Market Delta: Thursday August 13, Pre-Market

• Cisco Q4 revenue $17.3B (+18% YoY); full-year AI orders $9.3B; stock down in extended trading despite the beat

• July CPI print landed at 3.4% YoY yesterday, +0.1% MoM — cooled but stubborn; September rate-cut probability little changed

• Tomorrow 8:30 AM: PPI + initial jobless claims land together

• S&P and Nasdaq now up four consecutive weeks; index-level rally is the widest since the spring

Today's question: If someone asked what percentage of your portfolio sits in REITs, small caps, municipal bonds, covered-call income, or international equities, could you answer without checking — and does the answer feel intentional?

30-Second Brief

• Cisco reported $17.3 billion in Q4 revenue up 18 percent year-over-year and full-year AI orders of $9.3 billion exceeding management's own guidance, and the stock still fell in extended trading. See Theme 1.

• Tomorrow at 8:30 a.m. Eastern the July PPI print and weekly initial jobless claims land together. PPI is the leading indicator of the CPI print that already happened. See Theme 2.

• The bond market is priced for a soft PPI. If the print runs hot, duration gets repriced first and regional banks give back gains fastest. Don't add duration before the number. See Theme 3.

• Next week concentrates the calendar: FOMC minutes from the July 28-29 meeting drop Wednesday August 19, Walmart reports Thursday August 20. Two distinct reads of the same economy. See Theme 4.

• Gold has been unusually quiet through the summer rally. That is a tell about how professional money is positioned for the next Fed miscommunication. See Theme 5.

The interesting fact about Cisco's numbers last night is not that revenue grew eighteen percent. Revenue growth is the story most reporters will lead with this morning. The interesting fact is that the stock fell anyway in extended trading, and the reason it fell has almost nothing to do with the company and almost everything to do with the price the market had already put on it. That distinction is worth walking through carefully today, because the same pattern is about to repeat at Walmart, at Nvidia, and probably at the FOMC minutes reading next week. When a rally has run four straight weeks and every index is at or near a fresh high, a beat that would have moved the stock five percent in April moves it zero in August. The setups below start with what actually happened at Cisco, then move to the two prints tomorrow morning, then to the reading next Wednesday and Thursday, then close on the metal that has been telling you something all summer.

Cisco Beat, AI Orders Booked $9.3 Billion. Stock Fell Anyway.

Cisco (CSCO) reported fourth-quarter fiscal 2026 revenue of $17.3 billion, up 18 percent year-over-year, with product revenue up 24 percent and networking revenue up 28 percent. Full-year AI orders totaled $9.3 billion, exceeding management's own guidance. Total product orders in the quarter surged 35 percent. FY2027 revenue guidance came in at $72.2 to $73.4 billion. On any conventional reading, this is a strong print — the kind of number that moves a stock several percent in a normal tape. The stock fell in extended trading anyway.

What that fall tells you is not about Cisco. It is about the price of Cisco. Shares are up roughly fifty percent year-to-date going into the print, meaning the market had already paid for a beat of about this shape. When the beat arrives and matches what was priced, there is nothing left to buy. Plain English: the number was excellent and the reaction was small because excellence was assumed. This is what late-cycle rallies do, and it is why the "picks and shovels" AI-infrastructure trade — the networking equipment, the fiber, the power — is now a positioning story, not a fundamentals story. The $9.3 billion in AI orders is the fundamentals story. Retiree portfolios that already own Cisco or a networking sleeve should not chase into the extended-trading dip. Retiree portfolios that own zero networking exposure and have been watching the AI trade from the sidelines have a cleaner entry after the reaction settles.

🎯 Actionable Trade Setup

• Watching: CSCO, ANET (Arista Networks), JNPR (Juniper), CIEN (Ciena) as the networking-hardware sleeve.

• Bias: Constructive on the sleeve on any 3-5% pullback from current levels; sized as a small addition, not a fresh full position.

• Risk note: if AI capex actually slows in Q1 FY27 — the guidance quarter now beginning — the entire picks-and-shovels trade unwinds together. This is a positioning trade, not a conviction trade at current prices.

PPI Tomorrow At 8:30 A.M. The Print That Actually Matters More.

Producer Price Index — a measure of what wholesalers and manufacturers charge before goods reach the shelf — releases tomorrow morning at 8:30 a.m. Eastern. Consensus expectations are for a continued slowing on a month-over-month basis, though core PPI has historically had a habit of running hot when consumer inflation looks tame. The reason PPI matters more than the CPI print you already read yesterday is that PPI is the leading indicator. What wholesalers charge today shows up at the consumer level in six to twelve weeks. The Fed reads PPI as an early signal of where the CPI trajectory is heading, and it is now the number that most directly shifts the December rate-cut probability. Yesterday's CPI at 3.4 percent year-over-year and 0.1 percent month-over-month was cool enough that markets left September cut odds essentially unchanged. A hot PPI tomorrow flips that story before Powell has to say a word.

The mechanics for a retiree portfolio are straightforward. Duration sensitivity — how much your bond holdings move for a given change in yields — is highest in long-dated Treasuries like TLT and in intermediates like IEF, and second-highest in the regional bank ETF KRE because the entire regional bank business model reprices on the short end. If PPI runs hot at 8:31 tomorrow morning, TLT sells off first, IEF second, KRE third. If PPI runs cool, the opposite. Either way, the number you do not want to be doing at 8:29 is adding duration, and the position you do not want to be adding is regional banks on a bet-the-print basis. Wait for the print, then re-read the tape.

🎯 Actionable Trade Setup

• Watching: TLT (20-year Treasuries), IEF (7-10 year), KRE (regional banks), SHY (short duration) as the PPI-reaction map.

• Bias: Neutral into the print; the setup is about not moving before 8:30.

• Risk note: don't add duration before Friday's PPI. If print is hot, TLT gives back weeks of gains inside an hour. If cool, you have all day to add.

The Bond Market Is Priced For Something Specific. Read It.

The 10-year Treasury yield sitting near current levels tells you exactly what the bond market is expecting from tomorrow's PPI: a soft print that keeps the September cut on the table and preserves the December cut probability. That is a specific bet, and it is now the consensus bet, which means the pain trade on PPI is a hot print, not a cool one. When consensus positioning is heavy on one side and the number surprises the other way, the market takes back the move in a single session. If PPI comes in hotter than expected, the 10-year moves up ten to fifteen basis points inside the trading day, TLT drops one to two percent, and the KRE regional-bank ETF gives back a chunk of the summer's rally in an hour. If PPI comes in as expected, essentially nothing happens because the number was already priced.

What this means for a retiree portfolio is not to trade the print. It is to know what you actually own before the print lands. If your bond allocation is heavier in TLT and IEF than you intended — because duration ETFs have quietly rallied and pushed their share of your portfolio higher — tomorrow morning is a bad time to discover that. Do the check tonight. If your duration exposure feels intentional, hold. If it feels accidental because the sleeve grew into overweight, trim the overweight to your target before 8:30, not after.

🎯 Actionable Trade Setup

• Watching: your own portfolio's duration weight versus your target allocation, before Friday's open.

• Bias: bring duration back to target if it has drifted overweight; the check is more important than any bet on PPI direction.

• Risk note: this is a portfolio-hygiene move, not a market call. Don't confuse rebalancing with prediction.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 1: The 401(k) Employee Contribution Limit Rose To $24,500 For 2026

Under the OBBBA-adjusted contribution schedule, the standard employee 401(k) elective deferral limit is $24,500 in 2026, up from prior years, with an additional catch-up contribution of $7,500 for participants age 50-59 and an enhanced catch-up of $11,250 for participants ages 60 through 63 under SECURE 2.0 mechanics. The mathematical implication for a pre-retiree still working full-time is meaningful: the combined employee-plus-catch-up ceiling for a 60-through-63-year-old is now $35,750 in a single tax year, before any employer match, before any after-tax mega-backdoor contribution stream, and before the separate $7,000 IRA limit that stacks on top. For a two-earner household with both spouses eligible, the household's tax-advantaged capacity in 2026 exceeds $80,000 on the elective deferral side alone. Very few pre-retirees actually maximize this capacity even in strong income years, because the paperwork feels routine and the numbers get quietly larger without a corresponding push from the plan administrator.

Why it matters: the years between age 60 and 63 are structurally the highest tax-advantaged accumulation window in the current code. If a pre-retiree in that bracket is contributing at the 2023 limit because that is what payroll defaulted to when they set it, they are giving up several thousand dollars a year of deferred taxes on a decision that takes one phone call to fix.

Next Week Concentrates The Calendar. Read The Setup Now.

Next week is the calendar week that most reporters will treat as quiet because the individual days are quiet. That framing is wrong. The Federal Open Market Committee minutes from the July 28-29 meeting release Wednesday August 19 in the afternoon, and Walmart reports its second-quarter fiscal 2027 results pre-market on Thursday August 20 with consensus revenue near $186.9 billion and adjusted EPS of $0.73. Those are two distinct reads of the same American economy delivered eighteen hours apart. The Fed minutes read what fourteen policymakers actually think about the labor market and inflation, in their own words, with dissent visible for the first time. Walmart reads what a hundred and sixty million American shoppers actually spent, on what, and whether their basket size changed. Together they form the reading a portfolio manager cares about most: what the elite thinks the economy is doing, checked against what the economy is actually doing at the register.

For a retiree portfolio, this two-day sequence is a positioning question, not a trading question. If the FOMC minutes read more hawkish than the market currently prices — that is, if a meaningful number of governors are still worried about services inflation and want more evidence before cutting — the front end of the yield curve gets repriced within minutes of the 2 p.m. release. If Walmart guides cautiously the next morning, staples like XLP and defensive consumer names take a step back and quality-growth reads relatively stronger. Both cases favor a portfolio that is not overweight consumer discretionary or long-duration Treasuries walking into Wednesday. Neither case is a prediction; both are readings of what a portfolio should already look like by Tuesday night.

🎯 Actionable Trade Setup

• Watching: consumer discretionary weight (XLY) versus staples weight (XLP) in your own portfolio; front-end Treasury exposure (SHY, BIL) versus intermediate (IEF).

• Bias: Neutral-to-defensive positioning before Wednesday August 19 afternoon; don't add discretionary or long-duration going into the double-reading.

• Risk note: if the Fed minutes surprise dovish and Walmart guides strong, defensive positioning underperforms for a session. That is an acceptable trade-off for the asymmetry the other way.

Gold Has Been Unusually Quiet. That Is Also A Tell.

Gold spent the first half of 2026 running hard on the combination of central-bank buying, real-rate compression, and geopolitical risk premium. Through the last six weeks that run has flattened. Prices have consolidated in a tight range while equities have gone on to fresh highs. That flatness is not weakness. That is what a market looks like when professional money has already established the position it wants and is now waiting to see whether the next Fed communication validates the trade or forces a reset. The SPDR Gold Shares ETF (GLD) and the physical gold price it tracks are both quietly telling you the same thing the bond market is telling you: the smart-money read is that the Fed is closer to easing than the equity market has fully priced, and the hedge is already on. If FOMC minutes next week surprise dovish, gold breaks the range up. If the minutes surprise hawkish and the December cut probability slips, gold gives back some of the summer's gains. Retiree portfolios with zero gold or precious-metals exposure at all are the ones giving up the hedge; retiree portfolios with a defined small allocation (typically 3-8 percent for the sleeve) should sit and let the position work.

🎯 Actionable Trade Setup

• Watching: GLD, IAU (physical gold), GDX (gold miners) as the metals sleeve; SLV (silver) as the pro-cyclical variant.

• Bias: Constructive on the sleeve as a defined 3-8% portfolio allocation; not a fresh chase at current levels for accounts with zero exposure — wait for a 3-5% pullback.

• Risk note: gold miners (GDX) move roughly twice the underlying metal. Sized as the aggressive sub-sleeve, not the core hedge.

Trade Cheat Sheet: Thu Aug 13

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Market Delta: Wednesday August 12, Pre-Market

• Vanguard Real Estate ETF (VNQ) yielding 3.5% and outpacing the broader market YTD on rate-cut anticipation

• Russell 2000 (IWM) beating S&P 500 by the widest margin since 2003; most retiree portfolios still 90%+ large-cap

• MUB (national munis) yielding ~3.9% tax-free, roughly 5.1% tax-equivalent in the 24% federal bracket

• JEPI paying ~8.5% income; JEPQ ~10.5%; QYLD above 11%; three retiree-income sleeves compounding while you sleep

Today's question: If someone asked what percentage of your portfolio sits in REITs, small caps, municipal bonds, covered-call income, or international equities, could you answer without checking — and does the answer feel intentional?

30-Second Brief

• Vanguard Real Estate ETF (VNQ) is outperforming the broader market year-to-date on rate-cut anticipation, and higher-yielding real-estate sub-sectors are paying up to 10% distributions. See Theme 1.

• Russell 2000 small caps are beating the S&P 500 in 2026 by the widest margin since 2003. Most retiree portfolios that own only SPY or QQQ are missing this rotation entirely. See Theme 2.

• National municipal bond ETFs are yielding around 3.9% tax-free, which pencils to roughly 5.1% tax-equivalent yield in the 24% federal bracket and higher in top brackets, beating comparable-duration Treasuries. See Theme 3.

• Covered-call income ETFs like JEPI, JEPQ, and QYLD generate 8-11%+ annual distribution yields for retirees willing to cap equity upside. See Theme 4.

• International ex-U.S. equities have quietly outperformed U.S. YTD, and the average retiree portfolio remains 90-100% domiciled. One modest rebalance changes portfolio characteristics meaningfully. See Theme 5.

Six weeks of daily coverage on AI earnings and Fed pivots has almost certainly left the non-tech, non-Treasury sleeves of your portfolio under-examined. That is not a criticism of the coverage; the AI story genuinely mattered. It is a reminder that the eighty percent of the market that is not the Magnificent Seven has been doing its own thing all summer, and most of what it has been doing is quietly favorable for the specific portfolio structures retirees hold. Today's map does not touch a single AI ticker, does not preview a single Fed speaker, and does not reference this morning's CPI print. Instead, it looks at five specific sleeves — real estate, small caps, municipal bonds, covered-call income, and international — that most retiree accounts either own accidentally or do not own at all. The five setups are below.

The REIT Sleeve That Was Dead All Year Just Got Its Oxygen Back.

Real estate as an equity sector has spent most of the past two years under pressure from the rate-elevation cycle. Higher borrowing costs compress the value of every property held on a REIT's balance sheet, and higher Treasury yields compete with dividend income for the same retiree dollar. The Vanguard Real Estate ETF (VNQ) — the largest broad-based real estate exchange-traded fund by assets — drifted sideways through most of 2025 as a result. What changed is that the rate-elevation cycle now looks like it may be turning. The market pricing of rate cuts has moved materially in the past week, and REITs are the sector that mechanically benefits most from that repricing. VNQ is now outpacing the broader market year-to-date and yielding roughly 3.5% at current prices. Higher-yielding sub-sectors — mortgage REITs, healthcare-property REITs, some data-center REITs — pay distributions that can reach 10% at current entry points, with correspondingly higher credit and interest-rate sensitivity.

What this means for a retiree portfolio depends entirely on how much real estate exposure you actually hold. If your portfolio is built mostly around SPY, VOO, or QQQ, your REIT weighting is roughly 2-3% (real estate is a small S&P sector). If your intended allocation to real estate is closer to 8-12% — which is the range most fee-based financial planners recommend for retirement portfolios — you are probably underweight by a factor of two or three. The rate-cut trade is not a signal to chase the sector to a fresh overweight. It is a signal to check where your actual weighting sits versus where you would want it to sit if you were setting up cold today.

