Every important decision starts with a probability. Bellwether builds infrastructure that turns public markets, AI models, and collective intelligence into signals you can trust.
A research platform from Stanford GSB and the Hoover Institution, built to make prediction markets a public good.
GET /v2/markets/BWR-DEM-WIN-PRES-2028 { "bellwether_price": 0.42, "cost_to_move_5c": "$47,200", // robust "platforms": { "polymarket": 0.43, "kalshi": 0.41 } }
Markets across platforms can't be compared, most prices are trivially manipulable, and there's no standard way to identify what event a market is actually about.
The same question appears on Polymarket and Kalshi with different wording, different resolution criteria, and no shared identifier. If two markets resolve under different conditions, their probabilities are fundamentally different — and comparing them is meaningless.
98.7% of political markets lack the liquidity to resist manipulation. Most can be moved 5¢ for less than $1,000. A price that can be bought isn't a consensus — it's a vulnerability.
A market's price is just its last trade. That could be informed money, a fat finger, or someone trying to move a number before a screenshot goes viral. Without volume-weighting over time, you're citing an anecdote, not a signal.
One endpoint returns a canonical ticker, a manipulation-resistant price, and cross-platform context for any political prediction market.
Every market includes the dollar cost to move its price by 5¢. Instantly know whether a prediction reflects deep conviction or thin liquidity.
Our matching engine maps identical events across Polymarket and Kalshi into a single canonical ID. Same event, one price.
6-hour VWAP with automatic fallbacks to 12h and 24h windows smooths out noise and manipulation attempts. The price that reflects sustained belief, not the last trade.
Every market rated Compliant, Partly Compliant, or Non-Compliant based on whether its resolution criteria match our Market Blueprints. Know instantly whether platforms are resolving the same question the same way.
Live prices updated continuously with 30-second cache. Current order book depth, VWAP, and compliance scores for every tracked market.
{
"ticker": "BWR-DEM-WIN-PRES-2028",
"question": "Will the Democratic nominee win the 2028 presidential election?",
"category": "us_presidential",
"bellwether_price": 0.42,
"vwap_window": "6h",
"platforms": {
"polymarket": {
"price": 0.43,
"volume_24h": 1_640_000
},
"kalshi": {
"price": 0.41,
"volume_24h": 1_200_000
}
},
"compliance": {
"cost_to_move_5c": 47200,
"rating": "compliant"
}
}
The Tape API is where we started. These are the problems we're solving next.
Canonical tickers, manipulation-resistant VWAP pricing, and compliance labels for every political prediction market across platforms.
Open specifications for prediction market contracts. Event definitions, resolution criteria, and settlement rules that any platform can list — so critical questions get reliable markets.
Automated systems that improve price accuracy in thin markets where manual liquidity provision is impractical.
Structured signal extraction from prediction market activity, delivered as a feed for analytical systems and research pipelines.
Every metric in the API is backed by research from the Hall Lab at Stanford GSB and the Hoover Institution.
Volume-weighted averaging and manipulation resistance scoring for prediction market prices.
Read →A methodological note on truncation timing — recent applied work increasingly relies on resolution-relative prices, often without explicit discussion of the implications.
Read →Examining the rapid growth of political prediction markets, their governance challenges, and what they mean for how we understand political information.
Read →The Tape API is live. If you're a newsroom, researcher, or builder working with prediction market data, we'd like to hear from you.