Weekly review · Week 28, 2026

Regime trajectoryrisk-on grind-UP all week with one midweek wobble

Mon 07-06 opened the new-high push (SPY 751.28, VIX 15.57). Wed 07-08 a shallow flush on the hawkish FOMC minutes (QQQ 700.91, VIX intraday 18.91) was fully reclaimed. Thu-Fri sharp reassertion to SPY new closing high 754.95, VIX crushed to 15.03 (low 14.96, lowest in the sequence). Net Mon->Fri: TRENDING-up, low-vol, risk-on -- a regime FLIP from W27's CHOP/ROTATION (ret10 now positive and rising vs +0.52% last week). uw regime: risk-on, momentum-rebound tail live against shorts.

Sector leadership

FULL REVERSAL of W27's defensive/value rotation. Leaders (07-02->07-10): Energy XLE +3.49%, Semis SMH +3.16%, Tech XLK +2.87%, Comm XLC +1.86%. Laggards: Materials XLB -2.15%, Health XLV -1.77%, Industrials XLI -1.08%, Staples XLP -1.02%, Utilities XLU -0.76%. Cyclical/growth leadership fully back; W27's defensive winners (Health/Staples/Utils) are this week's laggards -- which is exactly why the W27 defensive-basket watch names (ABBV -5.64%) would have lost.

Notable flow

Trailing call-OI build, normalized by avg_30_day_call_oi with the pinned gates (Common-Stock only, liquidity + persistence floors, deal/earnings/near-high/listing-age carve-outs). Naive top-of-list all EXCLUDE: TECH 2.69x = confirmed M&A target (Merck KGaA $11.3B, 06-26); RSI 1.72x near-52w-high = uptrend build; TEL/SHOO/SLM/ENSG/CGNX = earnings inside the 2-4wk hold window; CBRS = IPO 05-14 (<60td listed); UNM = $3.8B Fortitude Re deal announced 07-06. HONESTY CORRECTION: last week's diary flagged CG at 3.77x -- a persistence check shows 97% of CG's build landed in a SINGLE day (06-29->06-30) via two ~100k-contract OTM call blocks = a lumpy institutional hedge, NOT persistent crowd accretion; the naive 2wk-sum ranking overstated it. The ONE clean, persistent, beta-neutral, non-deal, non-earnings-in-window fade is MAT (see Layer-2). Mega-cap tech (TSLA/MSFT/META/NVDA/AMD/AVGO/MU) call-build persists = QQQ-beta trap. Crypto-beta risk-on again (BTDR/MSTR bid).

Catalysts this week

2026-07-08June FOMC meeting minutes (Warsh's first meeting)Committee unanimously held funds at 3.50-3.75%; minutes revealed a deep 9-8 hawkish/dovish split with a few officials weighing HIKES ('nine hawkish dots'; Warsh's deliberate silence noted). Reaction MUTED on the day (futures held slightly negative, 2Y ticked up; QQQ dipped to its weekly low 700.91, VIX spiked to 18.91 intraday) but the hawkish tone did NOT hold: risk-on reasserted Thu-Fri to SPY new highs, VIX crushed to 15.03. Futures now price funds toward ~4% by year-end despite the soft June NFP -- equities SHRUGGED IT OFF (the reversal-after-catalyst tell in Layer-3).
2026-07-14Q2 bank earnings + June CPI (forward, next week)PRE-REGISTERED forward catalyst, not this week: big-bank Q2 prints and June CPI land the week of 07-13/14/15. The vol-crush into VIX 15.03 sets up an event-risk asymmetry for W29 -- cheap protection into a data-heavy week.

Weekly candles

Each candle is normalised to its own weekly range — the source is cross-sectional at incompatible price scales, so a shared axis would misrepresent it. Real prices are in the tooltip and the table below.

