How we test & rank prediction market apps

No platform is added to a Top-Apps list until our editorial team has tested it with real money, on real markets, end-to-end. We do not rely on press releases, vendor demo accounts, or a competitor's writeup.

What "tested" means

• •Funded with at least $250 of editorial capital per platform — never the operator's house money or a sponsored account.
• •A minimum of 25 live trades placed across three or more market categories (sports, politics, economics, crypto, weather, culture).
• •At least one deposit and one withdrawal tested per supported rail (ACH, debit card, crypto where applicable).
• •KYC flow tested with a real California ID — we document any friction, document re-uploads, and time-to-approval.
• •Customer support contacted at least twice through each channel the operator advertises (in-app chat, email, Twitter/X), with response time logged.

Test details — including the exact dollar amounts, contracts traded, and rails tested — are documented at the top of each platform's individual review.

Prediction-market law moves fast — sometimes weekly. We treat regulator action the same way a financial newsroom treats an SEC filing: as a primary source that overrides anything an operator's PR team says.

Sources we monitor

• •CFTC press releases, no-action letters, enforcement actions, and DCM order books.
• •PACER court filings for any active Kalshi / Polymarket / KalshiEX litigation, including state AG suits.
• •State gambling regulators in all 50 states — cease-and-desist letters trigger a same-week update on our state legality tracker.
• •California-specific channels: the California Bureau of Gambling Control, the California Gambling Control Commission, and the state Attorney General's gaming filings.
• •SEC filings for any prediction-market-adjacent tokens, custodians, or settlement rails.

When the law is unsettled — which it often is — we say so plainly. We will never present a contested legal interpretation as a settled fact, and we will not editorialise away genuine risk to make an affiliate link more clickable.

We earn affiliate commissions when readers sign up for prediction-market platforms via our outbound links. That revenue does not — and cannot — influence our rankings, scoring, or editorial language. Here's how we enforce it.

Journalists covering this beat can find pre-cleared quotes, brand assets, and contact details in our press kit.

We follow the FTC's Guides Concerning the Use of Endorsements and Testimonials in Advertising (16 CFR Part 255). In practice that means:

• •Affiliate relationships are disclosed above the fold on every page that contains commercial links — not buried in a footer.
• •Bonus claims are quoted from the operator's own published terms. We do not invent dollar figures or rollover requirements to make an offer sound better.
• •Language like "guaranteed," "risk-free," or "easy money" is never used. Trading involves real loss risk and we say so on every page that promotes a platform.
• •We do not market to anyone under 18. Every sign-up CTA carries the 21+ trade-responsibly badge with a link to Responsible Trading.

Every factual claim on this site falls into one of three tiers. We tag each source by tier internally and surface citations in the Sources block at the bottom of every guide.

Numbers we do not invent

User counts, trading volumes, valuations, and bonus values are quoted from the operator's own materials or a Tier-1/2 source — never estimated. If a number cannot be verified, we either omit it or describe it qualitatively ("deepest US liquidity in 2025") instead of inventing a figure.

Our rubric is the Venice Score: a 0–100 rating built from six weighted checks, all run from a California IP on a funded account. Cash-out reliability carries the single heaviest weight because it is the part of a prediction-market app that most reviews never test. A platform that fails a withdrawal test is capped at 60 regardless of the other five scores, and any editor override has to be written on the review page itself.

Prediction-market regulation, operator availability, and bonus terms change frequently. We publish a freshness commitment so readers know exactly how recent the page they are reading is.

• •Top-Apps list & individual reviews: reviewed at minimum every 30 days; bonus terms re-verified each cycle.
• •California + state legality pages: reviewed monthly; same-week updates when a regulator action, court ruling, or cease-and-desist is published.
• •News articles: dated at publication and not silently rewritten — material updates are added as a "Update:" block at the top.
• •Evergreen guides: reviewed quarterly; obviously dated content (e.g. discontinued products) triggers an immediate revision.

Every page carries a "Last updated" date. The freshness chip near our bylines turns from cyan to grey once a page is more than one month past its last review.

We will get things wrong. When we do, we want the correction process to be visible, fast, and honest.

1. 1.Acknowledge. Reader-reported errors are confirmed within 48 hours.
2. 2.Fix the page. The inline error is corrected in place — we do not silently delete the original wording.
3. 3.Add a correction note. A timestamped "Correction (May 2026):" block is added at the top of the page describing what changed and why.
4. 4.Material errors are escalated. If a mistake materially misled readers, we say so on the home page and on social.