole trick.
At -110 on both sides it's clean: 52.38 / 104.76 equals 50% each way. The no-vig line on a symmetric spread is a coin flip, which is exactly what you'd expect.
Asymmetric prices are where this earns its keep. Take a moneyline of -150 on the favorite and +130 on the dog. The favorite implies 60% and the dog implies 43.48%, for a combined 103.48%. Divide through: the favorite's fair probability is 60 / 103.48, or 57.98%, and the dog's is 42.02%.
Convert 57.98% back into American odds and the fair price on that favorite is roughly -138, not the -150 you're being charged. Now you have a real benchmark. If your handicapping says the favorite wins 61% of the time, you can compare 61% against 57.98% and know you genuinely disagree with the market, rather than just staring at an artifact of the juice.
Why props and parlays are harder to beat
That 4.5% toll is the friendly case. Main spreads and totals at major books are the lowest-margin products on the menu, and nearly everything else charges more.
Player props routinely sit at -115 or -120 on both sides. Run the same math on a -120 / -120 market and the total implied probability jumps to 109.1%, roughly double the overround of a standard spread. Your handicapping has to clear twice the toll just to tread water, in markets where information is thinner and limits are lower.
Parlays compound the problem rather than merely adding to it. Two -110 legs multiplied together pay about +264. But two true coin flips both hitting is a 25% proposition, and fair odds on 25% are +300. You're being paid +264 for a +300 shot because the margin got baked in twice, once per leg, and every leg you add multiplies it again.
None of this makes those markets unbeatable. It just raises the bar, and you should know exactly how high the bar sits before you jump.
Putting the conversions to work
During my first week as a projections analyst, the first spreadsheet I ever built was a vig remover. That was years ago, and I still keep a copy of it open in a browser tab most days. Old habits.
Practice on live prices until the conversions become automatic. Pull up today's games on the daily feed, pick any matchup, convert both sides, and compute the hold. Do it ten times and you'll start spotting expensive markets on sight.
This math is also why CAPTRACKER ranks handicappers by units won and ROI on the leaderboard instead of raw win rate. A 60% record sounds impressive until you notice it came at -150, where 60% is exactly breakeven. Prices decide everything, and the vig is the quiet tax sitting underneath all of it.
If any term here felt unfamiliar, the glossary has plain-language definitions, and my beginner guide to reading odds starts one step earlier than this post did. Learn the conversion, remove the vig, and you'll never mistake a price for a probability again.