Weekly candles — data table

Weekly candles: symbol, open, high, low, close, return
Symbol	Open	High	Low	Close	Return
SPY	748.74	755.42	739.51	754.95	+1.37%
QQQ	719.93	726.39	700.91	725.51	+1.81%
IWM	297.75	300.41	290.68	295.99	-0.53%

SPY1.37%

Open: 748.74
High: 755.42
Low: 739.51
Close: 754.95

clean uptrend continuation of W27's recovery; higher-high (755.42 > 751.31) AND higher-low (739.51 > 732.09); NEW closing high for the sequence, a hair below the 06-15 intraday high 756.68; big lower wick (lw 0.58) from the Wed flush that was fully reclaimed Thu-Fri; closed at 0.97 of range = maximal strength close; follow_through=True (extended the prior up week)

QQQ1.81%

Open: 719.93
High: 726.39
Low: 700.91
Close: 725.51

WEEKLY HAMMER / bullish reversal; sold Mon-Wed and UNDERCUT the prior week's low (700.91 < 705.17 = lower-low + lower-high, a bear-trap) then V-recovered Thu-Fri to close +1.81%, the BEST of the three; huge lower wick (lw 0.75); reclaimed mega-tech leadership after lagging in W27's defensive rotation; reversed W27's failed-rally doji

IWM-0.53%

Open: 297.75
High: 300.41
Low: 290.68
Close: 295.99

lower-high (300.41 < 302.72) AND lower-low (290.68 < 294.68), red week (-0.53%) while SPY/QQQ made gains; small-caps the clear laggard for a second week; closed mid-range (0.55) off the Wed low; the ONLY major to close red = rotation stayed OUT of small-caps

Weekly technicals pre-registered · 0 points · not sized

SPY new-high breakout (higher-high + higher-low)w_ret +1.37%, higher_high=True (755.42>751.31), higher_low=True (739.51>732.09), close_pos 0.97, NEW closing high 754.95 (intraday 755.42 still a hair below the 06-15 high 756.68), follow_through=Truewould signal: trend-continuation long IF it clears 756.68 and holds; a maximal-strength close at 0.97 of range with HH+HL is textbook breakout structure
QQQ weekly hammer / bullish-reversal bear-trapundercut prior-week low (700.91<705.17) then V-recovered to close +1.81% (best of the three); lower wick 0.75, close_pos 0.97, body_frac 0.22; reversed W27's failed-rally dojiwould signal: mega-tech leadership reclaimed / failed-breakdown long IF next week holds above the reclaim; the classic bear-trap-then-reclaim structure
IWM lower-high + lower-low laggard (2nd week)lower_high (300.41<302.72), lower_low (290.68<294.68), only major to close red (-0.53%), close_pos 0.55would signal: small-caps failing to confirm the new-high tape = narrow/large-cap-led advance; would signal caution on breadth IF it persists
reversal-after-catalyst (hawkish FOMC minutes shrugged off)hawkish 9-8-split June minutes (Wed) dipped QQQ to its low + spiked VIX to 18.91 intraday, then risk-on FULLY reclaimed it Thu-Fri to new highs, VIX crushed to 15.03would signal: 'buy-the-hawkish-dip' vol-crush regime-strength tell IF the pattern repeats; the tape overriding a hawkish macro print is a risk-on structural signal

Directional book1

MATMEDIUMShortlaneOI_FADE10dTHE ONE SCORED CALL. Clean OI_FADE short: persistent call-OI build every one of 10 days (accelerating into 07-09/10), normalized 0.90x avg_30_day_call_oi, max single-day share 0.40 (passes persistence). Beta-neutral (toys/consumer-disc, no cluster overlap), NOT mega-cap-beta, NOT an ETF/new-listing, NOT a deal, NOT near-52w-high (pct_52w_range 0.031 -- fresh 52w low 07-09). Fundamentals CONFIRM the short: Goldman cut to Sell 07-09 same day it hit a fresh low = crowd piling into bounce-bet calls on a battered name (the textbook crowded-embedded-leverage fade). Entry 07-10 close $13.33, liquid (~$65M/day $-vol). SIZE = starter/half-cap only, NOT full: shorting into a broad new-high VIX-15 tape is the standing headwind. EARNINGS 08-04 is ~18td out, inside an h21 hold -> run to h10 and FLATTEN before the print. Invalidation: call unwind, a real bottom/beat, or a regime-classifier strong-rebound thrust before entry.