🎯 Actionable Trade Setup

• Watching: VNQ, XLRE, MORT, REZ; higher-yielding specialty REITs for retiree-income sleeves.

• Bias: Constructive on the sleeve; sized as a check-your-weight question, not a fresh-overweight bet.

• Risk note: if rate-cut expectations reverse (a hot CPI reading, a hawkish Fed speaker), REITs give back gains fastest of any equity sector. The trade is the allocation size, not the entry timing.

Small Caps Beat S&P By Widest Margin Since 2003. Your Portfolio Is Missing It.

The Russell 2000 (IWM) — the standard small-cap benchmark — is outperforming the S&P 500 in 2026 by the widest margin since 2003. That was the year the market broadened out of the dot-com bust and mid-and-small-cap names led a multi-year cycle. There is no guarantee the 2026 rotation continues on the same trajectory, but the pattern of small-cap outperformance historically signals a market that is broadening its participation beyond the mega-cap winners of the prior cycle. That matters specifically for retiree portfolios because the standard "boring index fund" defaults — SPY, VOO, ITOT, VTI — are all market-cap-weighted and therefore mechanically concentrate exposure in the very same mega-cap names that led the last two years. If you own only those funds, you have essentially opted out of the rotation happening in the other end of the market.

The disciplined way for a retiree portfolio to fix this without whipsawing is not to sell large-cap for small-cap. It is to check whether your intended allocation to small caps — typically 5-15% for a diversified retirement portfolio, per most financial-planning frameworks — is actually reflected in your holdings. IWM is the most-owned small-cap ETF; IJR is a lower-fee alternative from Vanguard's iShares peer S&P Small Cap 600 ETF. If you have zero direct small-cap exposure, you are giving up an entire diversification lever, and 2026 so far has demonstrated the cost of that decision.

🎯 Actionable Trade Setup

• Watching: IWM, IJR, VBR (small-cap value) as ETF-level small-cap exposure vehicles.

• Bias: Constructive on rebalance toward intended allocation; sized as diversification, not directional bet.

• Risk note: small caps hurt more than large caps in genuine recession scenarios. If labor-market weakness deepens materially, this trade unwinds. The historical late-cycle rotation pattern is a probability, not a certainty.

The Municipal Bond Yield Nobody Is Advertising In Your Tax Bracket.

Municipal bonds are bonds issued by state and local governments, and the interest they pay is exempt from federal income tax (and often from state income tax for residents of the issuing state). That tax treatment is what makes them uniquely valuable for retiree portfolios in higher tax brackets. The iShares National Municipal Bond ETF (MUB) currently yields around 3.9% on a tax-free basis at current prices. For a retiree in the 24% federal bracket, that pencils to a tax-equivalent yield of roughly 5.1%. In the 32% bracket, it pencils to about 5.7%. Both of those numbers currently exceed the yield on a comparable-duration Treasury bond after federal tax. In high-tax states like California and New York, the differential widens further because state-specific muni ETFs (CMF for California, NYF for New York) exempt both federal and state tax.

What most retiree portfolios miss is that muni exposure is often absent from the standard "bond sleeve" that a retail brokerage default portfolio hands you. If you own AGG, BND, or a target-date fund's bond allocation, you have essentially zero muni exposure — those are all taxable-bond aggregate funds. For a retiree with meaningful taxable-account assets in a high tax bracket, adding a muni sleeve alongside the taxable bond allocation can meaningfully improve after-tax income without changing your risk profile. The mechanism is not glamorous; it is arithmetic. Run a tax-equivalent yield calculation for your specific bracket before your next quarterly rebalance and see whether the numbers now favor tilting a portion of the bond sleeve into munis.

🎯 Actionable Trade Setup

• Watching: MUB, VTEB (Vanguard equivalent), CMF (California), NYF (New York) for state-specific tax layering.

• Bias: Constructive as income sleeve in taxable accounts in higher tax brackets; irrelevant in tax-deferred accounts (IRA, 401(k)).

• Risk note: municipal credit quality varies by issuer. Diversified national ETFs mitigate single-issuer risk; direct state-specific bonds do not. If your bracket is 12% or lower, taxable bonds still win.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 14: Donor-Advised Fund Mechanics Under The Current Regime

A Donor-Advised Fund is an account you contribute to at a sponsoring organization (Fidelity Charitable, Schwab Charitable, Vanguard Charitable are the three largest) where you get an immediate charitable-contribution tax deduction in the year you contribute, but you can direct the actual grants to specific charities on your own timeline, years later if you choose. Under OBBBA's expanded SALT cap of $40,000 and the permanent higher standard deduction, "bunching" charitable contributions into a single DAF contribution every two or three years — instead of giving small amounts annually — has become a mechanically better tax strategy for many retirees. The reason is that in years without bunching, you take the standard deduction and get no incremental tax benefit from small charitable gifts. In a bunching year, you itemize the concentrated DAF contribution alongside your SALT and mortgage interest and clear the standard-deduction threshold by enough to make itemization meaningful. Then in the following one or two years, you grant from the DAF at your normal charitable pace while taking the standard deduction. The DAF holds the money, invests it (tax-free growth), and distributes on your instruction.

Why it matters: For a retiree who gives $10,000-$25,000 annually to charity out of taxable accounts, bunching two or three years of giving into a single DAF contribution during a bracket-favorable year can save four to five figures in federal tax over the four-year 2025-2028 SALT window. The DAF is the vehicle that makes the timing mechanically clean.

Covered-Call ETFs That Compound While You Sleep.

A covered-call ETF systematically writes call options against an equity portfolio and passes the option premium through to investors as monthly distributions. The trade-off is well-defined: you cap your upside in strong equity rallies (because the calls you wrote get exercised against you) in exchange for a much higher and more predictable income stream than the underlying dividend yield alone. For a retirement portfolio that needs cash flow but does not need every last percent of long-term price appreciation, this trade-off is often the right one. The JPMorgan Equity Premium Income ETF (JEPI) currently pays an income yield around 8.5%. The JPMorgan Nasdaq Equity Premium Income ETF (JEPQ) — the Nasdaq-focused sibling — pays around 10.5%. The Global X NASDAQ 100 Covered Call ETF (QYLD) pays over 11%. All three distribute monthly.

The tax nuance most retiree portfolios miss is that covered-call ETF distributions are often taxed as ordinary income rather than as qualified dividends, which means the after-tax yield can be materially lower than the headline yield in a taxable account. In an IRA or 401(k), the distribution is tax-deferred until withdrawal and the tax character does not matter until then, which makes the tax-advantaged account the natural home for these vehicles. For a retiree building an income sleeve inside a rollover IRA, allocating a portion (typically 10-25% of the income sleeve) to a covered-call ETF can meaningfully raise the monthly cash generation without changing the overall equity risk of the account.

🎯 Actionable Trade Setup

• Watching: JEPI, JEPQ, QYLD, XYLD, PUTW as options-income sleeve candidates.

• Bias: Constructive as income allocation inside tax-deferred accounts; awareness of ordinary-income tax treatment in taxable accounts.

• Risk note: covered-call ETFs underperform in strong equity rallies because the call-writing caps your upside. Sized as income allocation, not growth allocation. Not suitable as core equity.

The International Rotation Signal Most Retiree Portfolios Are Missing.

International equities (ex-U.S.) have quietly outperformed U.S. equities year-to-date in 2026, a reversal of the multi-year U.S. dominance most retiree portfolios were built around. The Vanguard Total International Stock ETF (VXUS) is the broadest single vehicle for this exposure and covers both developed markets (Europe, Japan, Australia) and emerging markets in one holding. The average retiree portfolio, according to industry-wide data, sits at roughly 90-100% U.S.-domiciled equity exposure, which is a home-bias overweight relative to almost any diversification framework that considers global market-cap weights (roughly 60% U.S., 40% international at present). Even a modest reallocation — moving 10-15% of equity exposure from a U.S. index fund into VXUS or similar — can meaningfully change portfolio return characteristics over multi-year windows without introducing exotic risk.

🎯 Actionable Trade Setup

• Watching: VXUS, EFA (developed ex-U.S.), VWO (emerging markets) as reallocation vehicles.

• Bias: Constructive on modest reallocation as diversification move; not a directional bet against U.S. equities.

• Risk note: currency risk is real. International ETFs are typically unhedged, meaning dollar strength eats into returns. If dollar strengthens materially, the trade underperforms.

Trade Cheat Sheet: Wed Aug 12

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Market Delta: Tuesday August 11, Pre-Market

• Monday closed with all three major indexes slipping as oil rose and Nvidia pulled back on no company news

• CoreWeave (CRWV) reports after close tonight; consensus revenue $2.56B (+136.8% Y/Y) on a $48.9B market cap; Lumentum (LITE) same evening

• Iran signals Strait of Hormuz reopening deal "very close"; Brent crude firmed to $84/bbl on the news

• July CPI due tomorrow 8:30 AM ET; consensus 3.4% headline / 2.5% core year-over-year

Today's question: When a company reporting tonight carries a Price-to-Sales ratio of 7.27 and its own insiders sold $1.03 billion of stock in the last three months, what does a "clean beat" have to actually look like for the tape to reward it?

30-Second Brief

• CoreWeave (CRWV) reports Q2 after close tonight. Consensus revenue is $2.56 billion, more than double the year-ago quarter. But the setup is uncomfortable: insiders sold $1.03 billion in the past three months, and the balance sheet flashes bankruptcy-risk signals despite the growth. See Theme 1.

• Lumentum (LITE) reports the same evening. This is the optical-networking layer of the AI stack that most retiree portfolios own only through broad-tech ETFs, so the print matters more than the coverage suggests. See Theme 2.

• Wednesday's CPI print at 8:30 AM ET is the actual anchor of the week. Consensus is 3.4% headline and 2.5% core. Do not reposition today for a print you have not yet seen. See Theme 3.

• Palantir shares moved sharply yesterday on a Bank of America upgrade and news of a U.S. agency withdrawing a competing military intelligence bid. Closed at $175.23. See Theme 4.

• Nvidia pulled back Monday on no company news. That is unusual heading into a print two weeks away and worth reading as a positioning tell, not a fundamentals move. See Theme 5.

Tonight brings the week's first serious test of the AI infrastructure story below the hyperscaler layer. CoreWeave prints Q2 after the close with consensus for $2.56 billion in revenue, up more than 136 percent year-over-year — a number that would be impressive for almost any company at almost any valuation, except CoreWeave carries a $48.9 billion market cap on a business that is still losing money and has watched its own insiders sell over a billion dollars of stock in the past three months. Lumentum reports the same evening on a much smaller footprint but the same AI-demand tape. Wednesday morning at 8:30 AM ET, July CPI resolves whether Friday's dovish payroll shock was the beginning of a new regime or a one-print anomaly. And Nvidia pulled back Monday on no company news at all — a positioning tell worth reading before tonight's prints, not after. The map is below.

CoreWeave Tonight: The Ultra-High-Growth, Ultra-Uncomfortable Setup.

CoreWeave (CRWV) reports Q2 2026 after tonight's close. Wall Street consensus is looking for revenue of $2.56 billion — more than double the $1.21 billion the company printed a year ago and a growth rate of roughly 136.8 percent year-over-year. Adjusted EPS is expected at a loss of $1.21 per share. The market cap sits at $48.9 billion. The Price-to-Sales ratio is 7.27, below the historical median for the stock but still elevated by any traditional valuation standard for a business without earnings. The company sells GPU-cloud capacity primarily to hyperscalers and large enterprises — the exact same customers who just told us on their own earnings calls that capex is climbing but that the tape now wants execution proof alongside the spending.

What makes tonight's setup particularly asymmetric for a retiree portfolio is the pattern in the ownership data. Insiders have sold approximately $1.03 billion of stock over the past three months. There have been zero insider purchases in the same window. Institutional gurus have been mixed — some added, some trimmed — but the insider sell-side signal alone is loud. Add a debt-to-equity ratio of 7.39 and an Altman Z-Score of 0.34 (which historically flags meaningful bankruptcy risk within two years), and you have a company that has to deliver a genuinely clean print tonight to hold its multiple. A beat plus raised guidance clears the bar the July earnings cycle established. A beat with soft guidance, or any hint that margin compression is accelerating faster than revenue growth, gets sold. A miss at these levels gets sold hard.

🎯 Actionable Trade Setup

• Watching: CRWV into 4 PM ET print; SMH and SOXX for sector read-across into Wednesday's open.

• Bias: Neutral into the print with downside skew for direct holders; sleeve-awareness for anyone holding CRWV through broad-tech ETFs.

• Risk note: insider selling of $1B in three months is not a normal signal. Do not add direct CRWV exposure before the print regardless of how the story sounds on the call preview.

Lumentum: The Optical Layer Most Retiree Portfolios Own Blindly.

Lumentum (LITE) reports Q4 fiscal 2026 the same evening as CoreWeave. Lumentum makes the optical components — lasers, transceivers, photonic switches — that physically move data between servers inside AI data centers. The company has been targeting $600 million in quarterly revenue by the June quarter on accelerating AI-driven demand. Analysts covering the name have been broadly positive on the AI-infrastructure tailwind heading in. What the print will resolve is whether the demand narrative in the space is broad-based (multiple hyperscaler customers ramping simultaneously) or concentrated (one or two customers driving most of the growth, which creates single-customer risk).

For a retiree portfolio, Lumentum is the kind of name most 55-to-75-year-old holders own without knowing it — through QQQ, SMH, technology sector ETFs, or thematic AI funds that sweep the whole supply chain. When it works, the position quietly compounds. When it breaks, the same broad ETFs absorb the drawdown without any signal to the underlying holder. Tonight's print, combined with CoreWeave's, will tell you whether the second-tier AI infrastructure story remains a source of quiet compounding or has quietly become a source of concentrated risk you did not know you were holding.

🎯 Actionable Trade Setup

• Watching: LITE into 4 PM ET print; COHR reporting Wednesday for confirmation or contradiction.

• Bias: Neutral into print; awareness on ETF-level concentration in optical / photonics names.

• Risk note: customer-concentration guidance is the tell more than revenue itself. If Lumentum flags one-customer dependency, the whole optical thesis re-rates.

CPI Tomorrow: The Anchor. Do Not Move Before You See It.

July CPI releases at 8:30 AM ET Wednesday August 12. Consensus expects headline inflation at 3.4 percent year-over-year and core CPI at 2.5 percent year-over-year. The June print was 3.5 percent headline and 2.6 percent core, so consensus is penciling in a slight continued cooling. Friday's payroll shock (-23,000 jobs against expectations of +83,000) already tilted the market toward pricing a September rate cut. If Wednesday's CPI confirms cooling, the September cut becomes near-consensus and the whole rate-cut trade — long duration, gold, rate-sensitive equity — continues extending. If CPI comes in hot, the market has a genuine contradiction on its hands and Warsh has an impossible-to-easily-resolve position going into Jackson Hole later this month.

The disciplined move today is to not adjust your positioning ahead of a print you have not yet seen. This is a specific temptation right now because Friday's move looked so decisive that adding to the dovish trades today feels rational. It is not. It is chasing a three-day-old repricing that could partially unwind in eighteen hours. What this means for your IRA: whatever bond-sleeve duration and gold-sleeve weighting you had before Friday, hold it into Wednesday's print. Rebalance after 8:30 AM ET, not before.

🎯 Actionable Trade Setup

• Watching: TLT, IEF, GLD, XLU into Wednesday's CPI release; the reversal pattern in the first sixty minutes after the print.

• Bias: Defensive on positioning changes before 8:30 AM ET Wednesday; constructive on gold as multi-scenario hedge.

