Quantitative research platform for building and testing systematic trading strategies with point-in-time backtests and walk-forward validation
We read the signals hiding inside markets
Stage
Startup
Growing
Founded
2025
2025
Based in
DE
US
Pricing model
Freemium
Free trial
Pricing
Starter €7.99/month, Practitioner €19.50/month (regularly €39), Research €34.50/month (regularly €69). All prices include VAT. Every new account gets 30 days of Practitioner free with no card required.
Free trial with full access to all signals; paid subscription required after trial.
Key features
Visual strategy builder with circuit diagram interface
Point-in-time backtesting over 20 years of history
Walk-forward validation and robustness testing
Indicator strategies with mean reversion and momentum
Stochastic methods including Kalman filter and Ornstein-Uhlenbeck models
Factor model and portfolio construction tools
Paper trading with daily marked prices
Python code export for backtests and strategies
Regime analysis and market regime detection
Real-time BTC 15-minute market analysis every 5 seconds
Clear YES or NO signals with confidence and edge metrics
85.5% live accuracy on first signals (30-day trailing average)
Public REST API and Server-Sent Events streaming
Trained on 15,000+ historical BTC markets
Ready-to-use prompts for coding agents (Claude, Cursor)
Replay feature showing past ten settled markets
What makes it different
Point-in-time data discipline prevents look-ahead bias and ensures strategies see only information available on each date
113 published research articles and studies with reproducible methodology and full audit trail
Circuit diagram interface makes strategy design accessible without requiring extensive coding
Full Python export enables strategies to be deployed outside the platform
User retains full control—signals are informational only, trades are placed by the user in their own Kalshi account
No custody of funds—Cotes never holds, accesses, or moves user money
Fast, objective signals updated every 5 seconds with transparent methodology