Political betting odds are a price, and every price can be turned into an implied probability. On prediction markets like Kalshi, Polymarket and PredictIt it’s easy: a contract trading at 62¢ implies about a 62% chance the event happens. Sportsbook-style odds need a formula (below), and in every format the probabilities across all outcomes add up to a bit more than 100%, so you remove that extra before comparing.
This guide covers the conversions, the quick table, how to strip out the vig, and what these numbers do and don’t tell you about an election.
How do prediction market prices work? #
A prediction market contract pays $1 if an outcome happens and $0 if it doesn’t. “Will Party X win the Senate in 2026?” might trade at 58¢ for Yes. Because the payoff is a dollar, the price in cents reads directly as the market’s implied probability: 58¢ ≈ 58%.
Two details make the real number slightly fuzzier:
- Bid and ask. You’ll usually see a highest bid (what buyers will pay) and a lowest ask (what sellers will accept). If Yes has a 57¢ bid and a 59¢ ask, the midpoint of 58¢ is the fairest single estimate.
- Yes plus No can exceed $1. If you can buy Yes at 59¢ and No at 43¢, the two add to 102¢. The extra 2¢ is the spread, and it works like the vig described below.
How do you convert American (moneyline) odds to probability? #
US sportsbooks quote odds relative to a $100 stake. A minus sign marks the favorite, a plus sign the underdog.
Negative odds (favorite): Implied probability = |odds| ÷ (|odds| + 100)
Example: −150 → 150 ÷ 250 = 60%
Positive odds (underdog): Implied probability = 100 ÷ (odds + 100)
Example: +150 → 100 ÷ 250 = 40%
At −150 you risk $150 to win $100. At +150 you risk $100 to win $150.
How do you convert decimal odds to probability? #
Decimal odds, common in Europe, Canada and Australia, show the total return per $1 staked, including the stake.
Implied probability = 1 ÷ decimal odds
Example: 2.50 → 1 ÷ 2.50 = 40%. A $1 stake returns $2.50 ($1.50 profit plus your dollar).
How do you convert fractional odds to probability? #
Fractional odds, traditional in the UK and Ireland, show profit relative to stake. For odds of A/B:
Implied probability = B ÷ (A + B)
Examples: 3/1 → 1 ÷ 4 = 25%. 1/4 (an “odds-on” favorite) → 4 ÷ 5 = 80%.
Quick conversion table #
| Implied probability | Market price | American | Decimal | Fractional |
|---|---|---|---|---|
| 90.9% | 91¢ | −1000 | 1.10 | 1/10 |
| 80.0% | 80¢ | −400 | 1.25 | 1/4 |
| 66.7% | 67¢ | −200 | 1.50 | 1/2 |
| 60.0% | 60¢ | −150 | 1.67 | 2/3 |
| 50.0% | 50¢ | +100 | 2.00 | 1/1 |
| 40.0% | 40¢ | +150 | 2.50 | 3/2 |
| 33.3% | 33¢ | +200 | 3.00 | 2/1 |
| 20.0% | 20¢ | +400 | 5.00 | 4/1 |
| 10.0% | 10¢ | +900 | 10.00 | 9/1 |
Why do the probabilities add up to more than 100%? #
Convert every outcome in a market and you’ll usually get a total above 100%. The excess is the overround, also called the vig, juice or margin. At a sportsbook it’s the house’s built-in profit. On an exchange it comes from the bid-ask spread and fees.
Example: a two-candidate race quoted at decimal 1.80 and 2.10.
- 1 ÷ 1.80 = 55.56%
- 1 ÷ 2.10 = 47.62%
- Total = 103.17%
To get fair probabilities, divide each by the total:
- 55.56 ÷ 103.17 = 53.85%
- 47.62 ÷ 103.17 = 46.15%
Now they sum to 100%. Do this every time you compare odds from different sources or line them up against a forecast model. Without it, a sportsbook favorite looks a few points more likely than the price really implies.
What does a 60% price actually mean? #
It means traders, collectively, think the outcome is somewhat more likely than not. It doesn’t mean the candidate is ahead 60–40 in the polls, and it doesn’t mean they’ll get 60% of the vote. A candidate tied in the polls can trade at 60% if traders believe structural factors favor them: the state’s lean, incumbency, fundraising, or an expected polling error.
It also means the other side wins 4 times out of 10. A 60% favorite losing is not a market “failure.”
What moves political odds? #
- New polls, especially in a race with little polling.
- Candidate news: entries, withdrawals, endorsements, scandals, health.
- Debates and major speeches, often within minutes.
- Economic data and approval ratings, for markets on chamber control.
- Legal and procedural rulings, such as court decisions on district maps.
- Large trades. In thin markets, one big order can move the price well beyond what the news justifies. See can election prediction markets be manipulated?.
Where do political odds go wrong? #
- Favorite-longshot bias. Traders tend to overpay for long shots, so a 3¢ candidate is often less likely than 3%.
- Thin markets. Obscure primaries and down-ballot races can have little money behind them, so the price reflects a few traders’ views.
- Who trades. Market participants aren’t a cross-section of voters, and markets can follow a popular narrative. In 2022, markets leaned heavily toward a Republican Senate takeover that didn’t happen. Our comparison of prediction markets vs. polls looks at the track record.
- Fees and capital costs. Money tied up for months until an election resolves has a cost, which can push prices on near-certain outcomes below their true probability.
Are political prediction markets legal in the US? #
The picture changed a lot in 2024 and 2025. Kalshi is a CFTC-regulated exchange that won the right to list election contracts after a 2024 court ruling. PredictIt received CFTC approval in September 2025 to operate as a regulated exchange and raised its per-contract limit from $850 to $3,500. Polymarket, which had blocked US users since a 2022 settlement, returned to the US in late 2025 through a CFTC-licensed exchange it acquired. Several states have challenged event contracts in court, mostly over sports. Rules and access are still shifting, so check each platform’s current terms. Our explainer on whether political prediction markets are legal has the details.
If you trade, remember that you can lose everything you put in. Treat prices as a signal about the race, not a reason to bet.
How to follow market odds without trading #
You don’t need an account on any platform to watch prices. Election Tracker, our free iPhone app, has a Market Sentiment tab that shows current Polymarket prices, read-only, for markets such as 2026 House and Senate control and the 2028 nominations, with 24-hour moves on the leading outcomes. A disclaimer at the top of the screen says the prices are trader sentiment, not forecasts, and the app takes no wagers. The same app’s polling tabs show race-by-race polling averages, so you can see when the market and the polls disagree and go looking for the reason.
Frequently asked questions #
How do I convert a prediction market price to a percentage? #
Read the price in cents as a percentage. A contract at 42¢ implies roughly a 42% chance. If there’s a gap between the bid and ask, use the midpoint.
How do I calculate implied probability from −200 odds? #
Divide 200 by (200 + 100): 200 ÷ 300 = 66.7%. For positive odds like +200, divide 100 by (200 + 100) to get 33.3%.
What is the vig in betting odds? #
The vig (vigorish) is the margin built into odds so the implied probabilities add up to more than 100%. Remove it by dividing each outcome’s implied probability by the total of all outcomes.
Are betting odds more accurate than polls? #
Sometimes, sometimes not. Markets react faster and fold in more information, but they’ve also missed badly, as in 2022. They’re most useful read alongside polling averages rather than instead of them.
Why do odds on different sites disagree? #
Different fees, different traders, different liquidity and, at sportsbooks, different margins. After removing the vig, reputable markets on the same event usually land within a few points of each other.