RTH only — 9:30 to 16:00 ET. Every number on every page under /vpoc-data/ is filtered to U.S. Regular Trading Hours. Pre-market, overnight, and post-close events are parsed from the session log but excluded from all aggregates. Counts, durations, distances, time-of-day distributions — all RTH-only.
Opening-bell buffer — first 60 seconds excluded by default on detail pages. The developing VPOC flickers heavily in the first minute of RTH (9:30:00–9:30:59 ET) as volume rushes in before the profile has any shape. Empirically, ~28% of all RTH shifts in the dataset occur in that single 60-second window, then the rate drops 96% to baseline by 9:31. Monthly detail pages default to "First 1 min" excluded — matching the same buffer used by the companion BMBridge tool — so the headline numbers reflect what actually happened *after* the opening flicker settled. The filter is toggleable per page: click None to see raw counts, or extend to 5/10/15/30 minutes for a more conservative view. Per-day CSVs (linked from each monthly page) remain raw and unfiltered so anyone can verify the math.
Source cutover. April and May 2026 retain their VPOC Migration Pro attribution. New periods come from the AMT ForeSight v1 event journal. The pipeline rejects any trading day claimed by both sources, so the cutover cannot double-count a session.
A journal segment can close on a session boundary, restart, or settings change. Public averages use the trading day as the denominator; restart counts are never treated as market sessions.
A VPOC shift is logged the moment the addon promotes a new price level to "current VPOC" — meaning that price now has more accumulated session volume than any other. A Naked touched event fires when price returns to a prior VPOC that hadn't been revisited yet. A DVPOC changed event fires when the Developing VPOC swaps to a different candidate. (Note: DVPOC = Developing VPOC throughout this site — the live intraday challenger to the current VPOC. Some traders use the same acronym for the Daily VPOC, the prior day's settled point of control; that's a different concept. What's the difference?)
These pages aggregate across sessions to surface patterns — direction skew, time-of-day activity, frequency of Naked touches — that aren't visible inside any single session. Sample size grows as more sessions log.