VWAP Pull-In — Actionable Playbook

Live decision card · distilled from the 7 VWAP Pull-In research pages into a tomorrow-morning workflow · Machine spec: vwap_pull_in_actionable_gate.json · Built 2026-06-09 · Working framework — forward-test before locking.
The thesis & the one decision A. Say this out loud (60-sec script) B. Green / Yellow / Red gate C. Live decision table D. 5-question pre-trade checklist E. Which of the 4 is it? F. Position sizing G. Post-entry management H. Example trade cards I. What James needs to improve → Where this lives in the trading system → Execution scorecard fields to add → Scorecard grade logic → VCET / live classifier output → Validation needed → Backtest before promotion Appendix: the data behind this

The thesis & the one decision

Plain VWAP Pull-In is only for calm, controlled tape. Hot tape is either Pull-In → Trend with confirmation, or it’s no trade. The one decision: “Do I have a calm VWAP pull-in, or am I chasing hot tape?”
Proven direction is in the data (with the sample sizes shown in the appendix)
Working a reasonable cut-point, not yet validated per-trade — forward-test it

Honest framing up front: the direction is solid — calm tape wins, hot tape loses, and how price behaves in the first few minutes after entry separates winners from losers. The exact numbers (RVOL 1.5 vs 2.5, etc.) are working rules pulled from group averages on a small winner sample (15 Pull-In→Trend trades). Use them as a discipline, not as gospel.

ASay this out loud before entry

The 60-second script. If you can’t answer all five fast, the trade isn’t ready — pass.

  1. Is this calm tape or hot tape?
  2. Am I taking a pull-in, a pull-in-to-trend, or am I chasing?
  3. What confirms the trade after entry?
  4. Where am I wrong?
  5. What size does this deserve?

BGreen / Yellow / Red gate working thresholds

🟢 GREEN
Normal pull-in allowed
  • RVOL 1.0–1.5
  • Gap < 0.5%
  • Controlled rotation around VWAP
  • Not emotionally extended
  • MTF / context supportive*
Take normal planned size. Manage as rotation unless expansion confirms.
🟡 YELLOW
Half size / confirmation required
  • RVOL 1.5–2.5 OR
  • Gap 0.5%–1.5%
  • Some heat but clean structure
  • Trade only if VWAP slope + higher lows + band expansion support it
Half size. State the trend-conversion thesis out loud first, or pass.
🔴 RED
Pass or reclassify
  • RVOL > 2.5 OR gap > 1.5%
  • First-hour emotional tape (< 10:00 ET)
  • Price extended far from VWAP / > +2 SD
  • Damage symbol: IBIT, AMD, ETHU, RGTI, CVX
Do NOT tag as plain Pull-In. Reclassify or pass.

*MTF/MA stack is a macro context filter, not a per-trade discriminator — in the data it separates the two slugs but not winners from losers within plain pull-ins. Use it to read regime, not to greenlight a trade. RVOL is the primary filter, but it’s missing on many historical rows; if RVOL is unavailable, lean on gap% + your visual tape read.

⚠ First-hour rule

09:00–10:00 ET held 91 of 127 plain pull-ins and −$401 of P&L proven. The hour isn’t the enemy — selection inside it is. In that window, demand RVOL < 2 AND gap < 1% or pass.

CLive decision table

Condition🟢 Green🟡 Yellow🔴 Red
RVOL1.0–1.51.5–2.5> 2.5
Gap %< 0.5%0.5–1.5%> 1.5%
Time of dayAfter 10:00, or calm openOpen hour but controlled09–10 ET hot / emotional
VWAP slopeFlat-to-rising (with dir)Rising, needs to holdAgainst you, or vertical
Band widthNormal / containedStarting to expandAlready blown out (> +2 SD)
Price locationAt / near VWAP pullback+1 SD, still structuredExtended far from VWAP
MTF contextSupportiveMixedConflicted / weak
ActionTake, normal sizeHalf size + confirmPass / reclassify

D5-question pre-trade checklist

  1. RVOL? — 1.0–1.5 ideal; > 2.5 is a stop.
  2. Gap %? — < 0.5% ideal; > 1.5% is a stop. (Calm/flat or even gap-down is where this works.)
  3. Time of day? — first hour + hot RVOL = double-strict or pass.
  4. VWAP / band structure? — controlled pullback into VWAP (OK) vs hot expansion already underway (manage for trend or pass).
  5. Management plan? — if it fails to confirm, where exactly do I exit? (VWAP / lower band / time stop — pick it before entry.)

