Prediction Market Odds: Polymarket and Kalshi

What the two biggest prediction markets are pricing right now, in one ranked feed: the implied probability of each outcome, how much is actually trading today, and when it resolves.

No API key, no login, no proxy.

What does this do?

Prediction markets are the closest thing to a live probability on a question nobody can poll — a rate decision, an election, a nomination. But Polymarket and Kalshi each publish their own book, in their own shape, with their own units, and every existing Actor covers one venue.

This reads both, normalises them into one table, and ranks by what is actually trading today.

flowchart LR
    A[Polymarket<br/>dollar volume] --> C[One normalised feed]
    B[Kalshi<br/>contract volume] --> C
    C --> D[Implied probability,<br/>activity, time to resolve]

Who is it for?

You are	You use it to
A trader	See where the money actually is, across both venues, in one list
A journalist or analyst	Quote a live market probability instead of a stale poll
Running a dashboard or bot	Pull a clean odds feed without a data licence
Researching sentiment	Track how a probability moves as news lands

What data do you get?

Field	What it is
venue, question, outcome	Which market, the question, and the specific outcome
probabilityPercent, priceUsd	The implied probability, and the raw price
confidence	The same number in words: genuinely uncertain, likely, all but ruled out
volume24h, volumeTotal, liquidity, openInterest	How much is trading now and in total
priceChangeSincePrevious	Movement, on Kalshi, without this Actor storing any history
priceChangeWeek, priceChangeMonth	Movement, on Polymarket
daysUntilResolution, resolvesAt	When it settles
decided	Whether the market has effectively answered the question already
category, url	Where it sits, and a link to the market

Example: one run, 24 September 2026

13,037 markets read (1,600 Polymarket outcomes, 11,437 Kalshi markets), run finished in nine seconds, filtered to the 100 most active, 50 from each venue:

Venue	Question	Implied	Traded in 24 h
Polymarket	Dota 2: Natus Vincere vs LGD Gaming	55.5% / 44.5%	$821,283
Polymarket	Valorant: Team Liquid vs Paper Rex	3.5% / 96.5%	$744,336
Kalshi	2028 U.S. Presidential Election winner: Gavin Newsom	8.7%	337,651 contracts
Kalshi	2028 Democratic presidential nominee: Andy Beshear	3.9%	90,491 contracts

Kalshi's busiest markets trade hundreds of thousands of contracts a day, and its election markets hold millions in open interest.

What it refuses to do

• It never adds dollars to contracts. Polymarket reports volume in US dollars; Kalshi reports it in contracts. They are not the same unit, so they are never summed, and every row states its venue.
• It does not let one venue bury the other. Sorting both into a single list by "24-hour volume" looks reasonable and is wrong: $821,283 on Polymarket and 337,651 contracts on Kalshi look comparable and are not, so a merged sort ranks by whichever unit happens to run larger. An early run ranked that way returned sixty Polymarket rows and no Kalshi rows at all. Each venue is now ranked on its own terms and the list takes turns between them.
• It ranks on activity now, not lifetime volume. The single largest market by lifetime volume was $83M with $134 traded in the last day, priced at 0.003 — long since decided. Ranking on lifetime volume fills the table with corpses.
• It will not claim two venues are pricing the same question unless they clearly are. Cross-venue matching needs several distinctive words in common. A question reduced to one word would otherwise match anything containing it and report a fictional price gap. In practice the overlap is thin — the two venues list largely different things — and the Actor says so rather than padding the output.
• It uses the endpoint that actually works. Kalshi's market list returns multi-leg parlay rows with unreadable titles and no prices; the event endpoint returns the real question with its markets priced underneath.

Popular prediction market reports

Ready-made runs you can open and start, or copy as a starting point:

• Prediction Market Odds Today
• Election Odds
• Presidential Election Odds
• Politics Prediction Markets
• Fed Interest Rate Odds
• Economics Prediction Markets
• Bitcoin Price Odds
• Ethereum Price Odds
• Sports Prediction Markets
• NFL Prediction Markets
• Entertainment Prediction Markets
• Oscars and Awards Odds
• Polymarket Odds
• Kalshi Odds
• Polymarket vs Kalshi Price Differences
• Markets Resolving This Week

Pricing

$0.002 per market returned. The hundred most active markets across both venues costs $0.20.

Input

Setting	What it does
Venues	Both by default, or one on its own
Only questions mentioning	Filter to a topic, for example "fed", "election" or "bitcoin"
Only this category	Kalshi labels events Politics, Economics, World and so on
Minimum activity in the last 24 hours	Skip quiet markets. Remember the units differ per venue
Hide markets that are already decided	Drops anything beyond 97% or 3%
Resolving within this many days	Optional, for near-term questions only
Look for the same question on both venues	Reports price gaps where genuine overlap exists
Maximum markets	How many to return, most active first

Integrations

Runs on a schedule and connects to Google Sheets, Slack, Zapier, Make or your own code through the Apify API. Run it hourly against a search term and you have a probability time series for whatever question you care about.