Graded Trades

The raw per-trade grading feed: every setup replayed as a real trade — enter on its trigger, exit at its target or its stop, whichever prints first. Setups that never triggered aren't counted either way. No cherry-picking, no deleted calls.

Read this before you read the numbers. A win rate on its own is not evidence. For the geometry these setups use, a coin flip on a driftless stock hits its target about 60% of the time — because the stop is closer than the target. A 60% win rate here is the baseline, not an edge. The number that matters is expectancy in R, measured against that baseline, and it lives on the Track Record page along with the confidence interval and the pre-registration.

These figures were rebuilt on 31 Aug 2026. The simulator behind them used to fill every trade at its trigger price even when the stock gapped past it, and floored every loss at exactly −1R — across 80 losing trades it booked −1.000R eighty times, which is a floor, not a market. Both caps erred the same way. Re-grading all 287 closed trades with gap-adjusted fills and real losses moved the published expectancy from +0.21R to −0.07R, and the win rate from 64% to 48%. Every number on this page is the corrected one. We are showing you the worse number because it is the true one.