The Nightcap White-Cap

Built ~4:55 PM ET · 2026-06-30 · validates the 6/30 regular-session close (Massive /v3/snapshot, COALESCE(session_close,session_price)) anchor: system-reminder 2026-06-30 == Bash date (agree) · static — regenerate to refresh

01Session Scorecard — 2026-06-30

5 setups in the open backlog; 2 finalized this run (today's Early Bird Curd, same-day) and 3 weekly Sunday Sundae setups left interim (window active through Thu 7/2 NFP) — 1 FIRE / 1 VOID / 0 MIXED / 0 NO_EVIDENCE. The morning called “choppy digest-the-rip”; the tape delivered a second leg up — a narrow, cap/semis-led melt-up. QQQ closed 736.40 (+1.70%) near its 737.62 high; SMH +3.78% (655.89) led, NVDA cleared $200 (+2.63%), SPY printed a fresh high (748.02 intraday, closed 746.77 +0.78%). Breadth stayed narrow — equal-weight RSP -0.13% lagged cap-weight, 4 of 11 sectors green. The MS-SEMI long FIRED (QQQ never threatened 720.85/716); the conditional LRT-QQQ short VOIDED — QQQ blew clean through the 730 level rather than rejecting, so the short correctly never armed.

Setup	Outcome	Evidence	Δ-ATR	Actual
MS L QQQ,SMHmm-260630-MS-SEMI-L	FIRE	Momentum-continuation long in semis/megacap-growth. QQQ held 720.85 and 716 all session (low 723.91), closed 736.40 (+1.70%) at the highs; semis led — SMH +3.78% (655.89), NVDA +2.63% (200.09), AVGO +1.42%; XLK +2.76% top sector. No kill (IWM reclaimed 300, never broke 294.68). Process caveat: QQQ opened 724.18 above the 720.85 trigger and never retested it — the clean opening-range pullback-reclaim entry never set up (the setup warned “NOT a chase after +3%”), so close-as-exit R overstates a real fill. SMH’s +3.78% trips a 2%-tol stale-flag vs the Mon build ref — the day’s real move, not a data error.	0.88	QQQ 736.40
LRT S QQQ,SPYmm-260630-LRT-QQQ-S	VOID	Conditional tag-and-reject short at the 730 round / 738 range high. QQQ tagged 730 and broke straight through to 737.62 (just shy of 738), closed 736.40 well above — no rejection. Kills triggered: QQQ held >730, IWM reclaimed 300 (300.45), semis led the breakout (SMH +3.78%, high 659.74). Megacaps mostly green (AAPL +2.71, MSFT +1.21, NVDA +2.63) — no Lag-7 RS failure. Short correctly never armed — 3rd straight session on the right side of the LRT level call. A blind short at 730 ≈ −0.22R.	0.36	QQQ 736.40

Still open — interim (weekly Sunday Sundae, window through Thu 7/2 NFP; not finalized)

Setup	Interim	Status as of the 6/30 close	Δ-ATR	Actual
OMR L QQQsun-260628-OMR-QQQ-L	OPEN	Working. Oversold mean-reversion long, entry ref 716 — QQQ closed 736.40, ~+2.8% above the trigger; no kill (QQQ<702 not hit). The “post-Tue quarter-end” arm condition is now satisfied. NFP Thu (7/2) is the gate before finalize.	—	QQQ 736.40
UCR L SPYsun-260628-UCR-SPY-L	OPEN	Working. Undercut-and-reclaim long, entry ref 737 — SPY closed 746.77 above 737 and held >716.58; no kill (close<716.58 not hit). NFP Thu gate.	—	SPY 746.77
EXF S XLVsun-260628-EXF-XLV-S	OPEN	Working directionally (WATCH). Exhaustion-fade of stretched defensives — XLV -1.29% (158.66), defensives broadly red (XLP -1.54, XLU -1.48, XLRE -1.98) as tech led; no kill (no broad risk-off; jobs Thu).	—	XLV 158.66

LensBoth same-day setups resolved correctly on a narrow cap/semis-led melt-up: the momentum long fired, the rejection short stood aside as 720.85/730 broke clean rather than rejecting. The honest knock is on the morning regime framing (“digest-the-rip” under-called a second leg up), not on setup selection. The weekly bullish-tech / fade-defensives lean is paying — all three Sundae legs working, none killed — but Thursday’s NFP (1pm half-day) is the real test before any finalize.

02Calibration

0.2418 Rolling Brier (last 50) ~flat from 0.2399 (6/29) · 0.25 = random

44% Hit rate (rolling 50) 41.6% over all 77 eligible

77 Calibration-eligible 2 finalized this run · 0 NO_EVIDENCE · 3 weekly interim

Rolling Brier held essentially flat (0.2399 → 0.2418, last 50) — the MS-long FIRE (prob 0.55) and the clean LRT-short VOID (prob 0.42) both scored better than the 0.25 random line, while hit-rate ticked up to 44% as the FIRE added a hit. No data-feed gaps this run: both finalized setups scored from confirmed Massive closes; neither leg was gated on $S5FI/$S5TH/AAII (the breadth kills referenced IWM/RSP — entitled ETFs), so the breadth read used confirmed price action, never inference.