Can’t answer all five quickly? Pass. The setup may not exist on this name today.

EWhich of the 4 is it?

1 · Plain VWAP Pull-In
  • Calm tape, controlled pullback
  • Scalp / rotation target (opposite band, +2 SD)
  • Do NOT assume trend
  • Stop = VWAP / reaction-point swing
2 · Pull-In → Trend
  • Starts as a pull-in; earns the hold only if confirmed
  • Confirm: higher highs/lows, VWAP slope rising, band width expanding, price accepting above +2 SD, first 5–10 min expands
  • Outcome label, not a guarantee — confirm with behavior
3 · Chase / Extended
  • Already too extended, FOMO-driven
  • Hot RVOL / gap / first-hour tape
  • Not an edge (net flat, one outlier carried it)
  • Reclassify — do NOT log as a Pull-In
4 · Pass / No Trade
  • No clear edge, no clear plan
  • Failed the gate
  • Can’t honestly name the setup
  • “I just want in” = a chase. Pass.

⚠ The trap to avoid

“Pull-In → Trend” is the strongest outcome bucket in the data (+$367, 93% win) proven — but when the mechanism was actually reviewed, only 1 of 6 classifiable trades was a true trend conversion; most were rotation scalps, and 64% of the bucket’s dollars are too old to verify proven. Translation: don’t hold a plain pull-in like a trend trade just because it might earn the label. Make it prove the trend after entry.

FPosition sizing working convention

GateSizeGradeDaily-Stop units*
🟢 GREENFull planned sizeA / A+up to 1 full unit
🟡 YELLOWHalf size, confirmation-onlyB≤ ½ unit
🔴 REDNo trade (or tiny review-only)C / no-grade0 (or ¼ unit if studying tape)

*1 unit = your pre-defined per-trade max risk (a fraction of your Daily Stop at /daily-stop/). The grade→size mapping is a working convention — confirm your own unit size. The point is simple: full size only where the edge is (green), reduced where it’s muddy (yellow), nothing where it leaks (red).

GPost-entry management

“Cut fades early; trail expansions only after they persist.”

Once you’re in, the tape tells you which trade you actually have. Manage to behavior, not to the label.

~3 minutes — the fade check proven primary
Is it fading against me with no band expansion?
If yes → reduce or exit. A plain pull-in that’s fading at 3 min wins only ~7%. Fades win ~25% at every window — cutting them early is the highest-confidence rule here.
~5 minutes — classify
Rotation, expansion, or wrong-side?
Flat band + flat slope → rotation scalp: target the opposite band, take the rotation, don’t extend stops.
Band expands + slope accelerates in direction → possible Pull-In → Trend.
State/slope flips against you → wrong-side: exit or reduce now (the P&L is luck, not signal).
~10–15 minutes — the continuation check
Does the expansion PERSIST?
Only keep holding for trend if expansion is still there — don’t blind-hold (only ~42% of early expansion survives to 30 min). If it persists, trail the runner. If it decays, bank it as a scalp. If price fails VWAP or loses the lower band, exit/reduce.

HExample trade cards

Illustrative — how the framework reads at the click. (MSFT and IBIT mirror real history in the data; AAPL/NVDA are hypothetical.)

