The share of Pool A application tiers whose predicted expected lots fall into the same log2 bucket (0–1, 2–3, 4–7, 8–15…) as the realized value, measuring the model's accuracy at the lot-count magnitude.
Methodology
Round each tier's expected lots and bucket into log2 bins (0–1, 2–3, 4–7, 8–15…); hit rate = matched tiers ÷ comparable tiers, averaged across stocks. Predictions are made at the realized oversubscription.
Current Value
85.2%
Sample Size
214
Methodology
This metric comes from a walk-forward backtest on stocks listed in 2025 and later — the anchors (market regime, index, sponsor heat, first-day P&L) use only data visible before each stock listed; the regression coefficients are fitted on the full history, matching production prediction.
This page documents the definition, methodology and current value of the prediction metric "Pool A Expected-Lots Bucket Hit Rate". The value is aggregated from the prediction_history table and refreshed by the backfill script, for objectively assessing model accuracy — not investment advice.