Seasonal Spread Strategies. Long one instrument, short a correlated one simultaneously. Return = %move(LongLeg) − %move(ShortLeg). Spreads are often more reliable than outrights because macro shocks tend to hit both legs — the seasonal captures the relative move between them, which is more structurally driven. Methodology inspired by MRCI seasonal spread research.

Seasonal Window Scanner. Brute-force searches every possible entry/exit combination for a given instrument across all 12 months — testing 960 windows per direction using the same Level 1 backtest engine. Filters by minimum win rate and data years. Splits history into two decades to test consistency. Deduplicates overlapping windows. Ranked by consistency then win rate.

Level 1 — Raw Seasonal Backtest · Yahoo Finance data. Enter at the open on the seasonal entry date. Exit at the close on the exit date. No stop loss, no filtering — pure price history. Each year = one trade. Green = price moved in the seasonal direction. All data sourced from Yahoo Finance via the Cloudflare Worker KV cache. Results auto-computed on page load for all supported instruments.