84 Points Ahead of Cathie Wood — and Rebalancing Anyway
The ARKK challenge is winning big. Why I'm still selling four names. Plus the macro read and a new momentum buy.
🌍 Macro
Liquidity and credit are still supportive. But Hormuz is closed again — higher-for-longer just went from tail risk to base case.
Growth solid: GDP nowcast ~2.8% and rising. No recession signal.
June inflation beat: CPI 3.5% (vs 3.8%), Core CPI 2.6%, Core MoM 0.0%. PPI −0.3%. Truflation real-time 2.04%.
The supportive backbone: High-yield spread 2.71% — near 1-year low. Fed credit conditions loose and still easing. Net liquidity at new highs ($5.99T), tracking the S&P.
What changed: Hormuz closed. Brent $88, +15.9% on the week, +37.4% over six months. Core PCE already 3.41% and rising. 10-year at 4.55%. Forward curve prices the Fed higher over the next year (~3.63% → ~4.06%).
Insight: Two forces now pull against each other. Liquidity and calm credit hold the market up. A closed Hormuz + sticky Core PCE push the inflation path — and the Fed — the other way. The June CPI everyone’s cheering is backward-looking; the forward inputs point up.
Implication: Higher-for-longer stops being a “risk” and becomes the working assumption. The oil premium is the transmission mechanism — it feeds headline CPI directly and keeps a Warsh Fed boxed in.
Implication: Prepare the trade, don’t just note the risk.
Energy is the first-order hedge — it wins on the oil premium itself. But it’s a geopolitics long, not a value long. Size it as a hedge, not a conviction hold; it unwinds the day Hormuz reopens.
The second-order play is what survives either outcome: short-duration quality and cash-generative names that don’t need rate cuts to work. Health care, insurance, financials — leadership that’s independent of the oil headline.
What gets hurt: long-duration growth priced for cuts. If the curve keeps pricing the Fed higher, those multiples compress regardless of earnings.
The tell: watch whether defensive-quality holds on days oil falls. If it does, you’ve got a rotation that survives a Hormuz reopening. If it doesn’t, the whole bid is just an energy trade in disguise.
📊 Index
S&P 500 trend fully intact. Three months sideways. Momentum fading at the edges, not breaking.
S&P 500 at 7,458, −1.55% on the week, ~2% off highs. +18.4% over the year.
Structure clean: price above all four weekly EMAs (10/20/30/40), stacked in order. That’s a textbook uptrend.
RSI 60.6 — strong, not overbought. MACD still positive, but the histogram is narrowing. Momentum is cooling, not reversing.
Nearly three months of lateral consolidation. No structural damage.
Insight: This is a pause, not a top. Sideways action while the moving averages catch up is how healthy trends refresh. The caution flag is honest, though: extended lateral pauses have sometimes preceded deeper pullbacks. The chart doesn’t tell you which — the macro does.
Implication: With EMAs stacked and RSI mid-60s, dips stay buyable while the structure holds. No reason to de-risk on price action alone.
Implication: The consolidation resolves on leadership, not the index. A cap-weighted market can grind sideways even as the average stock does fine — if the mega-cap generals stall and nothing large replaces them. So the resolution isn’t “does the S&P hold 7,300?” — it’s “does new leadership (energy, health, financials, cybersecurity) get big enough to carry the tape without tech?” That’s a breadth-and-rotation question the price chart can’t answer alone.
🔄 Sectors
Out of tech and semis, into energy and defensive-quality. Read the energy bid correctly — it’s oil, not value.
Out: Technology (XLK, −5.5% week), Semiconductors (VanEck Semiconductor, SMH, −16.8% 1-month — despite +36% over six months).
In: Energy (Energy Select, XLE, +4.7% week, +22.6% 6-month), Health Care (XLV), Pharma (SPDR Pharmaceuticals, XPH, +62% 1-year), Real Estate, Insurance.
Factors confirm it: this week Growth −5.5%, High Beta −4.2%, Momentum −1.7%. Quality, Value, Dividends held.
Insight: The tape looks late-cycle. But the cycle indicator reads mid-cycle and rising, with earnings +20% year-over-year. The conflict resolves once you separate the two buyers: the energy bid is a Hormuz premium, the health/insurance bid is earnings-quality rotation. Only the second one is structural.
Implication: Defensive-quality leadership is the durable signal — independent of oil, consistent with money moving toward earnings that don’t need rate cuts.
Implication: Ties straight back to Macro. If Hormuz reopens, the energy money must redeploy — back into tech (bullish) or deeper into health/financials/cyber (broadening). Either way, energy is a rented long. The names to own through the whole scenario tree are the ones that work whether or not oil works.
🎯 Themes
One growth theme still leading while the rest correct. That’s the signal.
Leading, alone: Cybersecurity (First Trust Cybersecurity, CIBR) — strength score ~29, +34% over three months, green on every timeframe.
