I want a robust, trend-following system for the Bitcoin spot market that I can both back-test and run live on QuantConnect. The core of the logic should revolve around Simple Moving Averages alongside three on-chain metrics I rely on—Puell Multiple, MVRV Z-Score, and the Realized Price model. Scope • Work with at least six full years of BTC/USD data. • Target performance: minimum 60 % win rate, 50 % CAGR or better, and no more than 20 % historical drawdown. These figures must be demonstrated in the QuantConnect back-test report. • Code has to be production-ready in QuantConnect (Python preferred, C# acceptable) so I can flip directly from research to live signal generation without rewrites. Deliverables • Well-commented source file(s) for QuantConnect. • Back-test notebook or script showing parameter selection and walk-forward testing. • Final back-test report meeting the win-rate, CAGR, and drawdown thresholds. • Brief README that explains how to launch both the back-test and a live deployment. Acceptance The project is complete once I can clone the repository, run the back-test inside QuantConnect, reproduce the stated metrics, and schedule live signals with no further code edits.