MSFTGREEN
Reason: RVOL 1.3, gap +0.3%, controlled pullback into VWAP, VWAP flat-to-rising, after 10:00.
Decision: Normal VWAP Pull-In allowed, full size. Manage as rotation unless the first 5–15 min confirms expansion. (MSFT was +$133 / 4 trades / 100% in the sample — calm-tape names like this are where it works.)
NVDAYELLOW
Reason: RVOL 2.0, gap +1.0%, strong structure but some heat.
Decision: Half size only. Needs higher lows + band expansion to become Pull-In → Trend. State the thesis out loud before entry, or pass.
AAPLRED
Reason: RVOL 3.1, gap +2.2%, first-hour hot tape.
Decision: Do not take as plain Pull-In. Reclassify as Chase/Extended or wait for genuine trend confirmation. Most likely: pass.
IBITRED · damage symbol
Reason: On the damage watch — IBIT was −$185 across 10 pull-ins (30% win), 51% of the entire plain-PI loss. Levered crypto-proxy tape.
Decision: Default to RED. Only consider if RVOL < 1.5 AND gap < 0.5% AND you can name a clean conversion thesis. Otherwise pass.
Calm name, fails to confirmGREEN entry → manage down
Reason: Entered green (RVOL 1.2, gap +0.2%). At 5 min: band flat, slope ~0, no expansion.
Decision: It’s a rotation scalp, not a trend. Target the opposite band and take the rotation — do NOT hold it hoping for a breakout that the tape isn’t offering.

IWhat James needs to improve

Part 2 — System integration
How this card stops being a standalone page and becomes a live component of the Strategy Library, Scorecards, and VCET. Design/plan only — no production code changes here.

Where this lives in the trading system

This playbook is not meant to sit alone in /system/. It is the bridge between the two halves of the system: the Strategy Library says what you’re allowed to trade; the Scorecards grade how you executed it. This card supplies the gate, the classification, and the grade logic that connect them.

LayerIts jobWhere it livesWhat this playbook feeds it
Strategy LibraryThe WHAT — source of truth for the setups you trade/strategies/ (data: /system/strategies.json; display mirrors _scientific_overlay.js)Becomes the official entries for the 4 slugs below
Execution ScorecardsThe HOW — grades execution quality per trade/scorecards/ (classifier _scientific_overlay.js#classifyEntry)gate_color, setup_classification, and the grade logic
VCET / live classifierThe WHEN — real-time gate + label at the ticketVCET engine (v1.1 + CBO)reads the JSON gate; supplies post-entry state/band fields
AMS system hubThe umbrella architecturesmb.a1prospect.com/ams/system/registers this as a reusable component

It should become the official Strategy Library entry for:

Draft slugLive DB slugWhat it actually is
vwap_pull_invwap-pull-inTradeable setup — calm-tape pull-in
vwap_pull_in_to_trendvwap-pull-in -> trendTradeable, but outcome-based — earns the hold by confirmation
chase_extendedchase-extendedWarning overlay, not a setup (off-book sentinel, not an edge)
pass_no_trade—Decision state — a logged no-trade, not a strategy

Draft object: vwap_pull_in_strategy_library_draft.json — full library metadata for all 4 (setup_type, market_context, required_conditions, disqualifiers, triggers, confirmation, risk, management, invalidation, scorecard_fields, reason_codes, validation_status), aligned to the real strategies.json schema. It is a draft: the underscore-vs-hyphen slug form needs reconciling, and chase_extended / pass_no_trade belong on the warning/decision axis, not the setup axis, before anything is promoted.

⚠ Don’t leave it isolated in /system/

The goal is to turn this into structured strategy metadata that the Library, classifier, scorecards, and VCET alerts all read from one source — so the gate on this page is the same gate that fires live and the same gate the scorecard grades against. Promotion is gated on the backtest below.