Lane status2

laneMOM_LONGBasket watchLongregime fit0.55

regime-fit IMPROVED vs W27 (0.45->0.55): the tape is now TRENDING-up (new highs, ret10 positive/rising) which is the condition this tail-driven lane needs. Still NOT sized: cohort is dilute (480 names, 46% ETFs), Financials/Healthcare cluster-correlation risk, and the edge is tail-driven (median ~0, hit-base -9.3pp) so it is BASKET-ONLY, never per-name. Gate to half-cap only if ret10 sustains >+1% for >=2wk and the cohort tightens. V-rebound novelty may fade.

laneMOM_SHORTStood downShortregime fit0.10

STOOD DOWN. New closing highs + positive/rising ret10 + VIX crushed to 15.03 = the momentum-rebound tail that inverts near-52w-low shorts (Daniel-Moskowitz). Confirmed by last week's W26 book resolving 07-10 with ALL THREE MOM_SHORT shorts squeezed (FXI -2.42, CME -5.16, MSTR -11.42 excess). No shorts emitted.

Lane glossary (3)

OI_FADE: OI-flow fade — fades persistent multi-day net call-OI building; heavy call-OI build precedes underperformance. The most robust lane measured (short, h10).
MOM_LONG: Momentum, long leg — near-52w-high relative strength. Tail-driven; basket-only, never sized per-name (h10).
MOM_SHORT: Momentum, short leg — near-52w-low relative weakness. Regime-gated against momentum-crash/squeeze rebounds (h10).

THREE layers, never blended. Layer 1 = diary (0 signal). Layer 2 = validated momentum/OI-fade lanes (scored in excess). Layer 3 = weekly technicals (PRE-REGISTERED, 0 points, never sized). Truth-set parquet is stale at 2026-06-26 → all W28 bars + carry-forward built path-aware from the live Yahoo chart API; OI build from the fresh chain-oi-changes panel (through 07-10).

1. The Week in Pictures (DIARY — documentation, 0 signal)#

Sym	O	H	L	C	w_ret	close_pos	body	structure vs prior
SPY	748.74	755.42	739.51	754.95	+1.37%	0.97	0.39	new-high breakout; HH (755.42>751.31) + HL (739.51>732.09); closed at 0.97 of range; big lower wick from the Wed flush, fully reclaimed
QQQ	719.93	726.39	700.91	725.51	+1.81%	0.97	0.22	weekly hammer / bear-trap; undercut prior-wk low (700.91<705.17) then V-recovered to close best of the three; lw 0.75; mega-tech leadership reclaimed
IWM	297.75	300.41	290.68	295.99	−0.53%	0.55	0.18	lower-high + lower-low; only major to close red — small-caps the laggard a 2nd week

