How to compare odds across US sportsbooks
US sportsbooks do not agree on a price, and the gaps between them are wider than most bettors assume. Comparing properly means comparing margins rather than headline numbers — and taking the best price consistently is one of the few improvements that requires no forecasting skill at all.
Published
How do you compare odds across sportsbooks?
Start by lining up the same outcome at every book that prices it, then convert each American price into the probability it implies. The price that implies the lowest probability is the better deal — you are being asked to believe the outcome is less likely, for a bigger payout.
implied probability (+odds) = 100 ÷ (odds + 100)
-110 → 52.4% · -105 → 51.2% · +100 → 50.0%
Headline comparison alone is not enough, though. A book can lead the market on one side and be well short on the other, so it is worth also adding up the whole market to see its hold. The book with the lowest total margin is charging you least across everything it prices, not just on the one selection you happened to look at.
Does taking the best price actually matter?
More than most people expect, because the effect compounds. A winning bet at -105 returns about 2.3% more than the same bet at -110. That sounds trivial on one bet and is not remotely trivial across a season — for many bettors it is larger than the edge they are trying to find in the first place.
It is also the rare improvement that costs nothing in skill. Finding a genuine edge is difficult and uncertain. Taking the best available price is arithmetic: the same bet, the same opinion, more money back when it wins. Whether it beat the market is then measured after the fact by closing line value.
Why do sportsbooks price the same market differently?
Because each is managing its own book, not tracking a shared truth. A sportsbook taking heavy action on one side will move that side's price and improve the other to reduce its liability, whatever competitors are showing. That adjustment has nothing to do with the true probability changing.
On top of that, smaller books often follow larger ones with a lag, and attention is unevenly distributed: a marquee moneyline is watched closely by everyone, while a single player prop may be priced once and left. Disagreements are therefore widest exactly where the least money is being bet.
How many sportsbook accounts do you need?
There is no correct number, and the honest answer is fewer than most comparison sites imply. You can only take a price at a book that is legal in your state, where you hold an account and have funds, so a comparison across every book is useful only to the extent you can act on it. More accounts also means more admin and more money sitting idle across balances.
The benefit tends to flatten well before you have opened everything available — the largest gains come from covering the books that genuinely lead markets in the sports you bet, not from breadth for its own sake.
Doing it by hand, or not
None of this requires software. A spreadsheet and a handful of browser tabs will compare a market perfectly well, and for a bettor placing a few bets a week that is often enough. What does not scale is doing it across every market, every book and every price change — the arithmetic is trivial and the volume is not.
HaterPicks compares 14 US sportsbooks and prediction markets on one screen for that reason, with the model's projection beside each line. It is information and tooling, not betting advice. You can also read how HaterPicks works, see how it compares to other tools, or work a single market through the EV calculator.