• Risk note: pre-CPI repositioning is a trade against a print you cannot see. The math is asymmetric against you regardless of which direction the print lands.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 11: The $15M Permanent Estate Exemption Plus Portability Mechanics

Yesterday's Rule 7 covered the headline number: $15 million per person, $30 million per married couple, permanent under OBBBA. Today the mechanics that most estate plans overlook. Portability is the provision that lets a surviving spouse claim the unused portion of a deceased spouse's estate exemption — but only if the executor of the first-to-die spouse's estate files IRS Form 706 within nine months of death, even when no estate tax is owed. Missing that filing forfeits the exemption forever, which under the OBBBA $15 million-per-person regime can mean giving up more than $6 million of tax-shielded transfer capacity permanently. Additionally, under OBBBA the portability election is now more valuable for a broader set of households, because previously many couples with $10 million to $20 million in combined assets did not think they needed to worry about estate taxes at all. Under the pre-OBBBA sunset schedule that number was heading toward $14 million per person, still safe for most. Under permanent $15 million per person, a $30 million combined estate is fully exempt only if portability is properly elected on the first death. A $32 million combined estate faces potential tax on the excess unless planning was done in advance.

Why it matters: If a spouse dies and the surviving spouse's estate attorney does not file Form 706 within nine months, the exemption is lost, full stop. This is the single most common estate-tax planning error retirees make, and OBBBA made the cost of the error larger, not smaller. Add "confirm Form 706 filing plan" to the estate-attorney review you scheduled from yesterday's Rule 7.

Palantir's Monday: A Reprice On Two Signals At Once.

Palantir (PLTR) closed Monday at $175.23, up 1.87 percent on the session after moving as much as ten percent in early trading. Two distinct signals drove the pre-market rally that partly unwound into the close. Bank of America published an upbeat post-earnings note repositioning the price target higher after last week's 93 percent revenue growth print. And separately, a U.S. agency withdrew its bid for a competing military intelligence contract that Palantir has been positioning around, effectively clearing the competitive field for a name that already carries a $32 billion market cap and stretched valuation multiples. The fact that PLTR gave back most of the pre-market gain into the close is itself a tell — it means the tape absorbed both bullish signals into existing positioning rather than chasing them into fresh highs.

For a retiree portfolio holding PLTR through diversified tech ETFs, Monday's move barely registered on the ETF level. For anyone holding PLTR direct, the takeaway is that the stock is now at a level where each incremental piece of good news adds diminishing marginal return, because the earlier reprice already absorbed the growth story. If tonight's CoreWeave print goes badly and the whole AI-services complex sells in sympathy, PLTR is one of the names that gives back Monday's gains and more in a single session. If tonight's prints go well, the ceiling on PLTR is now closer than it was two weeks ago.

🎯 Actionable Trade Setup

• Watching: PLTR reaction to tonight's CRWV print; XLK for read-across; concentration inside diversified tech ETFs.

• Bias: Neutral direct; awareness of AI-services complex correlation risk.

• Risk note: PLTR at $175 is priced for continued execution. A negative read from tonight's AI infrastructure prints can rerate this name faster than most retiree holders expect.

Nvidia's Quiet Monday Pullback Was A Positioning Tell.

Nvidia pulled back Monday on no company-specific news. Most sessions in a mega-cap tech name that move without a catalyst are noise. This one is worth watching because it happened fifteen trading days before Nvidia's own Q2 earnings on Wednesday August 26 — the specific print that will determine whether the entire AI complex re-rates higher or lower for the fourth quarter. When a name pulls back on no news two weeks before its own catalyst, it is usually because institutional positioning is de-risking ahead of the event. That is a signal, not a story: someone is preparing for the possibility that the print falls short of the capex-versus-execution template that Amazon, Microsoft, and Palantir cleared while Meta, SpaceX, and AMD failed.

🎯 Actionable Trade Setup

• Watching: NVDA into today's open and through tonight's CRWV/LITE prints; any acceleration of pullback into Wednesday's CPI.

• Bias: Awareness only. Sleeve consideration for direct NVDA holders through August 26.

• Risk note: institutional de-risking often front-runs the news it is de-risking against. If NVDA continues to slide on no news through this week, that is the tape telling you what it expects on August 26.

Trade Cheat Sheet: Tue Aug 11

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Market Delta: Monday August 10, Pre-Market

S&P futures firmer pre-open on softer jobs data and lower yields carrying into Monday
July CPI drops Wednesday 8:30 AM ET; consensus 3.4% headline / 2.5% core year-over-year
Four AI infrastructure earnings this week: CRWV and LITE Tue after close, COHR Wed, AMAT Thu
Houthi strike on Saudi Aramco's Jazan refinery Sunday; oil markets not yet pricing escalation risk

Today's question: When Friday's data flipped the September Fed trade in four hours and this Wednesday's data could flip it back, what is the disciplined way to position a retirement portfolio without doing both trades and paying the whipsaw cost twice?

30-Second Brief

• Friday's payroll shock (-23,000 jobs vs +83,000 expected) collapsed September Fed hike odds and drove Nasdaq to a record close on rate-cut anticipation. Monday opens with that positioning still fresh. See Theme 1.

• July CPI at 8:30 AM ET Wednesday is the second data point that either confirms or contradicts Friday's dovish read. Consensus is 3.4% headline and 2.5% core. Some analysts see hotter numbers coming. See Theme 2.

• CoreWeave, Lumentum, Coherent, and Applied Materials report this week — four independent tests of whether the AI infrastructure story that survived the July earnings gauntlet holds one more round. See Theme 3.

• Houthi militants struck Saudi Aramco's Jazan refinery Sunday. Oil markets have not yet priced escalation risk. The last comparable strike (Abqaiq 2019) moved Brent 5-8% in a single session. See Theme 4.

• Friday's rate-cut repricing started compressing money market fund yields at the major brokerages. If your cash sleeve was sized against 4.5%-plus assumed yields, this week is when that math starts to shift. See Theme 5.

Friday flipped the tape in four hours. The economy shed 23,000 jobs in July against expectations of an 83,000 gain, the September Fed hike odds collapsed, gold posted a new all-time high, and the Nasdaq closed at a record on rate-cut anticipation. Monday opens with all of that positioning still fresh, and with a specific problem: Wednesday's CPI print could either confirm the pivot or contradict it into a stagflation setup where the Fed cannot cut and cannot hike. Four AI infrastructure earnings land Tuesday through Thursday, testing whether the sector's July results were the story or a one-off. And over the weekend, Houthi militants struck Saudi Aramco's Jazan refinery — a fact the oil tape has not yet begun to price. The map for the week is below.

The Regime Flipped Friday. Monday Is When You Test It.

The July payroll shock was not incremental. Consensus expected 83,000 jobs added; the print was a loss of 23,000. The three-way hawkish dissent at the July 29 Fed meeting, the June dot plot showing nine of eighteen participants penciling in a 2026 hike, and the 64% September hike odds priced by Fed funds futures — all of it lost its data anchor in a single 8:30 AM tick Friday. Gold rallied 1.77% to $4,375.80, a new all-time high. The Nasdaq closed at a record on 0.86% strength. TLT and other long-duration bond funds marked up sharply after weeks of pressure. Any retiree portfolio that spent the past two weeks defending against a September hike found itself repositioned by the market rather than by choice.

The uncomfortable part of the Monday setup is that the pivot is only three trading days old. Wednesday's CPI print is the second data point, and if it comes in hot — above 3.5% headline or above 2.7% core — the Friday rally at least partly unwinds. The disciplined approach today is not to add fresh conviction to either side. It is to make one specific check: is your current positioning still the positioning you would choose if you were setting the portfolio up cold today, or is your book now overweight the trades that worked Friday because Friday just marked them up? The market rewards portfolio consistency across regime changes more than it rewards being right about the next data point. Sunday's kitchen-table check we discussed applies with more force today, not less.

🎯 Actionable Trade Setup

• Watching: TLT, GLD, QQQ into Wednesday's 8:30 AM ET CPI release; the pattern of first-hour reversal after the print.

• Bias: Constructive on rate-cut-adjacent trades; sized as position management, not directional conviction.

• Risk note: a hot CPI Wednesday partially unwinds Friday's move. Do not concentrate weekend gains into single-name direct exposure before the print.

CPI Wednesday: The Second Data Point, Not The Only One.

July CPI drops at 8:30 AM ET Wednesday August 12. Consensus expects headline inflation at 3.4% year-over-year and core CPI at 2.5% year-over-year. The June print came in at 3.5% headline and 2.6% core, so consensus is essentially penciling in another slight cooling. Some analysts are calling for a hotter number — the Seeking Alpha piece last week argued for a print that could accelerate what they termed an AI-adjacent unwind. What matters for a retiree portfolio is not the headline itself. It is the combination of Wednesday's CPI with Friday's payrolls to answer a specific question: is the U.S. economy transitioning to a slower-but-cooler regime the Fed can cut into, or to a slower-and-still-hot regime that puts Warsh in the impossible position we discussed Sunday?

The Thursday PPI print at 8:30 AM ET and the Friday retail sales print at 8:30 AM ET provide the confirming or contradicting color. Three data points in three consecutive sessions is unusual density. If all three lean dovish (cooling prices, weak producer inflation, softening retail spending), the September rate cut becomes near-consensus and October could get pulled forward as a possibility. If all three lean toward the stagflation reading (prices holding, producer costs elevated, retail spending sticky), the market has to price a Fed that would rather see recession than reignite the inflation fight. Two very different environments for your bond sleeve. Two very different environments for gold. One CPI print starts the sequencing.

🎯 Actionable Trade Setup

• Watching: TLT, IEF, HYG, GLD, XLU into CPI Wed 8:30; PPI Thu 8:30; Retail Sales Fri 8:30.

• Bias: Defensive on long-duration into the three-print sequence; constructive on gold as multi-scenario hedge.

• Risk note: three prints in three sessions is a compounding sequence, not three independent trades. Position for the sequence, not for each print.

Four AI Infrastructure Prints. Same Template. Different Layer.

CoreWeave (CRWV) and Lumentum (LITE) report Tuesday after the close. Coherent (COHR) reports Wednesday after the close. Applied Materials (AMAT) reports Thursday after the close. That is four AI infrastructure names inside three sessions, testing whether the sector's July prints were the story or a one-off. The capex-versus-execution template that Amazon, Microsoft, and Palantir cleared — and that Meta, SpaceX, and AMD failed — now applies to a different layer of the AI stack. CoreWeave is the GPU cloud tier that sits between hyperscaler capex and enterprise deployment. Lumentum and Coherent are the optical-networking and photonics layer that physically moves the data. Applied Materials is the semiconductor equipment tier that builds the tools that make the chips.

The read-through for a retiree portfolio is layered. If all four clear the template with clean beats and disciplined guidance, the infrastructure-beneficiary thesis we have been holding — NextEra Energy, Quanta Services, Vertiv, Eaton — gains a second wave of validation from the supply-chain end rather than just from the hyperscaler-capex end. If any of the four misses badly, the chip complex broader than just AMD gets marked down and the reflexive weakness carries through to power and cooling names in sympathy. Nvidia's own print on Wednesday August 26 becomes the final arbiter, but this week's four prints tell us whether Nvidia is walking into a supportive tape or a broken one.

🎯 Actionable Trade Setup

• Watching: CRWV, LITE, COHR, AMAT after each respective close; SMH and SOXX for sector-level; NEE, PWR, VRT for beneficiary read-through.

• Bias: Constructive structural on infrastructure-tier; neutral into individual chip prints.

• Risk note: if two or more of the four fail the template, the read-across to Nvidia's August 26 print becomes negative. That is a chain effect worth watching, not a single-print outcome.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 7: Estate And Gift Tax Exemption Raised To $15M Per Person ($30M Couples)

Under OBBBA, the federal estate and gift tax exemption is now $15 million per person, or $30 million per married couple using portability. This is a permanent increase from the roughly $14 million per-person exemption that would have prevailed under the pre-OBBBA sunset schedule. For a retiree household with meaningful assets, the practical effect is that the estate planning strategies designed to work around a lower exemption — specifically, aggressive lifetime gifting to lock in the higher exemption before it sunset — are no longer urgent, because the sunset is not coming. That changes the math on Grantor Retained Annuity Trusts (GRATs), Charitable Lead Annuity Trusts, and installment sale strategies to intentionally defective grantor trusts. Under a Fed that is now pivoting toward rate cuts, some of these strategies become materially more attractive because they rely on the "hurdle rate" published monthly by the IRS as the Section 7520 rate. When Treasury yields fall, the 7520 rate falls with them, and GRATs and similar vehicles get more efficient.

Why it matters: If your estate plan was set up in 2019-2023 around a $12 million exemption and pre-cut hurdle rates, the plan is likely due for review. Both the exemption number and the rate environment have moved. This is the quarter to schedule the conversation with your estate attorney, before the September Fed decision either accelerates the rate move or unwinds it.

Houthis Hit Aramco. Oil Didn't Even Blink.

Houthi militants struck Saudi Aramco's Jazan refinery over the weekend, an attack that in almost any prior six-month window would have moved Brent crude four to eight percent in a single session. The 2019 Abqaiq drone strike took Brent up 15% intraday and settled at plus-eight. This weekend's strike, so far, has moved oil markets only marginally. The reason the tape is not pricing escalation is that the parallel Iran-Oman Hormuz-reopening deal talks continued through the same weekend, with Iran reportedly demanding concessions but signaling a framework close. The market is choosing to price the deal path, not the escalation path, at least until Monday's open shows the volatility bid or absence thereof.

For a retiree portfolio that upsized energy sleeve exposure earlier in the summer as an Iran-tension hedge, the weekend attack is a reminder that geopolitical tail risk does not resolve cleanly on a policy timeline. The right-sizing we discussed last week — trimming Iran-hedge oil back to a baseline weighting — still applies, but with awareness that a second Houthi strike within the next two weeks would snap oil back sharply and the tape would price the escalation path it is choosing to ignore today. Gold, which held through every prior Iran-related test, continues to be the hedge that works across both deal-succeeds and deal-fails scenarios.

🎯 Actionable Trade Setup

• Watching: XLE, CL futures for the first-hour reaction to the Aramco strike news; GLD for continuation of Friday's move.

• Bias: Neutral tactical on oil; bullish structural on gold as multi-scenario hedge.

• Risk note: a second Houthi strike or a Iran deal collapse would each move oil sharply higher within one session. Sized as awareness, not directional add.

The Money Market Yield Just Started Compressing.

Friday's shift in rate-cut expectations started an effect that will show up in your Fidelity, Vanguard, or Schwab money market fund yield over the next several sessions. The Fed funds rate is still 3.50%, but money market fund yields are calibrated to the effective rate the market expects the Fed to hold, and when September cut probability moves from 20% to near-consensus over three sessions, the seven-day yields on the government-only funds start to compress. If your cash sleeve was sized against a 4.5%-plus assumed yield — and this is not an uncommon assumption for retirees who added to money market positions during the 2024-2025 rate-elevation window — this week is when that assumption starts to shift. Not a trade. Just a fact your Q3 planning needs to absorb before Wednesday's CPI potentially accelerates or partially reverses the move.

🎯 Actionable Trade Setup

• Watching: SPAXX, VMFXX, SWVXX seven-day yields Monday and Tuesday against your last-check baseline.

• Bias: Sleeve consideration only. Not a directional call; a "know what you are actually being paid on the cash sleeve now versus a week ago" check.

• Risk note: if the Fed cuts 25 basis points at the September 16 meeting, MMF yields drop within days. If they cut a second time in October or November, the compression accelerates.