The story — a risk-on grind to new highs, and mega-tech took back the wheel. The leadership ordering flipped clean out of W27: last week was SPY > QQQ > IWM (defensive/value rotation); this week is QQQ (+1.81%) > SPY (+1.37%) > IWM (−0.53%) — growth back on top, small-caps still lagging. SPY printed a new closing high (754.95), a hair under the 06-15 high 756.68, with textbook HH+HL breakout structure. QQQ is the tell: it undercut the prior week's low intraday (700.91 < 705.17) then reversed hard to close +1.81% — a weekly hammer / bear-trap that reclaimed mega-tech leadership after it lagged in W27. IWM couldn't confirm (lower-high, lower-low, red).
Catalyst anchored — June FOMC minutes, Wed 2026-07-08 (Warsh's first meeting):
- Funds held 3.50–3.75% unanimously, but the minutes showed a deep 9-8 hawkish/dovish split — a few officials weighed HIKES ("nine hawkish dots"; Warsh's deliberate silence). Futures now price funds toward ~4% by year-end despite the soft June NFP.
- Reaction: shrugged off. Muted on the day (QQQ dipped to its weekly low 700.91, VIX spiked to 18.91 intraday), but the hawkish tone did not hold — risk-on reasserted Thu-Fri to new highs, VIX crushed to 15.03 (low 14.96, the lowest in the sequence). A tape that overrides a hawkish macro print is itself a risk-on structural signal (§3).
Regime trajectory: TRENDING-up, low-vol, risk-on — a regime FLIP from W27's CHOP/ROTATION. Mon new-high push (SPY 751.28, VIX 15.57) → shallow Wed flush on the minutes → sharp Thu-Fri reassertion to SPY 754.95, VIX 15.03. ret10 now positive and rising (vs +0.52% last week). Momentum-rebound tail is live against shorts.
Sector leadership — a FULL REVERSAL of W27's defensive rotation. Leaders (07-02→07-10): Energy +3.49%, Semis +3.16%, Tech +2.87%, Comm +1.86%. Laggards: Materials −2.15%, Health −1.77%, Industrials −1.08%, Staples −1.02%, Utilities −0.76%. W27's defensive winners are this week's laggards — which is exactly why the W27 defensive-basket watch names would have lost (ABBV −5.64%, §4).
Notable flow (diary only): call-OI build normalized by avg_30_day_call_oi with the pinned gates. The naive leaderboard all excludes: TECH 2.69× (confirmed M&A target — Merck KGaA $11.3B, 06-26), RSI 1.72× (near-52w-high = uptrend build), TEL / SHOO / SLM / ENSG / CGNX (earnings inside the 2–4wk hold), CBRS (IPO 05-14, <60td listed), UNM ($3.8B Fortitude Re deal 07-06). Honesty correction: last week's diary flagged CG at 3.77× — a persistence check shows 97% of that build landed in a single day (two ~100k OTM call blocks) = a lumpy institutional hedge, not crowd accretion; the naive 2wk-sum overstated it. The one clean, persistent, beta-neutral, non-deal fade is MAT (§2). Mega-cap tech call-build persists = QQQ-beta trap; crypto-beta bid again (BTDR/MSTR).

2. Validated Weekly Setups (SCORED — excess) — ONE SCORED SHORT: MAT (starter/half-cap)#

Ticker	Lane	Dir	H	validated_excess	regime_fit	tier	note
MAT	OI_FADE	short	10	+0.0088	0.50	MEDIUM	the scored call — see below
MOM_LONG basket	MOM_LONG	long	10	+0.0235	0.55	WATCH	regime finally favorable, but tail-driven/dilute → basket-only, not sized
MOM_SHORT	MOM_SHORT	short	10	+0.0081	0.10	STOOD_DOWN	new-high tape = the momentum-rebound tail; stood down

MAT (Mattel) — the one clean OI_FADE short. Persistent call-OI build every one of 10 days, accelerating into 07-09/10, normalized 0.90× avg_30_day_call_oi, max single-day share 0.40 (passes the persistence check that killed CG). Beta-neutral (toys/consumer-disc, no cluster overlap), not mega-cap-beta, not an ETF/new-listing, not a deal, not near-high (pct_52w_range 0.031 — fresh 52w low 07-09). Fundamentals confirm the short: Goldman cut MAT to Sell on 07-09, the same day it hit a fresh low — i.e. a crowd piling into bounce-bet calls on a battered name, the textbook crowded-embedded-leverage fade (hit 0.61 vs 0.41 base, n=640). Entry 07-10 close $13.33, liquid (~$65M/day). Size = starter/half-cap only (shorting into a new-high VIX-15 tape is the standing headwind; OI_FADE is orthogonal to momentum so the tape is a haircut, not a veto). Earnings 08-04 (~18td) → run to h10 and flatten before the print.
Momentum: MOM_LONG's regime-fit improved (0.45→0.55) — the tape is finally TRENDING-up, the condition this tail-driven lane needs — but the cohort is dilute (480 names, 46% ETFs) with Financials/Healthcare cluster-correlation risk, and the edge is tail-driven (median ~0). Basket-only, never per-name → WATCH. Gate to half-cap only if ret10 sustains >+1% for ≥2wk and the cohort tightens. MOM_SHORT stood down: new closing highs + VIX 15.03 = the momentum-rebound tail that just squeezed all three W26 shorts (§4).
Vol book: non-directional; the VIX-15.03 crush sets up cheap protection into next week's bank-earnings + June-CPI cluster (pre-registered, W29).

