The Decision Quality Score is the average realised R of your resolved high-grade takes minus your resolved low-grade takes, so it needs at least one of each. Every metric on the panel states its own precondition rather than showing a value it has not earned — a named gap is the correct state until the data exists.
Decision Quality Score
It asks one question: when the engine said this was good and you took it, did you do better than when it said this was weak and you took it anyway?
It needs at least one resolved take in each family before it means anything, so until both exist the panel names the gap.
The rest of the panel
Each figure answers a distinct question about your own record.
- Capture rate — how much of the available move you actually took.
- Adherence share — followed takes against overrides.
- Expected calibration error — how far the confidence numbers were from what happened.
- Expectancy by grade family — whether higher grades actually paid better for you.
- Net R and week R — the totals, in R rather than in currency.
- R saved by passes — what declining the losers was worth.
- One behavioural insight, priced in R rather than described in adjectives.
Per-lens performance
Followed-versus-override expectancy is broken out per pack, and this publishes from your very first resolved trade — because it is your own record. The floors that govern platform-wide figures exist to stop a small sample being presented as a general claim; a statement about your own trading is not a general claim.
The weekly report
A rolling seven-day receipt: decisions, passes, R saved, net R, and your best and worst setup families. Email delivery depends on the deployment having a mail sender, and is named as conditional rather than implied.
Frequently asked
- Why does my Decision Quality Score show a gap instead of a number?
- It needs at least one resolved high-grade take and one resolved low-grade take. Until both exist there is nothing honest to compute, so the panel names the missing input rather than publishing a value built on half a comparison.
Updated Sep 1, 2026