Trade Cheat Sheet: Mon Aug 10

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Friday's payroll shock repriced the whole month of positioning in four hours. The Fed hawks who dominated the week's opening lost the September hike argument in a single 8:30 AM tick. But a single Friday does not make a regime. The regime gets defined over the next three weeks by three data points, and the first two land on Wednesdays. This coming Wednesday brings the July CPI print, which either confirms the dovish read Friday started or contradicts it into a stagflation setup. Seventeen days later, on Wednesday August 26, Nvidia reports Q2 earnings after the close. Thursday August 27, the Federal Reserve's Jackson Hole symposium opens. Friday August 28, Warsh delivers his first Jackson Hole speech. That is a rare convergence of three market-moving catalysts inside three consecutive sessions. What you do this Sunday matters less than being ready for those windows.

Further Reading from Bastion Stability

TThis Wednesday's CPI Print Either Confirms Or Contradicts.

The June CPI report, which released July 14, showed headline inflation at 3.5% year-over-year and core CPI at 2.6% year-over-year, with the monthly core reading landing flat at 0.0% versus consensus of 0.2%. That was a materially cooler print than the market had feared going in. Friday's payroll shock — the U.S. economy shedding 23,000 jobs against expectations of an 83,000 gain — adds a second dovish data point. If Wednesday's July CPI comes in at or below 3.4% headline and 2.5% core, the disinflation-plus-labor-weakness story becomes the operating framework for the second half of 2026, and the September rate cut becomes the base case rather than a possibility. In that scenario, long-duration bond funds continue the rally that started Friday afternoon, gold consolidates above $4,300, and rate-sensitive equity sectors (utilities, real estate, dividend-paying names) get a second wind.

The uncomfortable scenario is a hot CPI. If headline inflation reaccelerates above 3.6% or core moves back above 2.8%, the market has a contradiction on its hands: prices rising while jobs contract. That is the classic stagflation setup, and it puts a Fed chair thirty days into the job in an impossible position. Cutting rates supports the weakening labor market but reignites the inflation the FOMC just started to trust as easing. Holding rates protects inflation credibility but accelerates the jobs weakness that would otherwise resolve into recession. Warsh has said this Fed is reducing forward guidance, which means the market will not get advance signals about which way he leans in that scenario. Positioning for stagflation is different from positioning for either pure inflation or pure recession — it favors gold (which we already hold), short-duration Treasury income (which we have discussed), and defensive equity sectors (staples, healthcare) over both growth tech and long-duration bonds.

What this means practically for a Sunday at the kitchen table is that neither scenario requires you to trade tomorrow. Both scenarios require you to know what you would do if Wednesday's print comes in outside the expected range. That is a five-minute thought experiment: "if CPI prints 3.7% headline instead of 3.4%, what do I trim, what do I add, and in what order?" Write the answer down before the print, not after. The disciplined move on a print like this is to execute a pre-decided plan, not to react in real time.

📅 Forward Catalyst Calendar — Next Twenty Sessions

August 26 to 28: Three Catalysts, Three Sessions, One Sequence.

Nvidia reports Q2 earnings after the close on Wednesday August 26. Thursday morning August 27, the Federal Reserve's annual Jackson Hole symposium opens in Wyoming. Friday morning August 28, Kevin Warsh delivers his first Jackson Hole speech as Fed Chair. That is three market-moving catalysts of the largest possible magnitude landing in three consecutive trading sessions. Historical Jackson Hole data since 2015 shows the Fed chair's speech drives an average S&P 500 move of positive 0.74% on up days and negative 1.74% on down days. Semiconductor stocks show larger swings — AMD averages 3.46% absolute movement, and Nvidia posted a 9.23% single-day drop in the 2022 speech reaction. Nvidia's own earnings, independent of Jackson Hole, are historically the most-watched single-stock print of any quarter. The three events in sequence create a window where the tape reacts to Nvidia overnight, opens into Jackson Hole's first day still processing that reaction, and then meets Warsh's speech Friday morning before the first two reactions have fully settled.

What makes this sequence particularly consequential is that Nvidia is the anchor test of the capex-versus-execution earnings template we have watched define the past three weeks. Amazon, Microsoft, and Palantir cleared the template and got paid. Meta, SpaceX, and AMD failed it and got sold. Nvidia is the largest single AI-exposure holding in almost every diversified index a retiree portfolio might carry. If Nvidia clears the template with a clean beat plus disciplined capex plus execution proof Wednesday night, the broader tech complex re-rates higher into Jackson Hole's opening, and Warsh's Friday speech either confirms or complicates that move. If Nvidia fails it, the tech complex re-rates lower and Warsh's speech either rescues or amplifies the damage. In either sequence, the setup asks not "will you trade the print" but "is your positioning right for the sequence."

The Sunday check to do this weekend is not about the two Wednesdays. It is about the seventeen days between them. Bond sleeve duration: is it where you wanted it before Friday, and if the September cut becomes consensus this Wednesday, is it still where you want it heading into Jackson Hole? Direct Nvidia exposure: if you carry any, is the position sized so that a nine-percent single-day move against you does not damage the year? Gold sleeve: with a new all-time high in hand, are you still at your target weight or has the mark-up quietly overweighted the sleeve? Three checks, done this afternoon, are worth more than any Monday-morning reaction to Wednesday's print.

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

The week ended in the exact inverse position of how it began. Monday opened with Fed funds futures pricing a 64% probability of a September rate hike, the June dot plot showing nine of eighteen FOMC participants penciling in a 2026 hike, and the three Fed hawks who dissented at the July meeting looking prescient. Friday closed with July payrolls printing negative — the U.S. economy shed 23,000 jobs against expectations of an 83,000 gain — the September hike odds collapsing to near zero, gold posting a fresh all-time high at $4,375.80, and rate-cut expectations returning to the front of the yield curve for the first time in weeks. In four hours after the 8:30 AM release, the hawkish narrative that had shaped every trade for two weeks became yesterday's positioning. Anyone whose bond sleeve was heavily defensive going into Friday got the wrong side of a fast repricing.

Further Reading from Bastion Stability

The Week The Fed Hawks Lost

Friday's payroll number was not a small miss. It was a genuine shock. Economists surveyed by Reuters expected 83,000 new jobs; the actual print was a loss of 23,000. That is not "cooling data"; that is the labor market visibly contracting for the first time in months. The unemployment rate ticked to 4.1% from 4.2%, but the improvement is mechanical rather than real — labor force participation dropped to 61.4%, a five-year low, which means people are leaving the workforce faster than employers are cutting jobs. Local government education and retail trade were the largest sector declines. Under any reading Chairman Warsh gave the market last month, this print takes the September hike option off the table. The three-way hawkish dissent that framed the July meeting no longer has data to point to.

The market's reaction happened in the specific way that matters most for a retiree portfolio: the split. Nasdaq rallied 0.86% and closed at a new high, because long-duration tech benefits when rate expectations fall. The S&P 500 rose 0.33% and set a record close, on its strongest week since April. But the Russell 2000 — small caps — slipped 0.58%, because a labor market contracting is bad for small-cap earnings even if it is good for rates. Gold rocketed 1.77% to $4,375.80, an all-time high. Silver jumped a much larger 5.01% to $64.69. Bond yields fell. If your portfolio had the long-duration Treasury sleeve we spent all week describing as "defensive," Friday marked those positions up. If your bond sleeve was still short-duration and floating-rate as a hedge against the September hike, Friday marked you down. The regime shifted, and the positions that were right for Monday's regime became wrong by Friday's close.

What this means for the retiree portfolio through this weekend is that the tempting move — to reposition heavily into the new rate-cut regime before Monday's open — is exactly the move that will be reversed by the next data point. One month ago the Fed hawks were fringe voices; two weeks ago they were the dominant narrative; today they are gone. That kind of narrative acceleration means the market is going to overshoot the pivot, then correct, then overshoot again. The disciplined move is to note the regime change, refuse to trade it aggressively into the weekend, and let CPI on Wednesday and Warsh at Jackson Hole later this month provide the second and third data points that either confirm or reverse this Friday's reading.

📊 By The Numbers — Week Ending Friday August 7

Look-Ahead: CPI Wednesday, Nvidia August 27, And What Actually Matters

Next week is short on earnings and long on data. The July Consumer Price Index prints Wednesday morning at 8:30 AM ET. If CPI comes in soft (headline near 2.7% year-over-year, core near 2.9%), Friday's dovish jobs read gets a second confirming data point and the September rate cut becomes near-consensus rather than newly-emerging. If CPI comes in hot (headline above 3.0%, core above 3.1%), the market has a contradiction on its hands: jobs contracting while prices still rise. That is the classic stagflation setup, and it puts the Fed in an impossible position where cutting rates supports labor but reignites inflation, and holding rates protects inflation credibility but accelerates the jobs weakness. Every retiree portfolio that thinks it is "positioned" for one regime needs to consider what happens if the two indicators diverge on Wednesday.

The other date on the calendar is Nvidia's earnings on Wednesday August 27. That is the largest single test of the "capex-plus-execution" template that dominated the past two weeks. Amazon, Microsoft, and Palantir cleared the template and got paid. Meta, SpaceX, and AMD failed it and got sold. Nvidia is the anchor holding of every AI trade in every diversified index a retiree portfolio might carry. If Nvidia clears the template — revenue beat plus disciplined capex plus proof of execution — the broader tech complex re-rates higher. If Nvidia fails it, the broader tech complex re-rates lower, and this Friday's Nasdaq rally on rate-cut anticipation becomes the top of the range rather than the middle. That single print, three weeks from now, matters more than most of the intervening data.

What to do this weekend is not to trade either scenario preemptively. It is to check three specific things: whether your bond sleeve's duration is where you wanted it before Friday's move (if not, plan the trim or add for next week's calm sessions, not for Monday's open); whether your gold sleeve is at your target weight now that the price has moved to a new high (structural buyers are still bidding, but position-sizing rules apply); and whether any single-stock direct exposure you carry is small enough that a Nvidia disappointment on August 27 would not materially damage your year. Three checks, done at the kitchen table on Sunday, are worth more than any Monday-morning trade.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 12: 529-To-Roth IRA Rollover Under SECURE 2.0

Under SECURE 2.0, retained inside OBBBA, unused 529 education-account balances can now be rolled into a Roth IRA in the beneficiary's name, tax-free and penalty-free, subject to specific limits: the 529 account must have been open for at least fifteen years; annual rollovers are capped at the current Roth contribution limit ($7,000 for 2026, $8,000 age 50+); and the total lifetime rollover cannot exceed $35,000 per beneficiary. For grandparents and parents heading into back-to-school season, this changes the math on overfunding a 529. Before this rule, "leftover" 529 balances after college were locked into education-only uses or subject to penalty on non-qualified withdrawal. Now, any surplus after the beneficiary finishes school (or if they never go) has a legal path into that child's or grandchild's tax-advantaged retirement account, starting them decades ahead on Roth compounding.

Why it matters: The fifteen-year lookback means a 529 opened for a grandchild in 2026 becomes rollover-eligible in 2041. If you are contemplating opening a 529 for a young grandchild this back-to-school season, opening the account this year rather than next starts the fifteen-year clock immediately, even before contributions accumulate meaningfully.

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Market Delta: Friday August 7, Pre-Market

SpaceX (SPCX) closed +6.14% to $114.92 on unlock day; 254M shares traded, 109% above three-month average, on 36%-of-float short covering
Thursday close: S&P 500 -0.16% to 7,711; Nasdaq -0.06% to 26,348; Dow pulled back from record
July NFP due at 8:30 AM ET; consensus range 91K-120K jobs, unemployment ticking to 4.3% from 4.2%; average hourly earnings +0.3% expected
10-year yield holding near 4.67%; oil steady after three-session slide; Hormuz deal still pending

Today's question: When a single data point at 8:30 AM ET can move the 10-year Treasury yield 10 basis points in either direction, what is the disciplined move for a retiree portfolio — before the print, at the print, and heading into the weekend?

30-Second Brief

• July nonfarm payrolls print at 8:30 AM ET. Consensus range is 91,000 to 120,000 jobs with unemployment moving to 4.3%. This is the single data point that resolves the September Fed hike question. See Theme 1.

• SpaceX defied the $103 billion supply wall Thursday and closed up 6.14% at $114.92. The unlock failed to trigger a selloff because 36% short interest covered into the event and roughly half of the eligible shares remained mechanically locked. See Theme 2.

• This week confirmed the "capex-versus-execution" earnings template on its fifth print. Amazon, Microsoft, Palantir cleared the bar. Meta, SpaceX, AMD did not. The rotation now runs at the ticker level, not the sector level. See Theme 3.

• Oil closed a three-session slide on continued Hormuz-deal signaling. Gold held above $4,100 through the same de-escalation, confirming the structural-versus-tactical hedge distinction. See Theme 4.

• Discipline for the weekend: whatever moves in the first hour after the 8:30 print is often reversed by close. Any weekend repositioning should be planned pre-print, executed post-print calmly, not chased. See Theme 5.

The July payrolls print lands at 8:30 AM ET today, roughly two hours from this note. That single data point resolves the September Fed hike question that has driven the bond market's positioning all week. Consensus sits between 91,000 and 120,000 jobs added, with unemployment ticking to 4.3% from 4.2%. Thursday closed with SpaceX rising six percent through what was supposed to be a $103 billion supply wall — short covering plus a self-limiting staggered lockup structure absorbed what everyone had priced as a selloff catalyst. The Dow pulled back from its record. Oil finished a three-session slide on Hormuz-deal signaling. Now one number decides whether your bond sleeve gets marked up or down heading into the weekend. The map is below.

Two Numbers At 8:30 That Decide Your Bond Sleeve.

July nonfarm payrolls release at 8:30 AM ET. Continuum Economics is at 120,000 jobs added; CNBC consensus is closer to 91,000. Both project unemployment at 4.3%, up from June's 4.2%. Average hourly earnings are expected up 0.3% month-over-month, holding 3.5% year-over-year. The internals matter as much as the headline. June's payroll weakness came from a 61,000-person drop in leisure and hospitality that management economists attribute to seasonal-adjustment noise; a snap-back there would push the July print toward the upper end of the range. The 10-year Treasury at 4.67% is priced for a moderate result. The tape moves on any surprise larger than roughly 25,000 in either direction.

What this means for a retiree portfolio in the next two hours is that the disciplined move is no move. Any repositioning of the bond sleeve done between now and 8:30 is a guess about the print rather than a response to it. After 8:30, the first-hour tick is often reversed by close, especially on a summer Friday when institutional desks are thin. If the print comes in above 130,000 with unemployment holding, the September hike odds surge and TLT marks down — but the reactive selling can overshoot before the buy-side comes in on Monday. If it comes in below 60,000 with unemployment ticking to 4.4%, the September hike falls off the table and TLT rallies — but the buying can overshoot the same way. Position calmly after the tape settles.

🎯 Actionable Trade Setup

• Watching: TLT, IEF, HYG; the 10-year yield tick at 8:30 AM ET; the first-hour reversal pattern.

• Bias: Defensive long-duration going in; sized as duration management, not directional bet.

• Risk note: any position sized as a one-way bet on the print is asymmetric against you. Binary events with two-sided asymmetry reward calm, not conviction.

The Supply Wall That Wasn't. What SpaceX Just Taught The Tape.

