Monday plan: no trades into an unconfirmed broadening, with Hormuz risk re-armed
over the weekend
Friday delivered the best-quality tape in a week — tech led, small caps joined, energy was sold — but one session is not a regime, and the weekend put the Strait of Hormuz premium back on the table. The plan into Monday is zero trades: hold MSFT, JPM, AMZN, AVGO and NVDA at unchanged weights, let the open decide whether breadth confirms a second time or the inflation-supply rates row re-arms, and fund any forced response from the book's most multiple-sensitive asset rather than by buying hedges.
Grade: the prior report's plan — no trades, on the view that the book was already sized for chop — was right on outcome and half-wrong on reason. The book returned +0.93% against SPY's +0.61%, a spread of 0.32 points, but it earned that from the exact long-duration technology the report had flagged as the vulnerable side, while the energy bid it worried about reversed instead of extending.
The Call
No trades into Monday's open; the book stays MSFT 23%, JPM 23%, AMZN 20%, NVDA 18%, AVGO 16%. Friday was the healthiest tape in a week — leaders led, small caps joined, energy was sold — but the regime framework requires two consecutive confirming sessions before an upgrade, and the weekend put the Strait of Hormuz risk premium back on the table. When an improving tape cannot yet be confirmed, the correct response is to hold a book already positioned for it, not to add to it at Friday's prices.
Since Last Session
Friday, 7 August, was driven by the unwind of the energy and supply-inflation bid that had dominated the two sessions before it. XLE fell 1.13% and XOM 1.16%; with the supply premium leaking out, duration caught a bid (TLT +0.29%) and long-duration growth reasserted, with XLK +1.42% and QQQ +1.17% against SPY +0.61%. What made the session better than a simple mega-cap rescue was participation: IWM rose 1.11% and XLV 0.75%, so small caps and healthcare joined rather than being left behind. Financials were the other funding source, with XLF down 0.36%. VIX printed 14.9.
The odd one out was gold. GLD rose 2.26%, the strongest move on the macro board, with momentum20 at +6.3% — a hedge bid arriving on the same day volatility slipped below the complacency line. Tonight's packet cannot distinguish whether that is a policy-debasement bid or a geopolitical one, and an input that cannot be resolved lowers risk appetite rather than raising it. GLD is in the candidate universe and it is not in the book, and it will not be: this mandate expresses caution through which growth names it owns and at what size, never by buying hedges.
Inside the book, NVDA led at +2.27% on an 18% weight, AVGO added 1.71% on 16%, AMZN 0.82% on 20%, JPM 0.34% on 23%, and MSFT was effectively flat at +0.03% on 23%. Weighted, the book returned +0.93% against SPY's +0.61%. Era NAV finished at its peak, with drawdown at 0.00% and the drawdown protocol in its normal stage.
Regime Check
- Volatility state: VIX 14.9 — below the 15 line and into the framework's complacent, trend-friendly band, having fallen from 15.15. Trend-friendly on the level; the fall itself is the input, not the number alone.
- Rates impulse and driver: still net negative over the intermediate window — TLT momentum20 -0.74%, momentum60 -2.31% — though Friday's +0.29% and the energy unwind eased it. The driver is unresolved rather than settled: Friday looked like supply-premium decay, while the weekend re-armed it, with Iran denying direct US talks on reopening the Strait of Hormuz, the Houthis claiming an attack on a Saudi refinery, the UAE reporting one of its ships targeted by an airstrike, and oil rising into Monday.
- Leadership breadth: broadening. QQQ +1.17% and XLK +1.42% led, but IWM +1.11% and XLV +0.75% participated, and the down sectors were the recent winners being sold. That is the opposite of the rotation-without-breadth pattern that forced the downgrade.
- Credit and dollar: unverified. No spread series and no dollar series in tonight's packet. The nearest thing to a credit input is Moody's warning that banks' AI adoption is leaving them dependent on a handful of technology vendors — structural, not a spread signal. Unverified inputs can only lower risk, never raise it.
- Event proximity: no tier-1 macro date named in tonight's packet. Treated as unverified, so it cannot be used to justify more risk.
Regime: R2 — choppy rotation, maintained. The two strongest supporting inputs are the still-negative intermediate rates impulse attached to an unresolved, re-arming inflation-supply driver, and the unverified state of credit, dollar and calendar. The contradicting input is substantial and should be named plainly: sub-15 volatility plus genuine broadening is the textbook R1 signature, and the only reason this is not an upgrade tonight is the framework's two-session rule. One more session like Friday flips it.
