Every evaluation row we measured, side by side

Every measured row of every evaluation battery in the registry: 10,165 method rows and 323 grid variations across 68 tested accounts, with the pass rate, the time it takes and what it costs to get through — filtered and sorted however you like.

A row is one method variant at one stress arm, or one grid variation — the same row set that account’s own battery table prints, at the published slice (intraday touch walk). The Arm control starts on raw, the published basis; the degraded arms re-run the same book with halved PnL, winners capped at 1R or 1–2 ticks of slippage. ★ marks each account’s champion arm — the best pooled pass rate among its raw-arm method rows. It is a flag on a row, not a filter: “Champions only” collapses the table back to one row per account. None of these are live results.

Period re-reads the figures at one market regime: the value printed is that row’s own cohort in the frozen artifact, never a recombination of others. Entry price is structural and stays whole-window. A row whose artifact carries no cohort for a period shows an em dash there.

Compare: All eval accounts · All funded accounts · Full campaign chain · firms index

Expected spend to pass is only carried by one-time-fee evaluations; the monthly-billed accounts (MFFU, Topstep, TPT) have no per-attempt price to multiply, so those cells are em dashes and sort last. Untested accounts contribute a single dashed row — nothing is inferred from a neighbouring size.