3. Weekly Technicals — PRE-REGISTERED (documented, 0 points, NOT sized)#

Feature	This week	Would signal	Status	Pre-reg bar
SPY new-high breakout (HH+HL)	+1.37%, HH 755.42, HL 739.51, close_pos 0.97, new close 754.95 (<06-15 high 756.68)	trend-continuation long if it clears 756.68 and holds	PRE-REGISTERED	PR-WT (≥2yr / ≥30 wk / cross-regime / BH)
QQQ weekly hammer / bear-trap	undercut 705.17 to 700.91 then closed +1.81% (best), lw 0.75, close_pos 0.97	mega-tech leadership reclaimed / failed-breakdown long	PRE-REGISTERED	PR-WT
IWM lower-high + lower-low (2nd wk)	LH 300.41, LL 290.68, only major red (−0.53%)	small-caps failing to confirm = narrow/large-cap-led advance	PRE-REGISTERED	PR-WT
reversal-after-catalyst (hawkish FOMC shrugged off)	hawkish 9-8 minutes dipped QQQ + spiked VIX 18.91 Wed → fully reclaimed to new highs, VIX 15.03	'buy-the-hawkish-dip' vol-crush regime-strength tell	PRE-REGISTERED	PR-WT

These earn ZERO conviction and are never sized. Candlestick evidence is weak-to-null on liquid names (Marshall-Young-Rose 2006) and this panel is ~25 weeks — statistically meaningless for a weekly-pattern claim. They graduate to Layer 2 only by clearing PR-WT on the extended panel; /calibration-audit accrues them.

4. Carry-forward — W26 sized book, RESOLVED at h10 (marked 07-10; final resolves 07-13)#

Entered 06-26 close, marked 07-10 close (h9, effectively final). SPY +3.56% on the window. Excess: short beats SPY when the name underperforms.

Ticker	Lane	name ret	excess (final)	read
MARA	OI_FADE short	−13.34%	+16.90%	big win (partly idiosyncratic miner move)
NVDA	OI_FADE short	+9.57%	−6.01%	loss — flipped from a mid-flight small win; the mega-tech V-recovery hit it
FXI	MOM_SHORT short	+5.98%	−2.42%	loss — China/beta rallied
CME	MOM_SHORT short	+8.72%	−5.16%	loss — exchange squeeze continued
MSTR	MOM_SHORT short	+14.98%	−11.42%	loss — crypto-beta squeeze
ABBV (watch, unsized)	MOM_LONG	−2.08%	−5.64%	would-have-lost — the defensive rotation reversed

The lane split held to the end, with a sharpening lesson. MOM_SHORT went 0-for-3 (avg −6.3% excess) — shorting near-52w-low laggards into a VIX-crushed, new-high tape is the momentum-rebound tail the leg is gated against, confirmed the hard way. OI_FADE finished +5.4% avg but 1-for-2: MARA a huge win, but NVDA flipped from a mid-flight win to a −6.0% loss on this week's mega-tech V-recovery — a live reminder that even in the most-robust lane, a mega-cap-beta name (NVDA) is a QQQ-beta trap. Equal-weight the 5-name book ≈ −1.6% excess, entirely from the MOM_SHORT squeezes; strip them and OI_FADE is solidly green.
The takeaway sizes this week's MAT call: OI_FADE works when the name is idiosyncratic and beta-neutral (MARA idiosyncratic-win; NVDA beta-loss). MAT is toys, near its own 52w low on a downgrade — the beta-neutral kind. W27 had no sized calls. All W26 names roll into /calibration-audit (final at 07-13).

Layer-1 candles + carry-forward built path-aware from the live Yahoo chart API (truth-set parquet stale at 2026-06-26); OI build from the fresh chain-oi-changes panel (10-file trailing window, 06-26→07-10), normalized by avg_30_day_call_oi with persistence + Common-Stock + deal/earnings/near-high/listing-age gates; FOMC minutes web-anchored. Envelope: decision.json (weekly schema w1.0, validated). W28 outcomes unresolved (live panel edge).

Week 28, 2026

The week