Plan for Monday
The intended action count at the open is zero. The reason is not caution for its own sake — it is that both plausible paths into Monday argue against trading tonight.
If the open confirms Friday's broadening — technology leading with small caps still participating and volatility contained — the regime upgrades on the second confirming session, and the correct expression of that upgrade is letting NVDA and AVGO work inside the bands they already occupy. Buying them tonight, after a +2.27% and a +1.71% session, would be paying for a confirmation that has not arrived; the R2 rulebook forbids new theme entries on momentum precisely because momentum reverses intraweek in this regime.
If instead the Hormuz escalation prices in and oil extends, the energy and rates complex re-arms and the exposed side is the book's 57% information-technology weight. The response in that case is already written down: cut the most multiple-sensitive growth exposure first — Microsoft at 499.99 against a 20-day average of 423.84 is the book's furthest-extended asset and its named first source of funds — and concentrate what remains. It is not to buy energy, gold or duration. Buying the shock after it has been on the front page for a week is buying a hedge at the price of a thesis.
What the open must confirm, then, is the driver rather than the direction: whether the supply premium keeps leaking out of crude or comes back. Everything else in the plan follows from that one answer.
Positioning
| Holding | Weight | Thesis state | Action + tree branch | Evidence |
|---|---|---|---|---|
| MSFT | 23% | INTACT | Hold — INTACT branch; no pullback-to-structure, so no add | No Microsoft-specific news tonight; +0.03% while XLK gained 1.42%, and a flat session is not evidence either way |
| JPM | 23% | INTACT | Hold — INTACT branch | Moody's bank/tech-dependency warning is a sector structural flag, not a JPM falsifier; +0.34% against XLF -0.36% |
| AMZN | 20% | INTACT | Hold — INTACT branch | No Amazon-specific news tonight; +0.82%, with the early-August digestion resolved at 274.48 versus a 251.59 20-day average |
| AVGO | 16% | INTACT (competitive flag carried) | Hold — INTACT branch | No new information on the AMD/Taalas ASIC overlap since it was flagged; the written falsifier requires a program slip or in-sourcing, and neither occurred |
| NVDA | 18% | INTACT | Hold — INTACT branch; add explicitly declined after extension | Best performer at +2.27%, but a rally is not thesis evidence and R2 punishes momentum entries |
No actions — no state changes. Sector concentration sits at 57% information technology against the 60% cap, with financials 23% and consumer discretionary 20%; no single theme exceeds 23% against the 40% cap.
Institutional Signals
No filing changed: Berkshire, Bridgewater and Pershing Square are all dated 2026-05-15, and Scion's is 2025-11-03 and nine months stale. What is new tonight is a cross-read rather than a filing. Moody's warning that banks' AI push makes them dependent on a small group of technology vendors sits directly across Buffett's 9.5% BAC money-center exposure and this book's JPMorgan position, and it argues that the bank trade and the AI trade are less independent than their theme labels imply. That does not weaken the JPMorgan thesis — net interest income and credit quality are untouched by it — but it is a reason not to credit the position with more diversification value than it earns, and it is why conviction there is held rather than marked up.
What Could Break It
The single headline that would hurt the largest share of the book is a genuine closure or hard escalation at the Strait of Hormuz. That re-arms the inflation-supply rates row, which is the regime framework's explicitly negative line for long-duration growth — and MSFT, AMZN, AVGO and NVDA are 77% of this book between them. Tonight's packet already carries the ingredients: Iran denying direct talks, a claimed Houthi strike on a Saudi refinery, a UAE ship reportedly targeted by an airstrike, and oil rising into Monday. The distinction that matters is between a headline premium, which decays as Friday's did, and a physical disruption that reaches printed inflation data.
Second, a hyperscaler capex guide-down would hit AMZN, AVGO and NVDA — 54% of the book — through one shared mechanism, which is the honest cost of concentrating in the most contracted part of the AI complex rather than the most diversified.
Third, on JPMorgan: high-yield spread widening remains the named falsifier and remains unverified in tonight's packet. A stagflationary supply shock, unlike a simple reflation, eventually reaches credit — and the fact that this input cannot be checked tonight is itself a reason the book is not being pushed further out on risk.