The $103 billion SpaceX unlock Thursday was supposed to be a selloff catalyst. Instead SPCX closed up 6.14% at $114.92, on trading volume roughly 109% above the three-month average. Two things absorbed the supply. First, short interest had climbed to approximately 36% of the float ahead of the event, and traders who had positioned short into the unlock covered on any weakness — a mechanical rally rather than a fundamental one. Second, the lockup structure is not a single-cliff release. Roughly 455.8 million additional shares remain mechanically locked because the stock trades below its $135 IPO price. The staggered nine-stage structure means additional supply only enters the market if the stock price rises to unlock trigger levels — a self-limiting supply mechanism that most retail summaries missed.

The read-through for future mega-IPO unlocks matters for any retiree who holds broad-market index products. The Facebook 2013, Alibaba 2015, and Uber 2020 unlocks were single-cliff releases that produced the five-to-fifteen-percent single-session moves the analyst notes cited. SpaceX's structure is designed to prevent exactly that scenario. If the next mega-IPO your S&P index fund forced-buys uses a similar staggered structure, the "unlock volatility" event you have been trained to fear may not materialize at scale. The bigger risk moves from unlock day to path dependence: additional tranches unlock only if the stock rises, which creates a self-reinforcing supply ceiling. Something to remember when the next SpaceX-scale name comes to market.

🎯 Actionable Trade Setup

• Watching: SPCX for insider Form 4 filings next week; SPY / VOO for any residual passive-flow effect.

• Bias: Neutral to constructive on SPCX (unlock stress absorbed); awareness on staggered-unlock mechanics for future IPOs.

• Risk note: additional 455.8M shares unlock only if price rises above trigger levels. That creates a supply ceiling on rallies, not a floor on declines.

Five Prints, One Template: The Week That Rewrote Earnings Season.

Zoom out to the full earnings pattern of the past two weeks. Six major AI-adjacent names have printed. Amazon (AWS +37%) got paid. Microsoft (Azure past $100 billion, capex +execution) got paid. Palantir (revenue +93%, U.S. commercial +149%) got paid. Meta (revenue growth, insufficient capex discipline) got sold. SpaceX (revenue beat, $118 billion FY28 capex guide) got sold. AMD (revenue +50%, data center +107%, guidance beat) got sold. That is three-and-three, and the split is not random. The tape has moved from crediting a beat to interrogating whether the capex being spent is producing proportional returns. This is a new pricing framework, and it holds across six independent tests over sixteen calendar days.

What this means for your retirement account is that broad-tech sector ETFs — QQQ, XLK, SMH, SOXX — are now averaging winners and losers at market-cap weight in a way that neutralizes the winners. The rotation that used to be sector-level ("AI stocks are up") is now ticker-level ("some AI stocks are up, others are down, and the ETF gives you both"). The second-order infrastructure names — NextEra Energy, Quanta Services, MasTec, Vertiv, Eaton — continue to collect capex regardless of which hyperscaler clears the bar this week. That thesis strengthened another rung this week and has not broken once across the four prints.

🎯 Actionable Trade Setup

• Watching: NEE, PWR, MTZ, VRT, ETN; XLU and PAVE for ETF-level infrastructure exposure.

• Bias: Bullish structural on infrastructure-tier; neutral on individual hyperscalers; awareness on broad-tech ETF concentration.

• Risk note: if Nvidia's late-August print misses the same template, the whole complex re-rates lower. Infrastructure names sympathy-sell short-term but the thesis holds.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 10: QCD-To-RMD Interaction Under The $40K SALT Cap

A Qualified Charitable Distribution is a direct transfer of up to $108,000 (2026 limit, indexed) from a traditional IRA to a qualified charity, made by an IRA owner age 70½ or older. The QCD counts toward your Required Minimum Distribution but does not appear in your adjusted gross income, which means it does not trigger the phase-outs and tax cliffs that regular RMD-driven income creates. Under the new $40,000 SALT deduction cap for 2025-2028, this interaction becomes materially more valuable. If you were previously taking the standard deduction because your SALT was capped at $10,000, and you now itemize because SALT climbed to $40,000, your charitable-contribution deduction actually pencils again for the first time in six years — but only for donations you make with taxable dollars. A QCD achieves the same charitable outcome without needing to itemize, which is useful in the years you cannot itemize, and it also lowers your AGI, which reduces Medicare IRMAA surcharges and Net Investment Income Tax exposure. Under the current window, running your charitable giving through QCD versus itemized donation is a year-by-year calculation, not a one-time decision.

Why it matters: A retiree taking RMDs, giving to charity, and living in a high-tax state now has three levers that interact: SALT cap, itemization threshold, and QCD election. Optimizing the three together over the 2025-2028 window can save four to five figures across the four-year block that would otherwise be lost to Medicare IRMAA and NIIT.

Oil Unwound. Gold Held. Weekend Positioning Reset.

Oil closed a third consecutive lower session Wednesday and stabilized Thursday around Brent $78, WTI $75. President Trump continued signaling that a Hormuz-reopening deal "could happen tomorrow or the next day" — a statement that did not materialize into a Thursday deal but kept downward pressure on the tape. Gold, in the same session, held above $4,100. The structural-versus-tactical hedge distinction that we flagged two weeks ago has now been tested three times across three different geopolitical inputs, and it has held each time. The oil trade was pricing a specific event risk; the gold trade was pricing something the specific event risk was orthogonal to. Both worked the way they were designed to work.

What this means for the weekend is that if a deal announcement lands over Saturday or Sunday, Monday opens with oil gapping lower and energy ETFs following. If no deal, oil stabilizes near current levels and the trade quietly consolidates. Either way, the retiree who upsized energy exposure specifically as an Iran hedge earlier in the summer has already made most of the money that was available, and holding the position beyond this weekend adds carry risk without matching upside. Right-size the energy sleeve back to your baseline weighting. Hold the gold sleeve unchanged.

🎯 Actionable Trade Setup

• Watching: XLE and CL futures Sunday-night open; GLD, IAU for continuation.

• Bias: Defensive on Iran-hedge energy sleeve; bullish structural on gold.

• Risk note: a weekend deal collapse snaps oil back sharply. Right-sizing means reducing, not eliminating, energy weight.

What To Ignore After The 8:30 Print.

The first sixty minutes after any major economic release on a summer Friday are the least reliable trading window of the week. Volume is often thin, institutional desks are lightly staffed, and reactive orders overshoot in both directions before the actual price discovery happens. For a retiree portfolio, the discipline this weekend is to let the 8:30 tick land, watch the first hour without acting, and if any repositioning is warranted, execute it in the last ninety minutes when the tape has settled. Any weekend reposition should have been planned pre-print with a specific "if-then" framework, not chased after the fact. The market rewards patience on Friday afternoons more consistently than any other window of the week.

🎯 Actionable Trade Setup

• Watching: your own reactive impulses through the first hour after 8:30.

• Bias: Sleeve consideration only. Not a trade — a discipline.

• Risk note: the biggest weekend losses come from repositioning into a Friday close on a print that reverses Monday. Wait until the tape settles.

Trade Cheat Sheet: Fri Aug 7

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Market Delta: Thursday August 6, Pre-Market

Dow closed at record 54,488.68 +0.74% Wednesday; Nasdaq slipped -0.49% to 26,454; S&P flat at 7,741.68
AMD beat Q2 with revenue $11.5B (+50%), data center $6.7B (+107%) — and fell after hours; SPCX 911.5M share unlock hits at today's open
Eli Lilly (LLY) +4% Wednesday on Mounjaro/Zepbound record ~$15B in quarter; raised full-year guidance
Brent crude down to $78 (WTI $75) — third straight session lower — on Trump saying Hormuz deal "could happen tomorrow or the next day"

Today's question: When a billion shares of a mega-cap company hit the market in one morning, what actually happens to price discovery — and why should the average retiree be a spectator instead of a participant?

30-Second Brief

• SpaceX's first insider unlock releases 911.5 million shares this morning — roughly $103 billion of supply, more than doubling the public float overnight. Whatever the tape does today is amplified by pure mechanical selling pressure. See Theme 1.

• AMD beat every number Tuesday night: revenue up 50% year-over-year, data center revenue up 107%. Stock fell in after-hours anyway. That is four-for-four on the "beat plus soft sentiment" pattern this earnings cycle. See Theme 2.

• Eli Lilly delivered a clean beat plus raised guidance Wednesday morning, on Mounjaro and Zepbound generating nearly $15 billion in the quarter. LLY rose four percent and pulled XLV with it. See Theme 3.

• Oil closed the third straight lower session yesterday on Trump signaling a Hormuz reopening "could happen tomorrow." Brent at $78, WTI at $75. The Iran-hedge trade is now a full unwind. See Theme 4.

• Friday's payrolls print at 8:30 AM ET resolves the September Fed hike question. Consensus is 91,000 jobs and 4.3% unemployment, but the internals matter more than the headline. See Theme 5.

The single largest scheduled event in the equity market this quarter lands at this morning's open. SpaceX releases 911.5 million previously locked-up shares — roughly $103 billion of supply at current prices — more than doubling the public float overnight. AMD reported after Tuesday's close with a clean double beat and data center revenue up 107% year-over-year, and then fell in after-hours anyway. Wednesday closed with the Dow at a record, the Nasdaq down on chip weakness, and Eli Lilly rallying four percent on a genuinely dominant weight-loss quarter. Trump signaled that a Hormuz Strait deal "could happen tomorrow or the next day." Tomorrow morning at 8:30, the July payrolls print resolves the September Fed trade. Five things, one Thursday. The map is below.

SpaceX's $103 Billion Supply Wall Hits At The Open.

SpaceX's first insider lockup expires at this morning's open, releasing up to 911.5 million previously restricted shares onto the market. At Wednesday's close price around $116, that is roughly $103 billion of potential supply. The prior public float was smaller than the amount now unlocking, which means the total tradeable share count effectively doubles overnight. That is not a normal market condition. It is a pure mechanical event: supply has grown, and buyers have to absorb it before price discovery around fundamentals can resume. Historical precedent from earlier mega-IPO unlocks — Facebook 2013, Alibaba 2015, Uber 2020 — shows single-session moves of five to fifteen percent in the direction of the previous session's sentiment, followed by three to five sessions of stabilization.

What this means for a retiree portfolio depends entirely on how you already hold SpaceX. If you own it only through broad-market ETFs (VOO, SPY, VTI), the index absorbs the price change at market weight and rebalances mechanically over the next month. That is a passive event for you. If you added direct SPCX exposure since June's IPO, this morning is when the position is either validated by fundamentals or reset by supply pressure — and importantly, there is no clean way to trade the unlock in real time without chasing already-priced-in supply. The reasonable posture today is spectator. Read the tape, do not react to it. The stabilization week matters more than this morning's session.

🎯 Actionable Trade Setup

• Watching: SPCX open volume and price action; SPCQ / SSPC (inverse ETFs) for institutional selling signals.

• Bias: Awareness only. Sleeve consideration for meaningful direct exposure holders. Do not add new direct SPCX today; do not trim broad-market ETF exposure over this single event.

• Risk note: attempting to trade around the unlock window without a pre-established position is chasing already-priced supply. Post-unlock stabilization typically takes three to five sessions.

AMD Beat, AMD Sold. The Pattern Is Now Consistent.

AMD's Q2 print Tuesday night was a genuinely strong number by any earlier standard. Revenue reached $11.5 billion — a company record and up 50% year-over-year. Data center revenue alone hit $6.7 billion, up 107% year-over-year. Management beat both revenue and EPS estimates, and issued guidance that landed above consensus. The stock fell in after-hours anyway. That is now the fourth beat-plus-sell response in this earnings cycle, joining Meta, SpaceX, and Palantir's initial pre-print skepticism. The tape is no longer rewarding a beat. It is asking a specific question: "does the capex you are spending, or the growth you are printing, justify the multiple you carry going in." AMD's answer was apparently "not by enough."

What this tells you about the rest of the year is that every AI-adjacent earnings report from here has to clear a bar that keeps rising each week. Nvidia's next print will land against the same template. Broadcom's next print will land against the same template. For retirees who own these names through QQQ, SMH, or SOXX, the sector-level exposure is now inheriting the discrimination happening at the ticker level, and the sector ETFs are averaging losers with winners in a way that neutralizes the winners. If you want AI exposure that keeps working across this rotation, the second-order infrastructure names — the ones that get the capex regardless of which spender clears the bar — remain the cleaner sleeve.

🎯 Actionable Trade Setup

• Watching: AMD open, NVDA setup into its August print, SMH and SOXX for sector-level read.

• Bias: Neutral on individual chip names; constructive on infrastructure-tier (NEE, PWR, VRT).

• Risk note: sector ETF concentration is now working against you as the tape discriminates. Broader-market ETFs neutralize both the winners and the losers.

Lilly Delivered. Concentration Risk Just Paid Off, For This Quarter.

Eli Lilly reported Wednesday morning and delivered what the tape wanted. Mounjaro and Zepbound generated nearly $15 billion in the quarter alone, an all-time record for the weight-loss franchise. Management beat every headline number and raised full-year guidance. The stock rose approximately four percent on the day. For anyone holding XLV or a healthcare sleeve in a target-date retirement fund, that four-percent LLY move pulled the whole ETF with it, because LLY sits at outsized weight inside the healthcare index. The concentration that was a risk going in became a benefit on the print. This is what a retiree portfolio should hope for: outsized exposure to the specific name that delivered.

The forward question is now whether this trajectory holds. Novo Nordisk also reported Wednesday, CVS Health signed a weight-loss deal, and the competitive dynamic in the GLP-1 category continues to intensify. Lilly's guidance suggests management expects continued dominance, but that expectation is now priced. What this means for your IRA: if your healthcare exposure is a passive index sleeve, the LLY concentration will benefit you until it doesn't — and when Novo or a new entrant meaningfully compresses pricing power, you own that risk too. The concentration cuts both ways. Right now, this quarter, it cut in your favor.

🎯 Actionable Trade Setup

• Watching: LLY follow-through; XLV concentration; NVO for competitive read; CVS for distribution-partnership implications.

• Bias: Bullish structural on GLP-1 franchise this quarter; awareness that concentration is a symmetric bet.

• Risk note: XLV's LLY weight benefited you this Wednesday. That same weight will drag you when a competitive print eventually breaks the pricing-power story.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 8: SALT Deduction Cap Raised To $40,000 (2025-2028)

The State And Local Tax deduction cap is now $40,000 for tax years 2025 through 2028 for filers under the income phaseout (modified AGI up to $500,000 single / $600,000 joint). That is a quadrupling of the previous $10,000 cap set by the original TCJA. For a retiree in California, New York, New Jersey, Illinois, or any high-property-tax jurisdiction who carries property tax bills above $20,000 combined with state income tax on retirement-account withdrawals, the itemize-versus-standard-deduction math changed materially with tax year 2025. If you took the standard deduction the last six years because SALT was capped at $10,000 and itemizing did not pencil, run the calculation both ways for tax year 2026 — ideally before you make any large Roth conversion or gain-recognition decision this quarter. For many households in high-tax states, itemizing is now the better answer, and the timing of when you realize taxable income during the 2025-2028 window matters more than it has in a decade.

Why it matters: The expanded SALT cap creates a 2025-2028 window that sunsets back to a lower number after 2028 unless extended. Roth conversions, capital gain recognitions, and RMD-year sequencing all carry a tax-cost difference now that did not exist a year ago. Ask your tax advisor to model this four-year window as a distinct planning block, not as "the current bracket regime."

Hormuz Might Reopen Tomorrow. Oil Already Priced It.

President Trump said Wednesday that a deal to reopen the Strait of Hormuz "could happen tomorrow or the next day." Oil closed the third straight lower session on the statement. Brent settled at $78 per barrel; WTI at $75. The path from the mid-90s Brent print of ten days ago to today's $78 is essentially a full unwind of the Iran-tension risk premium that had been priced through most of the summer. For retirees who added long-oil positions or upsized energy-ETF exposure specifically as an Iran-hedge, the hedge worked when it needed to and is now unwinding as the crisis eases. That means the position stops paying you and starts costing you carry if you leave it upsized.

