Backtest
Backtest
Overview
Use this page to understand what the backtest is measuring, how to read the headline metrics, and where to go next for deeper validation details.
Backtest section
Read the validation story before you trust the output
The backtest evaluates the default dynamic DCA model against uniform DCA across rolling one-year windows. This page tells you what to read first and how to interpret the headline metrics.
Key terms on this page follow the canonical Glossary, including SPD and Win Rate.
As of: 2026-03-08
Source artifact: backtest-metrics.snapshot.json
Generation command: python backtest.py
Start Here
- Read SPD Calculation to understand the core performance metric.
- Continue to Performance Results for aggregate outcomes.
- Review Model Score for the composite benchmark.
Snapshot Metrics
Then Read
- Read Validation Checks for the consistency safeguards behind the reported metrics.
- Read Assumptions and Limitations for the conditions under which the backtest should not be over-interpreted.
- Use the Glossary when metric terms need a canonical definition.
Backtest Architecture
The system precomputes model features, then runs SPD calculations across 2,554 rolling one-year windows between 2018-01-01 and 2026-03-08.