Execution scorecard fields to add

For every VWAP Pull-In attempt (passes included), the scorecard should capture:

FieldValues
setup_classificationplain_vwap_pull_in / pull_in_to_trend / chase_extended / pass_no_trade
gate_colorgreen / yellow / red
rvol_at_entrynumeric
gap_pctnumeric
vwap_slope_staterising / flat / falling
band_width_statecontracting / flat / expanding / blown_out
price_location_vs_vwapsigned SD offset at entry
first_3min_resultholding / fading / expanding
first_5min_staterotation / early_expansion / wrong_side
first_10_15min_confirmationexpansion_persisted / expansion_decayed / never_expanded / failed
management_moderotation_scalp / trend_hold / failed_pull_in_exit / no_trade
rule_followedtrue / false
mistake_typechased_hot_tape / oversized_yellow / traded_red_gate / held_rotation_like_trend / ignored_3min_failure / no_clear_invalidation / good_execution
improvement_note_plain_englishfree text

Additive scorecard columns — design only. No gen_scorecard.py or _scientific_overlay.js change until approved.

Scorecard grade logic

Grade the process, not the P&L. A losing trade can grade A; a profitable trade can grade F.
Grades WELL — even at a loss
  • Gate was valid (green, or correctly-confirmed yellow)
  • Classification was correct for the tape
  • Size matched the gate (full / half / none)
  • Exit followed the checkpoint rules (cut the fade, trailed the persistent expansion, took the rotation)
Grades POORLY — even at a profit
  • It was a RED-gate trade
  • It was actually a chase but tagged a pull-in
  • Size was too large for a yellow/red condition
  • No invalidation was defined before entry
  • The 3–5 minute failure was ignored

Profitable-but-poorly-graded trades are the priority review target — they reward bad process and hide the leak. Each maps to a mistake_type so the scorecard can tally which error costs the most. (Same logic in vwap_pull_in_actionable_gate.json → scorecard_integration.grade_logic.)

VCET / live classifier output

VCET should give the one-glance answer at the ticket, not make you read this page live. The eventual live output is simple:

Example — GREEN
VWAP Pull-In Gate: GREEN
Setup Type: Plain VWAP Pull-In
Risk: Normal
Management Mode: Rotation scalp unless expansion confirms
Invalidation: VWAP fail or 3–5 min no-confirmation
Reason Codes: RVOL_OK, GAP_OK, VWAP_CONTROLLED
Example — RED
VWAP Pull-In Gate: RED
Setup Type: Chase / Extended
Risk: No trade
Management Mode: Wait for trend confirmation or pass
Invalidation: N/A — no trade
Reason Codes: RVOL_HOT, GAP_EXTENDED, FIRST_HOUR_EMOTION

The five fields behind the label

FieldExample value
Setup TypePlain Pull-In / Pull-In→Trend / Chase-Extended / Pass
Reason CodeCALM_PULLIN / HOT_NEEDS_CONFIRM / HOT_PASS / DAMAGE_SYMBOL / FIRST_HOUR_HOT
Suggested RiskFull / Half / None (grade A·B·C)
Management ModeRotation-scalp / Trend-trail-if-persists / Cut-on-fade
InvalidationVWAP loss / lower band loss / 3-min fade

Codes match vwap_pull_in_actionable_gate.json so VCET can read the gate directly. VCET (v1.1 + CBO) is also the eventual supplier of the post-entry state/slope/band fields that confirm Pull-In→Trend in real time — that’s the bridge between this card and the mechanism work.

KValidation needed

These are working rules. Here’s exactly what to test next so they become proven rules (or get fixed):

  1. Do the RVOL bands (1.0–1.5 / 1.5–2.5 / >2.5) actually separate winners from losers per-trade, not just on group averages? (Backfill RVOL where NULL first.)
  2. Does gap% add real separation beyond RVOL, or is it redundant with it?
  3. Does a 3-minute fade reliably predict a failed pull-in (the ~7% win-rate claim) on a larger forward sample?
  4. Does 10–15 minute persistent expansion confirm Pull-In → Trend — and how often does early expansion decay?
  5. Which symbols should be blocked vs merely size-reduced? Is the IBIT / levered-proxy pattern stable, and does CVX belong on the list? (AMD/ETHU/RGTI are tiny samples.)
  6. Should the thresholds differ by symbol liquidity tier, sector, or market regime (VCET state)?
  7. After 2 weeks of GREEN-only enforcement, re-run the diagnostic and measure the realized P&L delta vs the back-of-envelope +$500 family estimate.

