Execution-process measurement · not prediction
Execution Stability measures whether a trader's execution process stays consistent against their own baseline. The same certified measurement, two ways: on your historical data, run yourself in the browser, or as a guided live pilot we set up with you. A measurement, not a prediction. You set the tolerance; you decide.
If you could see that a trader was repeatedly leaving their own execution baseline without recovering, would that change how you assess them?
How decision pace deviates from the trader's baseline.
How sizing deviates from the trader's baseline.
Run your trade history in your browser.
A frozen baseline is built from the trader's first 200 trades.
The next session is classified against that baseline.
Needs ~350 trades: 200 build the baseline, the next 150 are the analysed session.
Input: a CSV with one row per trade, columns timestamp, pnl, position_value. Common export names (time/date, profit, volume/lots/size) are mapped automatically.
We set it up with you, no software for you to manage.
Sessions are monitored as they happen, against the trader's own frozen baseline.
Same classification, continuous instead of after the fact.
The same certified engine and the same measurement, set up as a guided pilot. Still a measurement, not a prediction.
Scalping: a second-resolution historical upload loses sub-second ordering when trades share a second, so those sessions are flagged rather than measured. We don't invent timestamps.
NOMINAL: recovery intact · ELEVATED: recovery weakened · BREACH: deviation persists without recovery.
Execution Stability measures execution-process consistency relative to a trader's own baseline. You decide what that means.
Same input, same output. Auditable and reproducible by your own team.
Your file is read inside your own browser. Never uploaded, never stored, nothing shared. No data-sharing agreement, nothing leaving your control.
A trader was evaluated against a frozen baseline built from the first 200 trades. Across multiple independent sessions, the execution process repeatedly classified as BREACH despite unchanged baseline, unchanged thresholds and unchanged engine. The case illustrates persistent loss of execution-process self-correction under a deterministic evaluation protocol.
Read Case Study 007 →For traders
See whether your own execution process stays consistent with your baseline, session by session, or whether it drifts without recovering. It measures whether your execution process remained consistent with your own baseline throughout the session. A private mirror for your own execution, run entirely in your browser.
Run it in your browser →For allocators & desks
An objective, auditable measure of execution-process consistency. Open to observational pilots on your own anonymised data.