Correcting: the entire AI complex — AI (Global X AI, AIQ, −12.5% 1-month), Robotics (ROBO, −10.1%), Quantum (Defiance Quantum, QTUM, −16.0%), Semiconductors (SMH, −16.8%).
Insight: The AI-adjacency selloff is a crowded-trade flush, not a thesis break. SMH −16.8% on the month but still +36% over six. That’s positioning unwinding, not the AI story ending.
Implication: The crowded first-order AI names — core semis, mega-cap AI — are getting de-rated. Healthy. It clears froth from the obvious winners everyone already owns.
Implication: Cybersecurity leading while AI corrects is the Layer-3 signal firing. The durable edge isn’t the crowded first-order trade — it’s the second-order demand AI creates. Cybersecurity is the expanding attack surface of an AI-saturated world. When the crowded trade flushes and the second-order beneficiary keeps making highs, that’s the map to where leadership migrates next. Watch whether cyber’s relative strength widens as the AI complex stabilizes.
📈 Portfolios
💼 10X Momentum Portfolio
The 10X Momentum Portfolio is my rules-based breakout book — real money, run mechanically. The mandate is: buy quality companies showing momentum, at the moment the chart confirms a Stage 2 breakout, with risk defined before entry. No discretionary overrides. The rules do the work.
Last week I sold ROKU for a 12% gain (79% annualized)
This week’s action: buying XERS (Xeris Biopharma).
XERS is exactly what this book is built to catch — a Stage 2 breakout in a leading sector (health care), backed by fundamentals that most breakout names don’t have. Adjusted ROA is already ~19% and rising to ~21% next year, and it trades at ~20x Adjusted forward earnings. That’s real quality already delivered, at a reasonable price — not a story I’m paying up for. Sector tailwind, clean chart, mispricing on my side.
The order:
Entry: Buy-stop @ $8.85 (triggers only if price confirms the breakout)
Size: 670 shares
Order duration: Open all week
First stop: $7.10
Initial risk~$1.75/share × 670 = ~$1,173 (~19.8% stop distance)
The buy-stop is the discipline: I don’t buy unless the market takes the price up through $8.85 — confirmation, not anticipation. If it never triggers, no trade, no regret.
Scorecard for XERS:
🚀 10X Best of ARKK — Q3 2026 rebalance
The challenge, recap: back in October 2024 I took Cathie Wood’s entire ARKK universe, ran my own screen to find the best ~10 names, and committed to rebalancing every quarter. The bet: disciplined selection beats the fund. [Original post here.]
The scoreboard after 642 days:
10X Best of ARKK ARKK Return since inception: +143.0%
+58.75% Edge vs ARKK +84.3 pts — Annualized: 65.6%
The method is beating ARKK by 84 full percentage points.
The internals are concentrated — AMD is carrying most of the performance — and this rebalance is about keeping the rules honest, not admiring the number.
How the screen ran this quarter. Same funnel, applied across all eight ARKK funds:
182 tickers → to 62: market cap >$50M, revenue >$50M, positive revenue surprise last quarter, revenue YoY >7%, forward revenue >15%.
62 → 13 tickers: clean adjusted-data gate — profitable, improving ROA, expanding assets, improving margins, undervalued-to-fair.
Four sells:
TOST and RXRX — both missed revenue estimates. A negative revenue surprise fails the screen at stage one, chart irrelevant. TOST had been a holder since inception and an earlier winner; it gets no sentimental pass. RXRX was already deep red — the screen just confirmed it.
MELI and CRWV — no longer clear the quality/valuation gate. CRWV was the book’s worst position; it fails cleanly.
Four buys: URGN, SPOT, NTLA, JOBY — all fresh from the screen. Note the pattern: URGN, NTLA, and JOBY are exactly the speculative small-cap ARKK names where this method has historically found its edge — the realized winners of past quarters (QSI, RKLB, COIN, SOFI) were all this type. SPOT is the quality-momentum name in the group. JOBY takes the 10th slot: it adds an uncorrelated urban-mobility sleeve to a book that’s otherwise heavy on semis, pharma, and fintech.
The AMD question — answered by the rule. AMD is +430% since April 2025 and carrying almost the entire book’s gain. The instinct is to trim a parabolic winner. But this book has one rule: screen-in, screen-out. And AMD still passes — profitable, improving ROA, still forward-growing, still fair on adjusted data. So it stays, full position. The concentration risk is real, and I’ll name it plainly: if AMD rolls over, the edge compresses fast. But you don’t sell a name that still qualifies just because the number got big. Discretion is the failure mode — not the rule.
The resulting portfolio (10, equal-weight):
Hold: AMD, LLY, TSM, AVGO, SYM, GENI
New buys: URGN SPOT NTLA JOBY
Fully invested, equal weight, 0% cash — as committed at inception. Proceeds from the four sells fund the four buys.
Next quarter’s watch item: whether AMD still clears the valuation gate after another leg up. The rule that let it run is the same rule that will sell it the day it stops qualifying.
Not investment advice. Do your own research and consult a licensed advisor before acting.