CoinThesis
Backtest cryptocurrency strategies on 9 years of Binance data — with AI parser and Monte Carlo analysis.
About the app
CoinThesis is a mobile backtester for cryptocurrency strategies. Just like ThesisEdge for S&P 500, CoinThesis allows you to describe a crypto strategy in plain language — AI will translate it into parameters, and the engine will simulate it on historical Binance data from 2017–2026.
The app supports over 100 coins (BTC, ETH, BNB, SOL, and others), daily OHLCV resolution, technical signals (MACD, RSI, Bollinger Bands, ATR, OBV), a BTC-dominance gateway, position weighting by market cap, and a rebalancing algorithm. All calculations take place offline, on the device.
In version 0.2.0, advanced analytical tools are available: parameter sweep with heatmap visualization, walk-forward analysis (out-of-sample validation), Monte Carlo simulation with a distribution of possible outcomes, and technical indicators as entry/exit signals. The data bundle (~4.6 MB) is downloaded once and does not require a constant connection.
Features
- AI Parser — describe your crypto strategy in Polish or English
- 100+ coins with Binance OHLCV data 2017–2026 (daily)
- Offline Dart engine — backtest without network connection after downloading the bundle
- Technical signals: MACD, RSI, Bollinger Bands, ATR, Stochastic, ADX, OBV
- BTC-dominance gateway: automatic exit to stablecoin during bear markets
- Parameter Sweep — overview of parameter combinations with CAGR heatmap
- Walk-Forward Analysis — strategy validation on out-of-sample data
- Monte Carlo — simulation of 1000+ scenarios, distribution of P10/P50/P90 outcomes
- Position weighting by market cap and fixed weights
- Universe filtering by category (Layer-1, DeFi, Meme, etc.)
- Metrics: CAGR, Sharpe, Sortino, Calmar, Max Drawdown, Alpha, Profit Factor
- 8 interface languages: EN, PL, DE, FR, ES, ZH, JA, PT-BR