The gold trade, by contrast, has continued to hold above $4,100 through the same de-escalation. That is the structural-versus-tactical distinction we flagged two weeks ago, and the past ten days have confirmed it. Gold's central-bank buyer does not care about the Iran headline. Oil's speculative long did. Sizing each sleeve based on which buyer is actually driving the price is now more useful than sizing based on "Iran hedge" as a category.

🎯 Actionable Trade Setup

• Watching: XLE, CL futures, GLD, IAU.

• Bias: Defensive on Iran-hedge oil exposure; bullish structural on gold.

• Risk note: a deal collapse snaps oil back sharply. Any right-sizing should reduce, not eliminate, energy weighting.

Tomorrow's 8:30 AM Print Is Where The Fed Trade Resolves.

Friday's July nonfarm payrolls print lands at 8:30 AM ET tomorrow. Consensus expects 91,000 jobs added and unemployment steady at 4.3%. The internals will matter more than the headline. A payrolls beat above 130,000 with unemployment holding gives the three Fed hawks and the nine-of-eighteen dot-plot participants the clean vindication they need to push September hike odds back above 60%. A payrolls miss below 60,000 with unemployment ticking to 4.4% removes the September hike from the table. Anywhere between and the ambiguity persists into Warsh's Jackson Hole speech later this month. The 10-year at 4.67% is priced for hawkish resolution; a dovish print is a compression trade, not a free trade. Your bond sleeve moves on the 8:30 tick, not on the day's later commentary.

🎯 Actionable Trade Setup

• Watching: TLT, IEF, HYG into tomorrow's 8:30 AM ET release; the 10-year yield print itself.

• Bias: Defensive long-duration going into the print; sized as duration management.

• Risk note: binary event with two-sided asymmetry. Any position sized as a directional bet on Friday is asymmetric against you.

Trade Cheat Sheet: Thu Aug 6

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Market Delta: Wednesday August 5, Pre-Market

SpaceX (SPCX) beat both revenue and EPS after Tuesday close — and dropped -7.46% after-hours to $115.98 on FY28 capex guide of $118.4B
Eli Lilly (LLY) reports this morning; consensus EPS $6.55, options priced for a 6.9% single-session move
JOLTS June came in at 7.36M openings vs 7.4M expected; Fed hike odds for September sit near 50% into Friday's payrolls
10-year yield holding near 4.67%; AMD reports after close tonight; SPCX 911.5M share unlock lands Thursday morning

Today's question: When a company beats every earnings number and still drops seven percent after hours, what is the tape actually asking for — and which companies in your portfolio might be about to fail that same test?

30-Second Brief

• SpaceX beat Q2 revenue and posted a smaller loss than expected, then declined 7.46% after hours on FY28 capex guidance of $118.4 billion. Wall Street is now punishing high capex without matching execution. See Theme 1.

• The pattern from the past two weeks is clear: Amazon and Microsoft got paid for capex + execution. Meta and SpaceX got sold for capex alone. The rotation is at the ticker level, not the sector level. See Theme 2.

• Eli Lilly reports this morning with consensus EPS around $6.55 and options pricing a 6.9% swing. If you hold XLV or a healthcare sleeve, this print rolls straight through your Wednesday close. See Theme 3.

• JOLTS came in slightly below consensus at 7.36 million openings but the internals were mixed. September hike odds hover near 50%. Friday's payroll print is where the trade actually resolves. See Theme 4.

• Tomorrow morning: SpaceX's first insider-share unlock releases up to 911.5 million shares — roughly $103 billion in supply at current prices. Whatever the tape does today, Thursday is when actual price discovery happens. See Theme 5.

SpaceX beat every number Wall Street had estimated for its first-ever public quarterly last night. Revenue landed at $6.81 billion. The loss came in at nine cents per share against consensus for a twenty-four cent loss. Regular session Tuesday closed up 9.43% at $125.33. And then the after-hours session opened, digested the FY28 capex guide of $118.4 billion, and marked the stock down 7.46% to $115.98. That is a company that beat and got punished. It is also the third such print in three weeks — Meta and, on some readings, Microsoft's initial reaction — and it tells you what the tape is actually pricing right now. The bar this earnings season is not revenue. It is capex per dollar of proven execution. The map for today is below.

SpaceX Beat, And The Tape Sold It Anyway.

SpaceX's Q2 print was a genuine beat by every standard measure. Revenue $6.81 billion, matching the $6.82 billion consensus. Loss of nine cents per share versus expected loss of twenty-four cents, a beat by roughly 60%. Adjusted EBITDA came in near $2.0 billion. Musk's letter to shareholders raised the timeline for reaching $1 trillion in annual revenue to 2030 — a full year ahead of previous guidance — with a "non-zero chance" of hitting the mark in 2029. On any earlier earnings cycle, that would have been a clean twelve-percent-up print. Instead, the after-hours session marked the stock down 7.46% because the same shareholder letter guided fiscal-2028 capex to $118.4 billion, up from $48.7 billion this year.

The reason this matters for a retiree portfolio is that the market has quietly moved the goalposts for what an "AI-adjacent" earnings beat has to look like. Revenue beats no longer clear the bar on their own. What the tape now wants is revenue growth plus capex discipline plus proof that the capex being spent is producing returns. Amazon's AWS +37% cleared that bar. Microsoft's Azure execution cleared it. Meta's revenue growth without matching capex-return proof did not clear it. SpaceX's beat with a capex ramp did not clear it. The pattern is now consistent enough across four prints that it is not noise. It is the new pricing framework, and any stock you hold that is about to report AI-adjacent numbers is being read against this template.

🎯 Actionable Trade Setup

• Watching: SPCX into today's open and through Thursday's unlock; META, GOOG for read-across; AMD after close tonight against the same template.

• Bias: Neutral direct SPCX; awareness-only for index-holders through Thursday's unlock resolution.

• Risk note: if AMD tonight beats revenue but guides capex higher without execution proof, expect a similar after-hours sell response — and Nasdaq broader pressure Thursday morning.

The Capex Punishment Rotation Has A Beneficiary.

The four-print pattern of the past two weeks does not just punish the losers. It creates a specific rotation, and the beneficiary is the second-order infrastructure name. If the market is going to reward AWS execution and punish SpaceX capex, both readings still assume the actual dollars are being spent on data centers, power, cooling, and short-haul transmission. Microsoft's fiscal-2027 capex guidance stepped up to $255-260 billion. Amazon's AWS growth of 37% implies its own capex step-up will follow. Add SpaceX's $118.4 billion FY28 guide, and hyperscaler-plus-infrastructure capex is now tracking toward $800-plus billion annually. That entire flow lands in the same set of companies regardless of which spender the tape is rewarding this week.

The names are the ones we discussed a week ago: NextEra Energy, Quanta Services, MasTec, Vertiv, Eaton. They are boring in a very specific and deliberate way — they build the physical infrastructure that everyone else's AI story requires. What has changed in the past week is that the case for owning them has strengthened, not weakened. The tape is now doing the discrimination job at the hyperscaler level. Below that, the money still has to be spent. Retirees who own these names indirectly through utility or S&P infrastructure ETFs (XLU, PAVE) are getting exposure that keeps working across both winners and losers of the capex debate.

🎯 Actionable Trade Setup

• Watching: NEE, PWR, MTZ, VRT, ETN into today's open; XLU and PAVE for the ETF-level positioning.

• Bias: Bullish structural on infrastructure-tier names. The capex debate at the hyperscaler level does not change the fact that the capex still has to physically land in these companies' order books.

• Risk note: if the broader Nasdaq weakens on Thursday's SPCX unlock, infrastructure names may sympathy-sell in the first hour. That is a buying window, not the trade thesis breaking.

Lilly Reports This Morning. Check Your XLV Weight Before The Print.

Eli Lilly reports Q2 results before this morning's open. Consensus EPS sits at approximately $6.55. Options markets are pricing a 6.9% single-session move on the print — unusually wide for a mega-cap healthcare name, which reflects both the growth-versus-competition question in the weight-loss category and the sheer weight LLY carries inside the S&P 500 Health Care sector. Lilly is the single largest position in XLV and one of the top-weighted names in almost every broad healthcare index product a retiree portfolio might hold. A seven percent single-session move in LLY drags the whole ETF, drags the whole sector reading in target-date funds, and can push into broader-market indexes given LLY's overall S&P weight.

What the tape is really pricing into the print is the trajectory question, not the current quarter. Weight-loss drugs are on pace to generate roughly $13 billion in the quarter alone. Does the market keep expanding at the 2024-2025 pace, or does Novo Nordisk's competitive response plus emerging generic-adjacent alternatives compress Lilly's pricing power? Management will guide against that framework. If the guidance holds the trajectory, XLV holders participate in the upside. If the guidance suggests peak pricing power has arrived, a lot of retirees who wandered into "safe healthcare" are about to learn how much of their sleeve is actually one name.

🎯 Actionable Trade Setup

• Watching: LLY at open; XLV and target-date funds for concentration passthrough; NVO for competitive read.

• Bias: Neutral into print; concentration-awareness for ETF holders. Not a "buy or sell Lilly" question. A "how much of my healthcare exposure is actually Lilly" question, checked before the print, not after.

• Risk note: LLY is heavy enough in the S&P that a 7% move contributes to broader-index direction. Position that fact into any planned Wednesday sleeve trims.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 15: The HSA Triple-Tax Advantage

Health Savings Accounts are the only account in the U.S. tax code that gives you three tax advantages in one vehicle: contributions are tax-deductible on the way in, growth is tax-free while inside, and withdrawals for qualified medical expenses are tax-free on the way out. Under OBBBA, the HSA contribution limits for 2026 sit at $4,300 for self-only coverage and $8,550 for family coverage, with an additional $1,000 catch-up contribution allowed for account holders age 55 and older. Unlike a Flexible Spending Account, HSA balances roll over year after year and can be invested, meaning a disciplined HSA holder can build a six-figure account that functions as a dedicated healthcare-cost sleeve inside their overall retirement plan. After age 65, HSA withdrawals used for non-medical expenses are taxed as ordinary income (no penalty), which effectively converts an HSA into a second traditional-IRA-style account with a healthcare escape hatch.

Why it matters: If the weight-loss drug category continues at current prices, out-of-pocket healthcare costs for retirees will keep rising. An HSA is the one tax-preferred vehicle that lets you save for that cost with pre-tax dollars, grow the balance untaxed, and spend it untaxed. For a household that is HDHP-eligible and has not maxed out the account, this is the highest-return tax move available to a retiree that has not been closed by rate changes.

Why Yesterday's JOLTS Didn't Change The Friday Trade.

JOLTS came in at 7.36 million June job openings against consensus of 7.4 million. That is a small miss. The May revision brought that month's opening figure to 7.54 million. On the surface, this is dovish data — fewer openings suggest cooling labor demand and lower urgency for a Fed hike. But the internals were split. Hiring strengthened even as openings fell, which is the pattern of a labor market where employers are converting existing job listings to actual hires rather than posting new ones. That is not the same signal as a genuinely weakening labor market, and Fed hawks will read it that way. September hike odds barely moved on the release — still hovering near fifty percent according to fed funds futures.

What this means for the bond side of your portfolio through the rest of the week is that Friday's payroll print remains the resolving data point. A payroll number below 60,000 with an unemployment tick to 4.4% takes the September hike off the table. A payroll above 130,000 with unemployment steady at 4.3% pushes odds back above 60%. Anywhere in between and the ambiguity persists into Jackson Hole later this month. What this means for your IRA: the 10-year at 4.67% is priced for the ambiguity to resolve toward hawkish, not toward dovish. A dovish resolution is not free money for long-duration — it is a compression trade. Sized as duration management, not as directional bet.

🎯 Actionable Trade Setup

• Watching: TLT, IEF, HYG into Friday's 8:30 AM ET payroll release.

• Bias: Defensive long-duration; short-duration constructive; floating-rate income sleeve holds.

• Risk note: single data-point resolution is a two-sided trade. Any duration position sized as a directional bet on Friday is asymmetric against you.

The $103 Billion Unlock Lands Tomorrow.

Tomorrow morning, SpaceX's first insider-share unlock releases up to 911.5 million shares — roughly $103 billion of supply at current prices, and more than doubling the public float overnight. Post-IPO unlock events at this scale have historically produced single-session moves of five to fifteen percent independent of company fundamentals, because the pure supply expansion overwhelms any real-time price discovery. The direction of that move usually amplifies whatever the previous session established. SPCX went into today's open down 7.46% after-hours from its earnings print. Thursday's unlock, coming after Wednesday's regular session, will do whatever Wednesday sets it up to do — only more so. This is the day to be a spectator, not a participant.

🎯 Actionable Trade Setup

• Watching: SPCX into today's close; SSPC and SPCQ (inverse ETFs) for pressure signals.

• Bias: Awareness only. Sleeve consideration for anyone with meaningful direct SPCX exposure.

• Risk note: attempting to trade the unlock without a pre-established position is chasing supply that already priced in. Post-unlock stabilization typically takes three to five sessions.

Trade Cheat Sheet: Wed Aug 5

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Market Delta: Tuesday August 4, Pre-Market

All three major indexes closed at record highs Monday: S&P 500 +1.12% to 7,573.53, Dow +1.01% to 53,014.36, Nasdaq +1.69% to 25,801.98
Palantir (PLTR) +15% after-hours on 93% Y/Y revenue growth and raised full-year guidance; SpaceX (SPCX) reports tonight after close
Oil sliding as Trump signals Iran negotiation over attack; ISM Manufacturing printed 55.6 for July, well in expansion
10-year yield holding near 4.67%; JOLTS due this morning as first labor-market read ahead of Friday's jobs report

Today's question: When Palantir prints 93% revenue growth and raises guidance, what has to be true about SpaceX's numbers tonight for the stock to justify a $1.4 trillion valuation — and what happens Thursday morning if it isn't?

30-Second Brief

• Palantir reported Q2 after Monday's close: revenue +93% Y/Y, U.S. commercial +149%, FY 2026 guidance raised to 82%. Stock up roughly 15% overnight. The bar just moved for every AI-adjacent print this week. See Theme 1.

• SpaceX (SPCX) reports its first-ever public quarterly tonight after close. Consensus revenue ~$6.82 billion, but the number the tape trades on is Starlink ARPU. The $103 billion insider unlock Thursday amplifies the setup. See Theme 2.

• Oil slid Monday as President Trump signaled calling off Iran military action to negotiate a deal. Gold held above $4,100 through the de-escalation — the structural bid thesis remains intact. See Theme 3.

• JOLTS job-openings data drops today, the first labor-market read ahead of Friday's payroll report. Fed funds futures still price a 64% September hike; the June dot plot showed 9 of 18 FOMC participants penciling in a 2026 hike. See Theme 4.

• Three record closes in one session. VIX near multi-month lows. SpaceX print tonight. The setup for cheap portfolio insurance against expensive concentration risk is as clean as it has been in months. See Theme 5.

Palantir reported after the close Monday and posted the kind of number that changes what everyone else in the AI complex has to deliver. Revenue up 93% year-over-year. U.S. commercial revenue up 149%. Full-year guidance raised to 82% revenue growth. Stock up roughly 15% in after-hours. Meanwhile the three major indexes each closed at record highs on Monday's regular session, oil slid on a Trump statement about pausing Iran military action to negotiate, and ISM Manufacturing came in at 55.6 — comfortably in expansion. Into all of this, SpaceX prints its first-ever public quarterly after the close tonight. The market that was already priced for perfection just got a benchmark that says perfection isn't good enough. The map for today is below.

