Stage A.1 - Pre-market Policy Recommendation

Run date 2026-06-07 · Stage A.1 §4 deliverable · Source: main.trade_context last 60 days · Anchor: Pull-In Research Hub

Recommendation: EXCLUDE pre-market trades from mechanism classification; label as insufficient_data_premkt

31 of 205 in-scope trades (15.1%) entered before 09:30 ET. These trades fail the v1.0-inherited session_active gate in the VCET engine and currently produce empty trajectory snaps. The simplest, least-risky Stage B path is to exclude them with a clear residual label: insufficient_data_premkt. Including pre-market in VWAP/SD computation has known data-quality and behavior issues (thin volume, wide spreads, gappy bars) and would require engine work that's out of scope for Stage A.

Pre-market net P&L is +$113.29 across 31 trades, but +$199.07 of that comes from a single trade (NVDA 2026-06-02 05:58 LONG). Excluding that outlier, pre-market trades net -$85.78 across 30 trades. There is no statistical case for including pre-market in the mechanism work today.

1. Pre-market trade quantification

31
pre-mkt trades (15.1% of 205)
+$113.29
net P&L (with NVDA 6/2 outlier)
−$85.78
net P&L (without NVDA 6/2)
04:24
earliest entry time (ET)

2. By time bucket

Time bucket (ET)nNet P&LNotes
< 09:30 (pre-market)31+$113.29+$199 from NVDA 6/2; without it -$85.78 over 30 trades
09:30 - 09:35 (open burst)27+$121.03Open hour zone
09:35 - 10:0065-$267.50Main morning leak
10:00 - 12:0059-$231.09Mid-morning leak
12:00 - 14:0019+$196.66Mid-day reset
14:00 - 16:006+$13.51Afternoon
after 16:003+$4.83Edge case; very few

3. By slug

SlugPre-mkt tradesEarliestLatest
vwap-pull-in1404:24:2409:19:59
chase-extended1705:58:0909:26:40
vwap-pull-in -> trend0——

Notably: no PI → Trend trades occur pre-market. The mechanism question (true conversion vs rotation scalp) for that bucket is unaffected by pre-market policy.

4. Engine behavior on pre-market entries

In the Stage A pilot, pre-market entries produced empty trajectory snaps. The cause is the session_active gate in compute_features():

typical = df.groupby("session")["volume"].transform("sum").shift(1).bfill()
df["session_active"] = df["cum_v"] >= (typical * cfg.min_vol_fraction)   # min_vol_fraction = 0.10

This gate requires cumulative session volume to exceed 10% of the prior session's TOTAL volume before any state classification (other than ROTATION with reason SESSION_INACTIVE) is allowed. Pre-market typically takes hours to accumulate 10% of an RTH session's volume, so the gate stays closed until well after most pre-market entries.

4.1 What gets emitted today for pre-market entries

vcet_state = "ROTATION"
reason_code = "SESSION_INACTIVE"
direction = ""
vcet_score = (low; computed from rotation formula with thin data)
band_width_pct, slope_deg, etc = computed but unreliable on thin volume

The engine emits something, but the mechanism plan's downstream classifier would have no signal to work with.

5. Options considered

Option A
Include pre-market in VWAP/SD computation
Option B - SELECTED
Exclude pre-market; label as insufficient_data_premkt
Option C
Compute separate ETH/RTH features

6. Impact on mechanism classification

QuestionUnder Option B (selected)
PI → Trend mechanism count15 trades (0 pre-mkt; no change)
Plain Pull-In mechanism count118 trades (14 pre-mkt excluded; 89% retained)
Chase-extended mechanism count41 trades (17 pre-mkt excluded; 71% retained)
Total mechanism-eligible population174 trades (15% reduction)
Impact on the 6-mechanism class studyNVDA 6/2 +$199 outlier excluded (cleaner statistics); BIRD 4/15 with gap=333.7% (likely corp-action artifact) and other pre-mkt noise excluded
What replaces the excluded tradesResidual label insufficient_data_premkt; trade still appears in the population summary but does not contribute to mechanism classification stats

7. Pre-market trade list (for reference)

DateSymbolTime (ET)SideP&LSlug
2026-05-29IBIT04:24:24LONG-$13.92vwap-pull-in
2026-06-02NVDA05:58:09LONG+$199.07chase-extended
2026-04-24IBIT07:32:47LONG-$65.50vwap-pull-in
2026-04-29IBIT07:36:31LONG+$1.08chase-extended
2026-05-07AAPL07:56:45LONG+$6.57vwap-pull-in
2026-05-05NVDA08:06:38LONG+$10.68vwap-pull-in
2026-05-28META08:12:06LONG-$47.73vwap-pull-in
2026-05-27NVDA08:21:49LONG-$100.51chase-extended
2026-05-27CRWV08:21:56LONG-$13.61chase-extended
2026-05-05WULF08:22:44LONG+$2.29chase-extended
2026-05-28NVDA08:29:38LONG+$7.70chase-extended
2026-05-13SOXL08:30:31SHORT+$57.18chase-extended
2026-05-13AMZN08:36:54SHORT-$9.11vwap-pull-in
2026-05-12NVDA08:37:15LONG+$35.87vwap-pull-in
2026-05-08CRWV08:39:04LONG-$79.53chase-extended
2026-04-13IBIT08:54:30LONG+$18.39chase-extended
2026-04-24SLV08:58:35LONG-$2.50vwap-pull-in
2026-06-02QQQ08:59:43SHORT+$10.42vwap-pull-in
2026-05-28ETHU09:01:46LONG-$5.00vwap-pull-in
2026-05-05COIN09:05:50LONG-$12.21chase-extended
2026-05-06QQQ09:06:47SHORT-$24.88vwap-pull-in
2026-05-14ETHU09:09:01LONG+$35.84chase-extended
2026-05-06ETHU09:09:10LONG-$96.01chase-extended
2026-05-06MU09:09:21SHORT-$1.43vwap-pull-in
2026-05-07MSFT09:17:11LONG+$83.95vwap-pull-in
2026-05-28GOOGL09:19:59LONG+$32.73vwap-pull-in
2026-05-05INTC09:20:25LONG+$27.61chase-extended
2026-05-04IBIT09:22:24LONG+$22.45chase-extended
2026-05-29RGTI09:25:30LONG-$23.20chase-extended
2026-05-14CRWV09:26:29LONG+$44.69chase-extended
2026-05-29AVGO09:26:40LONG+$11.91chase-extended

8. Future work (out of scope for Stage A/A.1)

9. What this report did NOT do

Links