Pattern performance — Component 13 (calibration-eligible; n≥3)

Pattern · Dir	n	Hit%	Mean prob	Mean Brier	Evidence class
LRT · S	23	56%	0.36	0.338	lore_pending
MS · L	19	37%	0.51	0.254	lore_pending
SRB · L	7	71%	0.41	0.299	lore_pending
SBD · L	6	0%	0.37	0.138	primary_source
GFD · L	5	60%	0.39	0.257	practitioner_backtest
MS · S	4	75%	0.44	0.318	lore_pending
SBD · S	4	0%	0.28	0.079	primary_source
LRB · L	3	33%	0.43	0.216	lore_pending

Plus six n=1 singletons (FBR-S, GFU-S, SRT-S, VAB-L, VBR-L, VSR-L) — too thin to read. The two workhorses moved in opposite directions tonight: LRT-S eased 59% → 56% over 23 (today’s conditional short voided — the level broke clean rather than rejecting, a wrong-direction outcome, not the pattern’s usual right-direction FIRE), while MS-L lifted 33% → 37% over 19 (a pure megacap melt-up where its chronic broad-tape/breadth kill didn’t bite). SRB-L (71% / 7) remains the cleanest high-n edge.

Calibration trend — Component 14 (rolling Brier by validation date)

06-220.271

06-230.272

06-240.261

06-250.2435

06-260.2411

06-290.2399

06-300.2418

Mechanism leak · C15 9/30 right-direction / wrong-mechanism (rate 0.300) — down from 0.321, still led by LRT-S (4). Neither of tonight’s two added to the leak: the MS long fired clean on-mechanism, and the LRT short was a wrong-direction VOID (QQQ ripped through 730, not a right-direction/clipped-by-kill miss), so the denominator grew (28→30) while leaks held at 9.

LensThe model is holding its multi-week gains (Brier ~0.242, well below the 0.25 random line) rather than extending them — a flat night, not a giveback. Tonight reinforces the same structural read: the LRT short’s edge is fading failed reclaims, so when a level instead breaks cleanly (today’s 730) the conditional correctly voids rather than shorting strength; and the chronically-MIXED MS continuation long has its good days precisely when breadth narrowness stops mattering — a pure megacap melt-up. Overconfidence risk still sits in continuation baskets carried through event risk — and Thursday’s NFP is exactly that.

03Tape & Rate Backdrop

Index	Close	Day %	Note
SPY	746.77	+0.78%	fresh high — intraday 748.02; closed near it
QQQ	736.40	+1.70%	led; low 723.91 never tested the 720.85/716 shelf
IWM	300.45	+0.49%	reclaimed the 300 round; small-caps participated modestly
RSP (eq-wt)	212.77	-0.13%	equal-weight red while cap-weight green = narrow
DIA	522.39	+0.14%	cyclicals/Dow lagged the Nasdaq melt-up

VIX eased to ~16.4 est. (web) from the 18.41 build-time reference — the risk-on melt-up pressed vol lower; not confirming a top.
Rates: 10Y ~4.38% area est. (web/FRED build-ref 4.38); the long end gave back ground intraday — TLT -1.18% (86.42, confirmed Massive) — yields firm-to-up on the risk-on tape. 2s10s ~+0.28 (build-ref).
Driver: quarter-end Tuesday + a semis/AI momentum bid (memory tailwind; NVDA cleared $200) extended Monday’s risk-on rather than digesting it.

Realized regimeNarrow, cap/semis-led trend-up (melt-up continuation) — not the morning’s “choppy digest-the-rip.” SPY to fresh highs, QQQ +1.70%, equal-weight red: a thin-breadth advance powered by the megacap-growth/semis complex. (Realized read, not a forward call.)

04Cross-Asset

Asset	Proxy	Close	Day %	Read vs equities
US Dollar	UUP	28.41	+0.14%	firm/flat — no tailwind, no drag
Crude	USO	106.44	-0.60%	soft — no energy bid (XLE red)
Gold	GLD	368.38	-0.05%	flat; GDX -0.30% — no safety bid
Long bonds	TLT	86.42	-1.18%	down (yields up) — confirms risk-on
Credit (HY)	HYG	79.97	-0.05%	calm — no credit stress under the rip
Bitcoin	IBIT	33.29	-2.60%	diverged — crypto sat out the equity melt-up

LensCross-asset broadly confirmed the equity risk-on — long bonds down (yields up), HY credit calm, gold/dollar quiet. The one notable divergence was crypto (IBIT -2.60%), which failed to join the tape; a small caution flag under an otherwise clean risk-on print, but not equity-confirming or -refuting on its own.

