See What's Possible with
Multi-Category Orderbook Data
High-frequency capture of prediction market microstructure across crypto, sports, economics, and weather. Full bid/ask depth, millisecond timestamps, and cross-category coverage — data that doesn't exist anywhere else.
What you can build
Six ways teams turn high-fidelity prediction market data into research, signals, and automated agents.
Market Making Simulation
Backtest market-making strategies across crypto, sports, and weather prediction markets with orderbook data at up to 10 Hz per crypto book
Learn moreSports Odds Analytics
Track live NBA, NFL, EPL, and FIFA World Cup game outcome probabilities. Analyze how odds shift as events approach
Learn moreWeather Derivatives
Monitor daily temperature prediction markets across 51 cities. Build weather-based trading models with historical orderbook depth
Learn moreEconomic Indicator Forecasting
Track FOMC decision probabilities and nonfarm payroll predictions. Capture market sentiment before data releases
Learn moreAI Agent Research
Feed live prediction market data to AI agents via MCP. Enable autonomous research across crypto, sports, and macro markets
Learn moreCross-Category Arbitrage
Detect mispricing across 114+ markets spanning 4 categories. Analyze spread patterns and liquidity dynamics
Learn moreWhy Data Fidelity Matters
Category-appropriate capture rates, sent to ClickHouse in batches every 2 seconds.
Event-Driven Capture vs Polling
Traditional data providers poll APIs at fixed intervals (1s, 5s, 30s), missing critical microstructure events between samples. Our pipeline reads WebSocket event streams and stores a snapshot when the book changes: up to 10 Hz for each crypto up/down order book (one every 100ms), up to 2 Hz for sports and up to 0.5 Hz for economic indicators. The writer sends the snapshots to ClickHouse in batches: every 2 seconds, or sooner when 800 rows are waiting. ClickHouse buffers each batch and makes it queryable up to about 30 seconds later.
AI Agents Meet Orderbook Data
Connect Claude, GPT, or custom agents to live market state over the Model Context Protocol.
What are the current NBA championship odds and NYC weather market?
{
"tool": "list_markets",
"params": {
"category": ["sports", "weather"],
"filter": "NBA championship, NYC temperature"
}
}I found 6 active NBA championship markets and 1 NYC daily temperature market. NBA Championship: The current frontrunner shows bid 0.32 / ask 0.34 with deep liquidity (18,500 contracts on the bid side). Three other teams are trading between 0.08-0.15. NYC Weather: The "above 72F" outcome for today is at mid price 0.61, with a tight spread of 0.02. Weather markets typically see increased volume closer to settlement time (6 PM ET).
{
"mcpServers": {
"polymarket-orderbook": {
"command": "node",
"args": ["mcp-server/index.js"],
"env": {
"API_KEY": "rm_your_key",
"API_URL": "https://api.resolvedmarkets.com"
}
}
}
}Three steps to live data
From signup to a live orderbook stream in your application — no boilerplate.
Get Your API Key
Sign up and generate an API key from the dashboard. Free tier available.
Navigate to /api-keys and click "Generate Key"
Make Your First Request
Query live markets with a single HTTP request.
curl -H "X-API-Key: rm_your_key" \ https://api.resolvedmarkets.com/v1/markets/live
Stream Live Data
Connect via WebSocket for the live orderbooks of one coin, sent every 2 seconds.
const ws = new WebSocket(
"wss://api.resolvedmarkets.com/ws/orderbook"
);
ws.onopen = () => ws.send(JSON.stringify(
{ type: "auth", apiKey: "rm_your_key" }));
ws.onmessage = (e) => {
const msg = JSON.parse(e.data);
if (msg.type === "auth") ws.send(JSON.stringify(
{ type: "subscribe", crypto: "BTC" }));
if (msg.type === "orderbook")
for (const m of msg.markets)
console.log(m.timeframe, m.up?.bestBid);
};Ready to Get Started?
Access prediction market orderbook data across every category at a resolution nobody else offers.