Prediction markets offer real-time crowd-consensus probability data anchored by real money. Citing them well requires knowing what you are quoting, where it came from, and what the caveats are. This guide covers all three.
A prediction market price is the aggregate willingness of participants to bet on an outcome at that probability. It updates in real time as traders buy and sell contracts.
It is not a poll, a statistical model, or an official projection. It is a market price — shaped by whoever is willing to stake money on it at that moment.
Always include a timestamp. A price is only meaningful at a specific point in time. "The market shows 62%" is incomplete without a date and time.
US regulated exchanges — such as Kalshi (CFTC DCM + DCO) and Polymarket (CFTC DCM (via QCX LLC)) — operate under CFTC oversight. International platforms may have different regulatory structures. Name the specific platform and state its regulatory status when relevant to your story.
Run through these before quoting any prediction market price in your article.
What exchange is this from?
Name the specific platform. “Prediction markets” is not a single data source — Kalshi, Polymarket, and PredictIt each run independent markets with different user bases, fee structures, and resolution mechanics. A price on one platform is not automatically the same as a price on another.
See all exchangesDoes this event trade on multiple platforms?
If the same question trades on more than one exchange, compare the prices. A gap larger than 5 percentage points is itself a story: it means different pools of informed traders disagree, or the market structures differ in ways that change the signal.
Cross-platform lookup toolHow liquid is this market?
A market with very little trading activity can be moved by a single large order and should not be cited as representative. Before quoting a price, confirm the market has meaningful depth — check volume at the platform's official site.
No single exchange has a monopoly on the signal. Here is what you gain — and miss — by citing only one.
PredictionMarkets.US pulls odds from Kalshi, Polymarket, and PredictIt simultaneously — so you can see whether they agree before quoting any single price. Open the journalism tools
Not every prediction market price is equally safe to cite. Avoid these without additional qualification in your copy.
Markets priced at 95%+ or 5% or lower
At these extremes, the outcome is near-certain and markets reflect what the crowd already knows. Citing a 97% price as news implies uncertainty where little exists.
Screenshots without a timestamp
Prediction market prices update continuously — sometimes within seconds. An undated screenshot is not a citable source. Require an exact date and time (with timezone) from whoever supplies the image.
Thin markets without volume confirmation
Low-volume markets reflect fewer participants and are easier to move. Confirm depth before citing as representative. If you cannot verify volume, note the uncertainty in your copy.
Aggregated “average” odds without source disclosure
If you are averaging across platforms, disclose that — and show the underlying prices. Unexplained averages obscure where the signal actually came from and make fact-checking impossible.
These are specimen examples — adapt the bracketed fields to your event. Each pattern can survive an editor's fact-check.
"Kalshi traders put the probability at 62% as of [date, time ET]. The exchange is regulated by the CFTC."
Include the exact timestamp. Name the exchange. State the regulatory body.
"Across major prediction markets, the probability ranged from 58% to 65% as of [date], indicating broad consensus but meaningful uncertainty."
Name the range, not an unsourced average. Dates on each platform reading help editors verify.
"Two platforms showed a 12-point gap — one at 62%, another at 50% — reflecting different user bases and settlement structures."
Divergence is a journalistic finding, not a problem to smooth over. Report it.
Every prediction market platform has a business interest in directing users — and journalists — toward its own data. A tool provided by a single exchange tells you what that exchange shows. It cannot tell you whether another exchange agrees.
The most defensible citation is one where you have checked multiple independent markets and can speak to where they agree, where they diverge, and what that divergence means. That requires a cross-platform view.
PredictionMarkets.US has no affiliate stake in any exchange outcome. The prices we display are pulled from platform APIs and shown as-reported. Our role is aggregation, not advocacy.
Continue exploring
