// Calibration report — estimate-vs-actual evidence behind the cones function CalibrationScreen({ onBack }) { const DS = window.CommiTeaDesignSystem_20e63b; const { Card, Badge, Icon } = DS; const c = window.CT_DATA.calibration; // scatter chart geometry const W = 420, H = 300, pad = { l: 36, r: 16, t: 14, b: 30 }; const maxD = 9; const X = (d) => pad.l + (d / maxD) * (W - pad.l - pad.r); const Y = (d) => H - pad.b - (d / maxD) * (H - pad.t - pad.b); const BiasBar = ({ bias }) => { if (bias == null) return n too small; return (
{c.n} closed issues with estimates · evidence, not opinion
estimated (x) vs actual days (y) · actuals inferred from git events, never tracked
You are not bad at estimating; you are optimistic in a very stable way. Stable, I can work with.