FX Engineer

We tested 75,864 SSI extreme events against a random baseline. The contrarian price path advantage is zero. A coin flip at every horizon, z-score magnitude, session, era. But SSI extremes do predict something. Just not direction. Full analysis (RP-004): fxeresearch.substack.com/p/price-path...

29 positioning events detected in the last 7 days. 25 of 29 produced a 15+ pip move in either direction.

AUDCAD retail positioning reached a statistical extreme (|z| = 2.12) — flagged 21:00 UTC, Oct 25. Resolved: confirmed.

Asset managers and leveraged funds sit on opposite sides of 15 out of 22 tracked markets. Largest spreads appear in 5Y and 10Y rates.

The model assigned 32% probability to elevated volatility on AUDCHF over the next 4 hours (move of 0.33% or more). Resolved: confirmed.

Resolution ledger. Everything that resolved in the window, both verdicts. Volatility events 6: 4 confirmed, 2 didn't materialize. Vol forecasts 12: 12 didn't materialize. Regime 2: 1 persisted, 1 transitioned. Full list in the ledger post. Running ledger on the card.

Across the last 6 similar episodes on AUDCHF, the typical path stayed within the p25–p75 band shown, out to 10 trading days. Checkpoints: d1 within band, d3 within band, d5 within band, d10 within band. Construction target ~50% within band.

EURGBP retail positioning reached a statistical extreme (|z| = 2.02) — flagged 21:00 UTC, Oct 22. Resolved: didn't materialize.

[COT Brief] Issue #22: cross-asset positioning across 22 markets as of 2026-10-20. Data tables only. Full analysis: https://fxeresearch.substack.com/p/weekly-cot-positioning-week-of-2026-86d #cot #positioning #weekly

CHFJPY sits at the 40th percentile. Retail traders are 14.9% long. This reading is more short than 60% of all data since January 2014.

The model assigned 39% probability to elevated volatility on AUDJPY over the next 4 hours (move of 0.39% or more). Resolved: didn't materialize.

16 positioning events detected in the last 7 days. 13 of 16 produced a 15+ pip move in either direction.

CADCHF retail positioning reached a statistical extreme (|z| = 2.27) — flagged 21:00 UTC, Oct 20. Resolved: didn't materialize.

GBPNZD sits at the 16th percentile. Retail traders are 25.4% long. This reading is more short than 84% of all data since January 2014.

The model assigned 39% probability to elevated volatility on AUDJPY over the next 4 hours (move of 0.39% or more). Resolved: didn't materialize.

Asset managers and leveraged funds sit on opposite sides of the 2-year Treasury futures market. The leveraged fund net position ranks in the 94th percentile.

The model assigned 27% probability to elevated volatility on AUDJPY over the next 4 hours (move of 0.39% or more). Resolved: didn't materialize.

Silver leveraged fund net positioning registers at the 6th percentile of the past 52 weeks while asset managers hold the opposite side of the trade.

The model assigned 37% probability to elevated volatility on AUDJPY over the next 4 hours (move of 0.39% or more). Resolved: didn't materialize.

Resolution ledger. Everything that resolved in the window, both verdicts. Volatility events 2: 1 confirmed, 1 didn't materialize. Vol forecasts 12: 1 confirmed, 11 didn't materialize. Full list in the ledger post. Running ledger on the card.

The model assigned 0% probability to elevated volatility on EURGBP over the next 4 hours (move of 0.23% or more). Resolved: didn't materialize.

The model assigned 24% probability to elevated volatility on AUDCHF over the next 8 hours (move of 0.47% or more). Resolved: didn't materialize.

[COT Brief] Issue #21: cross-asset positioning across 22 markets as of 2026-10-13. Data tables only. Full analysis: https://fxeresearch.substack.com/p/weekly-cot-positioning-week-of-2026-e10 #cot #positioning #weekly

AUDNZD sits at the 1st percentile. Retail traders are 23% long. This reading is more short than 99% of all data since January 2014.

The model assigned 50% probability to elevated volatility on NZDUSD over the next 4 hours (move of 0.37% or more). Resolved: didn't materialize.

EURJPY and USDJPY each flipped bias within the last three hours. Multiple simultaneous flips can reflect broad USD or risk sentiment shifts.

The model assigned 40% probability to elevated volatility on EURUSD over the next 4 hours (move of 0.26% or more). Resolved: confirmed.

MICRO-ES leveraged fund positioning hit a 52-week low with z-score of -3.18 while asset managers hold the opposite side.

The model assigned 29% probability to elevated volatility on USDJPY over the next 4 hours (move of 0.27% or more). Resolved: didn't materialize.

Leveraged funds just added 53,090 contracts to their short JPY position. Current net: -49,098 contracts at the 2nd percentile.

AUDCAD retail positioning reached a statistical extreme (|z| = 2.06) — flagged 21:00 UTC, Oct 11. Resolved: confirmed.

Leveraged-fund net positions in 5Y and 30Y Treasury futures both hit 52-week highs this week while asset managers remained opposed. https://fxengineer.com/intelligence

The model assigned 31% probability to elevated volatility on CADJPY over the next 4 hours (move of 0.36% or more). Resolved: didn't materialize.

Resolution ledger. Everything that resolved in the window, both verdicts. Volatility events 4: 3 confirmed, 1 didn't materialize. Vol forecasts 12: 12 didn't materialize. Full list in the ledger post. Running ledger on the card.

Across the last 11 similar episodes on USDCHF, the typical path stayed within the p25–p75 band shown, out to 10 trading days. Checkpoints: d1 within band, d3 outside band, d5 outside band, d10 within band. Construction target ~50% within band.