The original hypothesis was NOT supported on this pilot sample. Pull-In → Trend winners did not cluster in EARLY_EXPANSION / EXPANSION_CONFIRMED / TREND_UP at entry — they entered in ROTATION. The biggest single Pull-In loss in the sample (-$216 IBIT 6/3) entered in EXPANSION_CONFIRMED. The state label alone does not separate winners from losers at v1.0 / slope=2.0 on this sample.
However — VCET numeric fields (slope, band_width, vcet_score, compression_score) do show systematic differences between winners and losers within the same slug. The signal is in the geometry, not the categorical label. See §4.
Verdict on next step: RVOL + gap (the existing Quality Gate) remains the primary actionable filter. VCET v1.0 at this sample size is too noisy to displace it. See §7 for the recommendation.
vwap-pull-in 127, vwap-pull-in -> trend 15, chase-extended 56).vcet_state, reason_code, vcet_score, vcet_grade, vwap_band_width_pct, vwap_slope_deg, price_distance_from_vwap_sd, expansion_score, range_compression_score.| Slug | Total | With VCET | NO_BAR | Snap rate |
|---|---|---|---|---|
vwap-pull-in leak | 127 | 35 | 92 | 27.6% |
vwap-pull-in -> trend edge | 15 | 9 | 6 | 60.0% |
chase-extended sentinel | 56 | 11 | 45 | 19.6% |
| TOTAL | 198 | 55 | 143 | 27.8% |
| State at entry | n | Total P&L | Avg | Win % |
|---|---|---|---|---|
| ROTATION | 22 | -$91.77 | -$4.17 | 22.7% |
| TREND_DOWN | 5 | -$38.87 | -$7.77 | 0.0% |
| TREND_UP | 4 | -$41.40 | -$10.35 | 0.0% |
| FAILED_EXPANSION_TENTATIVE | 2 | -$18.35 | -$9.18 | 0.0% |
| COMPRESSION | 1 | -$2.10 | -$2.10 | 0.0% |
| EXPANSION_CONFIRMED | 1 | -$216.15 | -$216.15 | 0.0% |
| State at entry | n | Total P&L | Avg | Win % |
|---|---|---|---|---|
| ROTATION | 5 | +$109.49 | +$21.90 | 100.0% |
| TREND_UP | 3 | +$10.69 | +$3.56 | 66.7% |
| TREND_DOWN | 1 | +$12.25 | +$12.25 | 100.0% |
5 of 9 PI → Trend winners entered in ROTATION, not in EXPANSION_CONFIRMED as the hypothesis predicted. The biggest PI → Trend winners (+$68 DAL 5/26, +$22 DAL 5/27) both entered in ROTATION_FLAT_BANDS.
Interpretation: the conversion to trend happens after entry, not at entry. VCET-at-entry captures the moment before the trend fires — that moment is rotation. The state evolves to TREND_UP in the 5-15 bars following entry. The right question may not be "what state at entry" but "what state at exit" or "what state transition occurred between entry and exit."
| State at entry | n | Total P&L | Avg | Win % |
|---|---|---|---|---|
| ROTATION | 11 | +$96.15 | +$8.74 | 54.5% |
| State at entry | W | L | Win % | W P&L | L P&L |
|---|---|---|---|---|---|
| ROTATION | 16 | 21 | 43.2% | +$504.74 | -$390.87 |
| TREND_UP | 2 | 5 | 28.6% | +$22.45 | -$53.16 |
| TREND_DOWN | 1 | 5 | 16.7% | +$12.25 | -$38.87 |
| FAILED_EXPANSION_TENTATIVE | 0 | 2 | 0.0% | +$0.00 | -$18.35 |
| COMPRESSION | 0 | 1 | 0.0% | +$0.00 | -$2.10 |
| EXPANSION_CONFIRMED | 0 | 1 | 0.0% | +$0.00 | -$216.15 |
ROTATION is the only state with net-positive P&L (+$114 across 37 trades). Every other state lost money on this sample. TREND states (TREND_UP / TREND_DOWN) were 17-29% WR — the opposite of the hypothesis.
The hypothesis predicted EXPANSION_CONFIRMED + TREND states would house winners. They did not. The state machine appears to be calling regime AFTER the move has matured (and after the entry was already taken), making "VCET state at entry" a lagging indicator on this sample.
| Slug | Out | n | vcet_score | band_width_pct | vwap_slope_deg | dist_from_vwap_sd | expansion_score | compression_score |
|---|---|---|---|---|---|---|---|---|
| vwap-pull-in | W | 5 | 60.4 | 1.18 | +3.95 | +0.65 | 31.0 | 26.0 |
| vwap-pull-in | L | 29 | 48.0 | 2.54 | -1.61 | +0.33 | 36.0 | 13.1 |
| vwap-pull-in -> trend | W | 8 | 54.0 | 3.83 | -1.48 | +0.15 | 26.9 | 7.5 |
| vwap-pull-in -> trend | L | 1 | 100.0 | 9.60 | +70.07 | +1.04 | 35.0 | 0.0 |
| chase-extended | W | 6 | 59.3 | 1.59 | -3.26 | +0.39 | 23.3 | 5.0 |
| chase-extended | L | 5 | 55.6 | 8.00 | -5.10 | -0.14 | 15.0 | 26.0 |
Plain PI: winners enter on rising VWAP slope (+3.95°) and narrow bands (1.18); losers enter on flat/falling slope (-1.61°) and wider bands (2.54). Score 60 vs 48. The slope sign and width predict outcome more cleanly than the state label.