Backtest before promotion to official Strategy Library

✓ First validation pass done (2026-06-09) → VWAP_PULL_IN_GATE_VALIDATION_AUDIT.html

Run read-only against the Liberator backfill. Verdict: PARTIALLY SUPPORTED — revise before promotion. Data-proven: |gap|>1.5% (the strongest filter), the 3-min fade-exit, 10–15 min expansion confirmation, and instrument type (megacap-tech works; crypto-proxy/small-cap fail). Not supported: the RVOL thresholds — bucketed, the calmest <1.0 RVOL bucket was the worst (−$320), so "hot tape = the leak" doesn't hold. Promote the proven parts; rework the RVOL axis. Still needs an out-of-sample pass.

▶ Forward-test capture is LIVE (2026-06-09) → VWAP_PULL_IN_FORWARD_TEST_TRACKER.html

The next step isn’t another backtest — it’s collecting 20–50 live attempts under the frozen rules. Use the tracker to log each pull-in (gate, classification, 3-min / 5-min / 10–15-min checkpoints, mistake tag). Promotion stays HOLD until the sample fills.

Before this becomes an official Strategy Library rule (promoted into strategies.json / the classifier), validate:

  1. Do RVOL buckets separate winners and losers?
  2. Do gap buckets separate winners and losers?
  3. Does the first 3-minute fade predict failure?
  4. Does 10–15 minute expansion confirm Pull-In → Trend?
  5. Are some symbols repeat offenders?
  6. Are thresholds different for ETFs, megacap tech, small caps, crypto proxies, or high-beta names?
  7. Does Green/Yellow/Red improve expectancy versus the old generic VWAP Pull-In label?
  8. Does the scorecard mistake-tagging explain the P&L leak?

Until these pass, every threshold here stays working and the library draft stays validation_status: working_rule_needs_backtest. (Section K is the same list framed as the immediate research queue; this is the promotion gate into the live system.)

Appendix — the data behind this

So this card stays honest. Source: main.trade_context through 2026-06-03 (n=310 classified). Everything below is from the linked research pages; nothing is invented.

The split that drives everything proven

BucketnNet P&LAvgWin%
Plain vwap-pull-in (leak)127−$364.47−$2.8735.4%
vwap-pull-in → trend (edge)15+$367.67+$24.5193.3%
Family net (both)142+$3.20—41.5%

The discriminator: calm vs hot proven (as group means)

GroupRVOLGap %
Plain PI losers (n=80)4.01+2.11%
Plain PI winners (n=45)1.38+0.35%
PI→Trend winners (n=14)1.14−0.88%

The exact gate cut-points are working — these are averages, not validated per-trade boundaries.

Where the loss concentrates proven

Why “Pull-In → Trend” needs the caveat proven, n tiny

Of 15 PI→Trend trades, only 6 were mechanically classifiable: 1 true conversion (+$12), 3 rotation scalps (+$96, the bulk of the dollars), 2 wrong-side escapes. Six more (+$235, 64% of the bucket) are older than the bar window and unverifiable. The bucket is real and profitable, but the name overstates “trend conversion” by roughly 5×. Hence: confirm after entry, don’t assume.

Post-entry behavior is the real management edge corroborated, n=320 backfill

First-window behaviorWin %Avg P&L
Early expansion71%+$485
Fade26%−$700
Follow-through > +1σ79%—
Follow-through < −1σ19%—

From the deeper Liberator full-window backfill (vwap-band-entry-surface research). This is why the management section leads with “cut fades early.”

Source pages

Read-only distillation · designed as a system component (Strategy Library ↔ Scorecards bridge) · draft metadata in vwap_pull_in_strategy_library_draft.json + vwap_pull_in_actionable_gate.json · no DB / classifier / strategies.json / _scientific_overlay.js / Scorecards changes · thresholds are working rules pending the backtest gate.