05Sector & Breadth (realized)

XLKTech+2.76

XLIIndus+1.35

XLBMaterials+0.34

XLYDiscr+0.14

XLFFins-0.20

XLCComm-0.69

XLEEnergy-0.88

XLVHealth-1.29

XLUUtils-1.48

XLPStaples-1.54

XLREREITs-1.98

4 green / 7 red — a narrow, top-heavy tape. Tech (XLK +2.76) carried the index; the green list thins fast after it (XLI, then barely-green XLB/XLY).
Defensives sold, again: XLRE -1.98, XLP -1.54, XLU -1.48, XLV -1.29 — the rotation out of last week’s defensive bid continued.
Equal-weight vs cap-weight: RSP -0.13% < SPY +0.78% < QQQ +1.70% — classic narrowing, the advance concentrated in the megacaps. IWM +0.49% a modest small-cap offset.
Breadth internals ($S5FI/$S5TH/$ADRN/$TICK/$TRIN) refresh-required — Massive 403s index symbols and BarChart isn’t reachable headless; the breadth read leans on the sector strip + RSP/SPY + IWM proxy (no leg was gated on internals this run).

LensA textbook narrow melt-up: the index made a new high on 4-of-11 green sectors and a red equal-weight. Powerful while the megacap/semis bid holds, but thin participation is the standing caveat into Thursday’s jobs print.

06Single-Name Movers

Illustrative of the day’s leadership only — never trade plans. From confirmed Massive closes (close-bell newsletters had not landed at the ~4:55 PM build).

Name	Close	Day %	Note
SMH (semis)	655.89	+3.78%	the engine — memory/AI complex led the tape
AAPL	289.36	+2.71%	megacap strength, not just semis
NVDA	200.09	+2.63%	cleared $200
TSLA	420.60	+2.13%	high-beta growth bid
AVGO	377.75	+1.42%	semis breadth
MSFT	373.02	+1.21%	steady megacap
GOOGL	357.37	+1.05%	participated
META	563.29	+0.12%	laggard among the Mag-7
AMZN	238.34	-0.75%	the only Mag-7 name red

LensLeadership was semis-first, megacap-broad — SMH +3.78% with AAPL/NVDA/MSFT/GOOGL all green undercuts the morning “Lag-7 losing RS” thesis that underpinned the rejection short; only AMZN (-0.75) and a flat META dissented. That breadth-within-megacap is exactly why the LRT-S short correctly stood aside.

07Morning Lens vs Reality

The Early Bird Curd framed today as a “choppy digest-the-rip, two-sided” session after Monday’s +2.48% QQQ melt-up — expecting digestion and high dispersion. Reality was a second leg up: QQQ +1.70%, SPY to fresh highs, semis +3.78% — continuation, not consolidation. The regime framing was too cautious; that’s the honest miss.

What the lens got right still drove the P&L read: (1) its conviction that “semis are the only complex with genuine fundamental wind” was vindicated — SMH led outright; (2) the “Lag-7 / defensives unwinding” structure held (XLV/XLP/XLU/XLRE all red); and (3) both emitted setups were on the correct side — the MS-SEMI momentum long fired, the LRT-QQQ rejection short correctly stood aside as 730 broke clean. Lens caution = the miss; leadership read + setup selection = the hits. The one carried risk — a thin-breadth continuation into Thursday’s NFP — is now the week’s open question.

LensLoop closed: a too-cautious regime call wrapped around two correctly-directional setups. The week’s bullish-tech tilt is intact and paying; the unresolved tension is whether a narrow, megacap-only advance can survive a jobs print — the standing overconfidence-through-event-risk flag from §02.

08Execution Debrief — realized execution on closed setups, not a forward trade plan

Setup	Profile	Entry	Stop	Target	Realized R
MS · L — mm-260630-MS-SEMI-L	swing	720.85	691.48	779.59	+0.53R
LRT · S — mm-260630-LRT-QQQ-S	swing	730.00	759.37	671.26	−0.22R

Realized R = (close − entry) / (entry − stop), self-signed, exit = the 6/30 close (QQQ 736.40). Aggregate (Component 16, all closed setups carrying exec): n=31 · mean realized R +0.68 · win-rate 67.7% · stop-rate 9.7%.

LensHonest read on the long: the +0.53R is a close-as-exit artifact — QQQ gapped above the 720.85 trigger and never pulled back, so a disciplined limit entry would have sat unfilled all day; the swing target (779.59) is a multi-week objective, not a day-1 fill. The short’s −0.22R is the cost a blind fade at 730 would have eaten — the conditional’s tag-and-confirm discipline avoided it entirely. Realized execution on closed setups — not a forward trade plan.