Chase-extended: winners have narrow bands (1.59), losers have very wide bands (8.00) — a 5× ratio. Wide bands = volatile environment = chase doesn't work.
PI → Trend: the lone loser (LUNR -$11) had extreme slope (+70°) and wide bands (9.60) and a perfect score (100) — a forced trend signal that didn't follow through. The 8 winners had calmer numerics (slope -1.48, bw 3.83).
| Slug | reason_code | n |
|---|---|---|
vwap-pull-in | STICKY_DEFAULT | 16 |
| SESSION_INACTIVE | 12 | |
| TREND_DOWN_SLOPE_STRONG | 2 | |
| ROTATION_FLAT_BANDS | 2 | |
| COMPRESSION_HOLD | 1 | |
| TENT_LOST_RANGE_LEVEL | 1 | |
vwap-pull-in -> trend | STICKY_DEFAULT | 3 |
| TREND_UP_SLOPE_STRONG | 2 | |
| ROTATION_FLAT_BANDS | 2 | |
| SESSION_INACTIVE | 2 | |
chase-extended | SESSION_INACTIVE | 7 |
| STICKY_DEFAULT | 3 | |
| ROTATION_FLAT_BANDS | 1 |
| Entry time | Symbol | P&L | VCET state | reason_code | score |
|---|---|---|---|---|---|
| 2026-05-26 13:56 | DAL | +$68.25 | ROTATION | ROTATION_FLAT_BANDS | 90 |
| 2026-05-27 13:56 | DAL | +$22.50 | ROTATION | STICKY_DEFAULT | 52 |
| 2026-05-28 11:56 | BB | +$18.40 | TREND_UP | TREND_UP_SLOPE_STRONG | 88 |
| 2026-05-28 09:31 | JOBY | +$12.45 | ROTATION | SESSION_INACTIVE | 0 |
| 2026-06-03 10:38 | HL | +$12.25 | TREND_DOWN | STICKY_DEFAULT | 38 |
| Entry time | Symbol | P&L | VCET state | reason_code | score |
|---|---|---|---|---|---|
| 2026-06-03 09:41 | IBIT | -$216.15 | EXPANSION_CONFIRMED | RANGE_BREAK_SHORT | 94 |
| 2026-05-27 09:39 | AMD | -$59.30 | ROTATION | STICKY_DEFAULT | 0 |
| 2026-05-28 08:12 | META | -$47.73 | ROTATION | SESSION_INACTIVE | 0 |
| 2026-06-03 09:51 | NVDA | -$36.86 | ROTATION | STICKY_DEFAULT | 0 |
| 2026-06-02 09:48 | RGTI | -$33.58 | TREND_UP | STICKY_DEFAULT | 70 |
Entry: LONG IBIT at 09:41 ET, 6/3/2026. VCET state at entry: EXPANSION_CONFIRMED, reason RANGE_BREAK_SHORT, score 94. The state machine said "confirmed expansion, breaking the range to the downside" — correctly. The trade was taken LONG. The trade lost $216 because the short-side break followed through. VCET was right; the trade was wrong-side of what VCET said. One data point doesn't make a rule, but this single trade nearly accounts for the entire vwap-pull-in / EXPANSION_CONFIRMED loss bucket.
At v1.0 / slope=2.0 on this 55-trade pilot sample: NOT BY STATE LABEL. The hypothesis (PI → Trend winners in EARLY_EXPANSION / EXPANSION_CONFIRMED / TREND_UP, plain PI losers in ROTATION / FAILED_EXPANSION) did not hold. Winners and losers both cluster in ROTATION.
However, BY NUMERIC GEOMETRY: yes, partially. Plain PI winners have rising VWAP slope (+3.95° vs -1.61° for losers); chase-extended winners have narrow bands (1.59 vs 8.00 for losers); PI → Trend winners have moderate slope/bands, the lone loser was an extreme-slope false signal.
The Quality Gate (RVOL < 2.5 AND gap < 1.5%) likely captures most of what VCET geometry would. Until v1.1/CBO lands AND we have full historical bars, VCET v1.0 state at entry is not the right additional filter on top of the existing gate.
| Rule | Basis | Confidence |
|---|---|---|
V1. Plain Pull-In: prefer entries with VCET vwap_slope_deg > 0 and vcet_score > 55 | winners avg +3.95°/60.4 vs losers -1.61°/48.0 | ⚠ low — n=5 winners |
| V2. Plain Pull-In: avoid entries when state is TREND_UP or TREND_DOWN (paradoxical 0% WR) | 9 trades 0 wins; likely "trend already extended" condition | ⚠ low — n=9 but counter-intuitive |
V3. Chase / Extended: require vwap_band_width_pct < 3.0 | winner bw 1.59 vs loser bw 8.00 | ⚠ low-medium — n=11 |
V4. If VCET says EXPANSION_CONFIRMED with reason RANGE_BREAK_SHORT, do not take LONG Pull-In (counter-side break confirmed) | IBIT 6/3 -$216 specifically | ⚠ very low — n=1 |
| V5. Continue using RVOL+gap Quality Gate as primary | established earlier; VCET v1.0 doesn't displace it here | high — from larger Pull-In diagnostic |
main.vcet_bars infrastructure yet. v1.0 alone hasn't shown the discriminating power that would justify the build. Wait for v1.1 (with CBO) AND better historical bar source before investing in the table + backfill.