Palantir Just Told You What SpaceX Has To Beat Tonight.

Palantir's Q2 print is the cleanest AI-services number the market has seen this cycle. Revenue grew 93% year-over-year. U.S. commercial revenue, which is the "AI is landing in real enterprise workflows" line item, grew 149%. Management raised the full-year revenue guide to 82% growth, and U.S. commercial guidance to 134%. The stock reacted the way you would expect — up roughly 15% in after-hours to trade above $144. But the story the print tells that matters for the rest of the week is that the bar for every AI-adjacent earnings report has just been raised. Wall Street was already skeptical of the $32 billion Palantir valuation heading in. Palantir met that skepticism with numbers that made the valuation look reasonable.

For SpaceX tonight, the read-through is uncomfortable. SPCX carries a roughly $1.4 trillion valuation. Consensus expects Q2 revenue near $6.82 billion, but the wide analyst dispersion — EPS estimates range from a loss of $1.26 to a profit of $0.33, with 12-month price targets from $62 to $800 — tells you the market has no clean view of what to expect. Palantir just showed that when an AI-adjacent name delivers a clean beat with raised guidance, the tape rewards it. It also, implicitly, showed the consequence of missing: a name at a stretched multiple with disappointing execution has nowhere to hide. SpaceX has to beat both revenue and guidance to hold the tape. Anything short opens the trap door for Thursday's unlock.

🎯 Actionable Trade Setup

• Watching: PLTR follow-through today; SPCX into the 4 PM ET print; XLK for sector-wide read.

• Bias: Constructive on PLTR post-print for holders; neutral-with-downside-risk on SPCX into tonight. Do not add new SPCX before the print — the setup is asymmetric.

• Risk note: if PLTR fades from its after-hours highs into Tuesday's regular session, that is a sentiment tell — the market is not willing to pay for even validated AI-services execution at these multiples. SPCX tonight becomes even more fragile in that scenario.

SpaceX Tonight: The Numbers That Actually Move The Stock.

SPCX closed Monday near $112, roughly half its post-IPO peak of $225.64. Wall Street consensus expects Q2 revenue around $6.82 billion, up from $4.7 billion in Q1. Starlink generated $11.4 billion of revenue and $4.4 billion of operating income in calendar 2025 — margins that more than doubled year-over-year. AI capital spending reached $7.72 billion in Q1 2026 alone, with Anthropic paying approximately $1.25 billion per month for compute through 2029. Launch operations added roughly $5.2 billion in 2025 revenue on 40 orbital launches in Q1 alone. Every one of those lines has its own tell, and the market will read them in real time on the earnings call.

The single number that will drive the after-hours move most is Starlink average revenue per user. It has slid from around $99 per month in 2023 to roughly $66 per month in Q1 2026. If Q2 shows ARPU stabilizing above $65, the volume-versus-pricing debate resolves in favor of "management is managing the mix." If ARPU prints below $60, the story becomes "growth is coming at the cost of pricing power," and the $1.4 trillion valuation loses one of its core supports. Two trading days after tonight's print, up to 911.5 million shares unlock — roughly $103 billion worth at current prices. Whatever the tape does tonight, Thursday morning is when the actual price discovery happens.

🎯 Actionable Trade Setup

• Watching: SPCX into 4 PM ET; monitor Starlink ARPU, U.S. commercial guidance, Anthropic contract renewal language.

• Bias: Neutral into print; sleeve-only awareness for index-holders. Direct SPCX exposure carries asymmetric event risk through Thursday.

• Risk note: Starlink ARPU below $60 is the "sell first, read second" line. Above $65 with beat-and-raise guidance clears the bar Palantir just set.

The Iran Trade Just Flipped Without A Headline That Matched.

Oil sold off Monday on a Trump statement that he was calling off a military action against Iran to negotiate a deal instead. Brent that had been near $92 slid into the Monday close. On the surface, that reads like a straightforward de-escalation trade. What is more useful for a retiree portfolio is what didn't move in response. Gold held above $4,100 through the news. The central bank buying that drove gold there was structural, not reactive to any specific geopolitical event, and it absorbed a de-escalation the same way it would have absorbed a re-escalation. This is the exact test the gold thesis needed. It passed.

What this means practically is that the two hedges most retirees hold as "Iran insurance" — long-oil positions through XLE or specific energy names, and gold through GLD or IAU — are now doing different things. The oil sleeve was tactically hedging a geopolitical spike that has now paused; that hedge unwinds. The gold sleeve was structurally hedging fiscal and reserve-currency questions that never depended on Iran in the first place; that hedge holds. If your energy weighting was upsized specifically because of Iran, this week is the week to look at it. If your gold weighting is where you want it, you have less to reconsider than the headlines suggest.

🎯 Actionable Trade Setup

• Watching: GLD, IAU, XLE, CL futures.

• Bias: Bullish structural on gold (thesis validated by de-escalation test); defensive on tactical Iran-hedge oil exposure.

• Risk note: a sudden re-escalation on either side flips both trades quickly. Sized as sleeve rebalancing, not directional bet.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 3: 401(k) Plans Now Open To Private Equity And Alternative Assets

OBBBA changed the rules that historically kept 401(k) plans out of private-equity, private-credit, and alternative-asset positions. Plan sponsors can now offer participants access to alternative-asset allocations inside their standard 401(k) menu, subject to fiduciary standards and specific liquidity provisions. In practical terms, this means the "pre-IPO exposure" question — the one that used to require a Wall-Street relationship or accredited-investor status — is opening up for ordinary retirement account holders. A company like SpaceX, which spent years as the largest privately-held name most retirees could not directly own, went public in June. But the next SpaceX-scale private company sitting on retirement-account waiting lists is now potentially reachable through your 401(k), if your plan sponsor adds the option. Ask your plan administrator whether alternative-asset sleeves are being evaluated for 2026 open enrollment.

Why it matters: The forced-buying mechanics that repriced SPCX at IPO in June happen faster when index funds have to buy at scale. Being positioned in the private company BEFORE the IPO through a 401(k) alternative-asset sleeve captures the value that gets transferred at listing — instead of just paying it as an index-fund buyer.

What JOLTS Tells You Before Friday's Jobs Report.

The JOLTS report — Job Openings and Labor Turnover Survey — drops this morning. JOLTS is the Bureau of Labor Statistics measure of open positions employers are trying to fill, plus hires, quits, and layoffs. It is a look at labor demand two weeks before the payroll report. For the Fed's three hawkish dissenters and the nine of eighteen FOMC participants who penciled a 2026 hike into the June dot plot, JOLTS is the first fresh data point they can point to. A job-openings number well above the roughly 7.4 million recent trend supports the hike case. A print below 7.0 million pushes the September hike odds down from the current 64% priced by Fed funds futures.

What this means for your IRA over the next four sessions is that bond fund duration risk is set to move whichever direction JOLTS and Friday's payroll report resolve toward. TLT (the 20+ year Treasury ETF) has been the pressure point for most retirees. If JOLTS surprises high and Friday's payrolls print above 130,000, the 10-year likely tests 4.75%. If JOLTS surprises low and Friday's payrolls print below 60,000, the 10-year could retreat toward 4.55%. In either case, the move happens before you can react to it on Friday alone. This morning's JOLTS is the first piece of the setup.

🎯 Actionable Trade Setup

• Watching: TLT, IEF, HYG; the 10-year yield reaction to JOLTS at 10 AM ET.

• Bias: Defensive on long-duration Treasury exposure into JOLTS and Friday payrolls. Match your bond-fund duration to the compensation you are being paid to hold.

• Risk note: any single data point can surprise either direction. The two-print setup (JOLTS today, NFP Friday) is where the trade actually resolves.

The Record-Close Setup Nobody Is Hedging.

Monday closed with all three major indexes at record highs on the same session, an ISM Manufacturing print in expansion territory, and Big Tech leading the rally. The VIX — the market's fear gauge — sits near multi-month lows, which means portfolio insurance is priced as if the setup is stable. Tonight SpaceX prints. Thursday $103 billion of shares potentially unlock. Friday delivers the July jobs report. Any one of those three events could move the tape three percent in a session; all three landing in the same 96 hours makes for asymmetric downside risk against exposures that are at record highs. Cheap insurance against expensive concentration is the trade that almost never looks needed until it does.

🎯 Actionable Trade Setup

• Watching: VIX, 1-month SPY puts, hedged-equity ETFs.

• Bias: Sleeve consideration only. Not a directional bet against the tape; a "does my downside protection match my upside exposure" question this week specifically.

• Risk note: long-vol positions cost carry. Sized as portfolio insurance against three specific events in four days, the premium is the price of the option, not a failed trade.

Trade Cheat Sheet: Tue Aug 4

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Market Delta: Monday August 3, Pre-Market

S&P futures firmer pre-open as earnings optimism outweighs rate jitters into blockbuster week
SpaceX (SPCX) trading near $108 into Tuesday's first-ever public earnings; 20% insider unlock two sessions later
10-year yield holding 4.67%; Fed funds futures still price 64% September hike
Brent near $92 on cooling Iran tensions; oil desks flag long-Iran trade "getting trickier" into new week

Today's question: When five different catalysts land in the same five days, which two actually change what a retiree portfolio should hold — and which three are noise you can skip without missing anything?

30-Second Brief

• SpaceX (SPCX) reports its first-ever quarterly as a public company Tuesday after the close. Consensus revenue $6.82 billion. The number the tape will trade off is not revenue — it's Starlink subscriber growth and ARPU. See Theme 1.

• Palantir (PLTR) reports Tuesday night as well. Two AI-story earnings in the same session gives us the cleanest read on services-AI vs infrastructure-AI in months. See Theme 2.

• Eli Lilly (LLY) prints Wednesday, with weight-loss drugs on pace for roughly $13 billion in the quarter. Any healthcare-ETF-holder has LLY at outsized weight. See Theme 3.

• The Fed's June dot plot showed nine of eighteen participants penciling in at least one 2026 rate hike. Wednesday's three-way hawkish dissent was consistent with a much broader hawkish tilt on the committee. See Theme 4.

• Jobs report Friday: consensus 91,000 payrolls, unemployment 4.3%. The print resolves whether Warsh has the labor-market slack he wants to leave rates alone into Jackson Hole. See Theme 5.

Five separate catalysts land in the next five sessions. SpaceX prints its first-ever public quarterly Tuesday night, alongside Palantir on the same day. Eli Lilly reports Wednesday, with the weight-loss story either extending or breaking. AMD prints the same day, giving the second read on hyperscaler capex landing at the chip end. Friday closes the week with the July jobs report — the first labor-market data the Fed's hawks can use to justify or reverse their case for a September hike. Most retirees do not need to trade every one of these. Some of them change what your portfolio should hold. The other ones are noise. The map for the week is below.

SpaceX Prints Tuesday. This Is Not Nvidia Week Redux.

SpaceX (ticker SPCX) reports its first-ever public quarterly Tuesday after the close. Wall Street consensus is looking for roughly $6.82 billion in Q2 revenue, up from $4.7 billion in Q1. That is a beat-or-miss headline the tape will react to for one session. The number that matters longer than one session is Starlink subscriber growth — 10.3 million at year-end 2025 from 8.9 million a year earlier — combined with Starlink's average revenue per user, which has slid from around $99 monthly in 2023 to roughly $66 in Q1 2026. If subscribers keep growing but ARPU keeps dropping, the growth story that supports a $1.4 trillion valuation starts to look like a volume story pretending to be a pricing story.

Two trading days after SPCX prints, twenty percent of eligible insider and employee shares unlock. That is mechanical selling pressure regardless of the earnings outcome. If Tuesday's print beats and management guidance is credible, the unlock is absorbed. If Tuesday's print misses or the Starlink ARPU decline accelerates, the unlock lands into an already-weak tape and the reflexive selling compounds. For a retiree who holds SPCX only through broad-market ETFs, this is a pass-through event — the ETF absorbs whatever price the tape prints. For any retiree who added direct SPCX exposure since the IPO, this week is when the position is either validated or reset.

🎯 Actionable Trade Setup

• Watching: SPCX into Tuesday's close and the 48 hours after; SPY / VOO for index-inclusion effect.

• Bias: Neutral into the print for direct holders; awareness-only for index-holders. The 20% insider unlock two sessions later is the asymmetric event, not the print itself.

• Risk note: if Starlink ARPU trends below $60 in the reported quarter, the multiple compression that follows will not be a one-session move. Size direct exposure accordingly.

Palantir Same Night. Services-AI Meets Infrastructure-AI.

Palantir (PLTR) also reports Tuesday after the close. That gives the market two AI-story earnings in the same session — SpaceX on the infrastructure side (Starlink, launch, connectivity), Palantir on the services side (data operations for the government-and-large-enterprise buyer). Last week's earnings gave us the hyperscaler capex numbers — Microsoft's fiscal-2027 capex guide at $255-260 billion, Amazon's AWS growing 37%. Tuesday tells us whether that capex is actually landing in the AI-services layer that the retiree portfolio owns most often through diversified tech ETFs.

What to read for is the language Palantir uses about government contract renewals and commercial-segment growth. If Palantir shows accelerating commercial revenue and confirms multi-year government contracts, the read-through is bullish for the whole AI-services complex — and for the infrastructure names funded by that spend. If Palantir shows commercial deceleration, the market will start pricing an AI-services glut, and the hyperscaler capex boom becomes a supply-side story with no clear demand pull. Same week. Same complex. Two very different trades.

🎯 Actionable Trade Setup

• Watching: PLTR, AMD (Wednesday), plus SMH / SOXX for chip-sector read-through.

• Bias: Neutral into Tuesday's print; look for hyperscaler-benefiter language in the call. Palantir's commercial-vs-government mix on the guidance is the tell.

• Risk note: PLTR has a persistent retail-hype tail that can push single-session moves double what fundamentals justify. Don't chase either direction in the first hour after print.

Eli Lilly Wednesday. The Weight-Loss Print That Matters For Your IRA.

Eli Lilly (LLY) reports Wednesday morning, and the weight-loss drug segment is on pace to bring in roughly $13 billion in the quarter alone. Options-market pricing implies a 6.9% single-session move on the print, which is unusually wide for a mega-cap healthcare name. The reason the print carries this much weight is that Lilly is the single largest position in the S&P 500 Health Care sector ETF (XLV) and one of the top holdings in most broad healthcare index products. If you hold XLV, or a healthcare sleeve inside a target-date retirement fund, Lilly's earnings result flows directly through to your Wednesday close.

What the tape is really pricing into Wednesday is not just the current quarter. It is the trajectory question: does the weight-loss market keep expanding at the pace 2024 and 2025 suggested, or does the competitive entry from Novo Nordisk and the emerging generic-adjacent options start to compress the pricing power? Lilly has to guide to a market that stays big enough to support a valuation that is already priced for continued dominance. The risk-reward around the print is asymmetric in a way that matters for anyone who wandered into a "safe healthcare ETF" and did not check the LLY concentration on the way in.

🎯 Actionable Trade Setup

• Watching: LLY, XLV, plus NVO for competitive-read-through.

• Bias: Neutral into print; concentration-awareness for ETF holders. Not a "buy or sell Lilly" question — a "how much of my healthcare exposure is actually Lilly" question.

