Inspired by Thorp and Kassouf (and G. Polya’s plausible reasoning), I’m experimenting with applying their approach to European BTC options on Deribit. Currently focused on daily call contracts (the most liquid), I plan to build six models (calls/puts × daily/weekly/monthly expirations). The goal is to explore arbitrage strategies, mainly short straddles, by spotting overpricing in these derivatives. Why Thorp-Kassouf? Practical, linear regression-based, and transferable to BTC options. Classical Black-Scholes assumptions may not hold for crypto, but this approach is simple yet powerful for prototyping. Data: Historical BTC data from 2017-01-01, 5-min granularity, stored in MongoDB; processed with Python libraries (NumPy, Pandas, SciPy, Statsmodels). Disclaimer: Not a Python expert; code evolved from a proof-of-concept. Can be optimized/refactored, focus is on modeling and insights for short-straddle strategies. Next steps: Explore ways to “learn” when a contract fits the model. GitHub: [https://github.com/dradicchi/kassouf-btc-options](https://github.com/dradicchi/kassouf-btc-options) Kassouf paper: [An Econometric Model for Option Price with Implications for Investors' Expectations and Audacity](https://www.jstor.org/stable/1910443) Feedback on modeling, reproducibility, or ideas for backtesting would be appreciated. |