• Risk note: a 7% single-session move in a top-5 S&P holding will drag the whole index. If LLY misses hard, expect broader-market weakness Wednesday afternoon regardless of AMD's after-close print.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 9: RMD Strategy Under Permanent TCJA Brackets

Required Minimum Distributions from traditional IRAs and 401(k)s used to be a "take what the IRS makes you take" event with limited planning flexibility. Now that the Tax Cuts and Jobs Act brackets are permanent under OBBBA — the 10%, 12%, 22%, 24%, 32%, 35%, and 37% structure — the multi-year RMD planning horizon changes. Instead of treating each RMD year in isolation against a "sunset that may come," retirees can now plan a fifteen- or twenty-year distribution schedule against a stable bracket regime. That opens strategies that were previously too uncertain to commit to: partial Roth conversions in the 22% bracket to reduce future RMD size, spreading capital gains harvests across multiple years to stay under the 32% threshold, and timing charitable distributions (QCDs) against the years an RMD would otherwise push into a higher bracket. None of these strategies work in a one-year snapshot. They work only against a multi-year plan, which the permanent brackets now let you build.

Why it matters: With the 10-year Treasury at 4.67% and the Fed openly hawkish, higher-for-longer rates change the after-tax yield on bond-heavy portfolios. RMD sequencing under permanent brackets is one of the few levers still fully under a retiree's control that can meaningfully move the after-tax outcome over a decade.

The Dot-Plot Number That Was More Hawkish Than The Vote.

Wednesday's three-way hawkish dissent from the FOMC was not the whole picture on how the committee actually leans. The June dot plot — the individual projections each FOMC participant submits alongside the meeting decision — showed nine of eighteen participants penciling in at least one rate hike for 2026. That is half the committee. The three dissenters last week were simply the most public voices for a position the majority of the committee already carries privately. Fed funds futures now price a 64% probability of a September hike, up from around 20% a month ago, which means the market has already absorbed roughly two-thirds of the way toward pricing the dot plot's median view.

What this means for a retiree portfolio is that the risk premium on long-duration Treasury exposure just moved. If the majority of the FOMC actually acts on the hike, the 10-year moves higher and TLT-type positions mark down again. If they hold, the position pays as it stands. Either way, the risk is asymmetric — you are paying a duration premium against a committee that is telegraphing higher rates. What this means for your IRA: check the duration of any bond fund labeled "total return" or "aggregate." If the duration is above six years and you cannot articulate why, the position is probably drifting past what you are being paid to hold.

🎯 Actionable Trade Setup

• Watching: TLT, AGG, BND, plus HYG for credit read.

• Bias: Defensive on long-duration; neutral on 3-7 year; constructive on short-duration Treasury and floating-rate. Match the duration to the compensation.

• Risk note: a soft Friday jobs report (below 60K payrolls, unemployment 4.4%+) would reverse this trade quickly. The Fed hawks lose their justification if the labor market cracks.

What To Ignore This Week.

Five catalysts in five days does not mean a retiree portfolio needs five trades. The two prints that actually change portfolio construction for a 55-to-75-year-old holder are Friday's jobs report (because it moves the bond sleeve) and Wednesday's Eli Lilly print (because it moves the healthcare concentration inside index products). SpaceX is a direct-exposure decision, not an index-holder event. Palantir is a signal read for AI-services allocation, not a required trade. AMD is a chip-cycle datapoint, useful for context but not a portfolio-mover for most retirees. The discipline this week is not "which trade do I take." It is "which three catalysts do I read the recap of without doing anything."

🎯 Actionable Trade Setup

• Watching: your own bond-duration and healthcare-ETF concentration this weekend.

• Bias: Sleeve consideration only. Discipline is not doing more; discipline is knowing which reports demand action and which are context.

• Risk note: attention has a cost. Every catalyst you actively trade this week is a catalyst you gave account energy to that could have been better spent on the two that actually matter.

Trade Cheat Sheet: Mon Aug 3

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

This week ended a trade that has driven every retirement account with equity exposure for the last eighteen months. The Federal Reserve held rates at 3.50% Wednesday, and three of its officials dissented in favor of a hike — not a cut. Fed funds futures now price a 64% probability of a September hike, a number that stood near 20% a month ago. Big Tech earnings, delivered against that backdrop, split the Magnificent 7 into two piles for the first time in a long time. Amazon rallied more than fifteen percent for the week. Apple sold off more than seven. The idea that "you own the index and the tide lifts everything" — the trade that made passive investing look like magic since the last cut cycle — needs new pricing for the second half of 2026.

Further Reading from Bastion Stability

The Week That Killed The Dovish Trade

The Fed's Wednesday hold at 3.50% was consensus. The three-way hawkish dissent was not. Chairman Kevin Warsh, in the press conference, called it a "good family fight" — the phrase choice was deliberate. Meanwhile the June PCE inflation print released Thursday showed core prices easing by 0.1%, with consumer spending up 0.3%. Cool inflation data. Hawkish central-bank dissents. Those two things do not usually happen together, and when they do, it means the majority Fed view and the risk-oriented Fed view are pulling apart. That kind of internal disagreement typically resolves toward the more hawkish side, because dissenters need to justify their vote at each subsequent meeting, and the majority has an incentive to move partway toward them rather than face repeated three-way splits.

The Big Tech tape is telling the same story in earnings numbers instead of policy statements. Amazon reported AWS growth of 37% and rallied 15% for the week. Alphabet added nearly seven percent. Apple beat every headline number but guided Q4 revenue below consensus, citing services softness and continued weakness in Greater China — the stock fell more than seven percent for the week. Microsoft's Azure crossed $100 billion annually and its fiscal-2027 capex guidance stepped up to $255-260 billion. Meta grew revenue 28%, but capex is pressuring profits. Same sector. Same week. Four different outcomes. This is what a market looks like when it stops crediting theme exposure and starts asking who is actually executing.

What this means for your retirement portfolio is that the two positions most retirees added over the last eighteen months — heavier passive index exposure in equity, and longer-duration Treasuries in fixed income — both need to be re-examined. QQQ and SPY averaged Amazon's beat with Apple's guide this week, so the concentration you thought was helping you was neutral at best. TLT and long-duration bond funds got hit as the 10-year held 4.67%. The question a retiree needs to ask this weekend is not "should I sell." It is "am I still being paid for the risk in the specific positions I hold, given that the regime just shifted." Ask that of each sleeve.

📊 By The Numbers — Week Ending Friday July 31

Look-Ahead: SpaceX Prints Tuesday. Jobs Print Friday. Both Matter.

The single largest scheduled event of next week is SpaceX's first quarterly report as a public company, due Tuesday August 4 after the close. The stock — ticker SPCX — trades near $108 today, down roughly 48% from its post-IPO high of $225 earlier this year. Starlink is the revenue engine, printing $11.4 billion in calendar 2025 and growing 50% year-over-year. This will be the first real read on whether the pre-IPO forced-buying cycle that ran through Q2, when index funds were required to purchase SPCX at market weight, has left behind a stock that can hold its bid on fundamentals rather than mechanical demand. The number to watch is Starlink subscriber growth. It hit 10.3 million at year-end 2025 from 8.9 million the year prior. Any material slowing of that trajectory removes the growth story that supports the 80x sales valuation.

Two trading days after SPCX reports, twenty percent of eligible insider and employee shares unlock. That is a mechanical selling-pressure event regardless of whether the earnings print is good or bad. What makes the setup asymmetric: if SpaceX beats and gives guidance the market accepts, the unlock is absorbed quietly by inflows and reinvestment. If SpaceX misses or guides softly, the unlock lands into a weak tape, and the reflexive selling can accelerate the decline. For a retiree who holds SPCX inside a broad-market ETF — which most do by index inclusion at this point — this is the first time actual company performance meets forced-buying mechanics in the same window. The week after August 4 is when we find out which is stronger.

Friday August 7 delivers the July jobs report. Consensus is looking for 91,000 payrolls added, with unemployment at 4.3%. This is the first labor-market data the Fed's three hawkish dissenters can point to as either vindication or ammunition. A print above 130,000 combined with any tick-down in unemployment would give the September hike case its clearest fuel. A print below 60,000 with unemployment at 4.4% or higher would take the September hike back off the table. Either outcome resolves the ambiguity that made this week feel unsettled. Bond funds will move first; equities will follow. What this means for your IRA: if you have not yet checked how much of your bond sleeve sits in long-duration paper, this weekend is the weekend to do it — before Friday resolves the trade one way or the other.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 5: Tax Cuts & Jobs Act Brackets Made Permanent

The 2017 Tax Cuts and Jobs Act brackets — 10%, 12%, 22%, 24%, 32%, 35%, and 37% — which were originally scheduled to sunset after 2025 and revert to higher pre-TCJA rates, are now permanent under OBBBA. That means the current bracket structure is now the baseline for all multi-year planning, not the temporary window it functioned as for most of the last eight years. The permanent status changes the math on Roth conversions, capital-gain harvesting, and RMD sequencing in a specific way: strategies that only worked because you were "beating the sunset" no longer have that urgency, but strategies that leverage the low brackets across a fifteen- or twenty-year retirement horizon just got more attractive relative to income timing.

Why it matters: With the Fed now openly divided on hikes and the 10-year at 4.67%, a longer planning horizon under permanent TCJA brackets interacts with your bond allocation directly. If you had planned tax moves against a 2025 sunset that never came, those plans need to be re-examined this quarter rather than left on the calendar.

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.

Market Delta: Friday July 31, Pre-Market

Amazon +9% after-hours on AWS +37%; Nasdaq futures firmer into the open
Gold +0.1% to $4,112/oz — first close above $4,100 in a month
10-year yield holding 4.67%, near one-year high after Wednesday's Fed
Brent crude near $92 as U.S. pauses Iran airstrikes; Hormuz tanker news adds cross-currents

Today's question: When Amazon and Apple move in opposite directions in the same week, is the market pricing "AI" — or something more specific that your QQQ position doesn't distinguish?

30-Second Brief

• Amazon posted AWS +37% and stock +9% after-hours. Apple beat earnings but guided Q4 soft; shares down roughly six percent. Same week, opposite trades. See Theme 1.

• Microsoft's Azure crossed $100B in annual revenue with a fiscal-2027 capex guide of $255-260 billion. The power-grid names have not fully caught up to this number yet. See Theme 2.

• The Fed held at 3.50% Wednesday, but three officials dissented in favor of a hike — hawkish dissents, not dovish. See Theme 3.

• Gold closed above $4,100 for the first time in a month, even as the U.S. paused airstrikes overnight. The buyer holding the bid is not who most retail thinks. See Theme 4.

• Money market fund yields are still north of 4% while the 10-year sits at 4.67%. The cash sleeve of a retirement portfolio has an earnings yield that briefly rivals what parts of the equity market pay. See Theme 5.

The last day of July closes a week that split Big Tech into two piles the market has not sorted for months. Amazon reported AWS growth of 37% and rallied 9% after-hours. Apple beat every headline number and sold off six percent on soft Q4 guidance. Same sector, same week, opposite tapes. Meanwhile the Fed held at 3.50% Wednesday with three hawkish dissents — the first three-way hawkish break Wall Street has seen since before Powell's exit. Gold quietly held $4,100 through a foreign-policy de-escalation that should have compressed it. And the money market fund sitting in your Fidelity account is paying you nearly 4.5% while equity multiples close at records. The map for today is below.

The Big Tech Split Nobody Was Priced For

Amazon reported Q2 revenue above $200 billion, with AWS growing 37% year-over-year — the fastest cloud growth print of the current cycle. The stock rallied 9% after-hours. Apple, twenty-four hours earlier, delivered a Q3 beat on the top and bottom lines and then guided Q4 revenue below Street consensus, citing softness in services and continued weakness in Greater China. Shares dropped six percent by the close and continued lower through the pre-market Friday. Same week. Both Mag 7. Opposite tapes.

This matters because index-fund exposure — the way most retirees own Big Tech — does not distinguish between AI winners and AI stragglers. QQQ owns both names at their market-cap-weighted share. Your IRA gets Amazon's beat and Apple's guide averaged. For six months, the concentration risk in mega-cap tech was the trade; this week, the concentration became the drag. The market is now asking the question the last two years let investors avoid: which name inside the basket is doing the actual work, and which one is riding the theme.

🎯 Actionable Trade Setup

• Watching: AMZN, AAPL, MSFT, META; QQQ concentration relative to individual names.

• Bias: Neutral on the aggregate; constructive on the split. Discrimination between winners and stragglers within Mag 7 has more room to run in Q3 than it had in Q1 or Q2. Do not add new QQQ into today's open on the assumption "the whole complex is fine."

• Risk note: if chip weakness accelerates further (Nasdaq had chip pressure this week), the concentration trade cuts both ways fast. A basket that averaged out this week averages out both directions.

The $255 Billion Number That Repriced The Power Grid

Microsoft's Q4 FY26 numbers on Wednesday included Azure crossing $100 billion in annual revenue and, buried inside the guidance, a fiscal-2027 capital expenditure range of $255 to $260 billion. The last number is the one that matters. Combined with Amazon Web Services growing 37% — implying its own capex step-up will follow — hyperscaler capex is now trending toward and above $800 billion aggregate for calendar 2026 into 2027. This is no longer an "upside surprise." It is the baseline every serious analyst worksheet is now built on.

The second-order tells are in power generation, cooling, and short-haul transmission. NextEra Energy, Quanta Services, MasTec, Vertiv, Eaton — these are the boring names that have to physically build the infrastructure the AI names use to compute. They are also names that a retiree portfolio often owns indirectly through utility ETFs or S&P infrastructure sleeves rather than at the individual-ticker level. The slow money has been rotating into them for a week; the fast money has not fully caught up.

🎯 Actionable Trade Setup

• Watching: NEE, PWR, MTZ, VRT, ETN into the open.

• Bias: Bullish structural; calibrated on entry. These names are catch-up trades to a hyperscaler capex number that has already been printed, not a bet on the AI theme itself. Sizing question, not timing question.

• Risk note: if the chip weakness in Nasdaq deepens today, infrastructure names may sympathy-sell in the first hour. Better prints, faster catch-up — but not by 10 a.m.

Three Dissents For A Hike. Think About What That Means.

The Federal Reserve held its policy rate at 3.50% on Wednesday, which was the consensus expectation. What was not consensus: three officials dissented in favor of a hike. Chairman Kevin Warsh, in the press conference, called it a "good family fight," a phrase he chose deliberately. Three-way hawkish dissents are rare. The last time the FOMC saw a three-vote hawkish break, the market repriced the whole front end of the curve within two sessions. Meanwhile the June PCE inflation print, released Thursday, showed core prices easing by 0.1%. Consumer spending rose 0.3%. Dovish data. Hawkish dissent.

This gap matters for a retiree portfolio for one specific reason: bond fund duration. The 10-year Treasury closed Wednesday at 4.67%, holding near its one-year high. If Warsh signals openness to a September hike in his next speech — and the dissent pattern says he might — the long end of the curve breaks higher. That marks any long-duration bond fund down instantly, and TLT (the 20+ year Treasury ETF that many retirees own as "safe income") has been the exact position getting hit this cycle. What this means for your IRA: if you hold a duration-heavy bond sleeve, the risk premium on it just went up, and you have not been paid for holding it.

🎯 Actionable Trade Setup

• Watching: TLT, IEF, AGG, HYG.

• Bias: Defensive on long-duration Treasury exposure; neutral on 3-7 year duration; constructive on floating-rate income. The trade is not "sell all bonds." The trade is "match the duration to the risk you are being paid for."

• Risk note: a dovish surprise from Warsh at Jackson Hole in August would compress this trade quickly. Sized as duration management, not directional bet.

🔒 OBBBA Rule Watch: Tax Rule of the Day

Rule 6: New $6,000 Annual Deduction for Filers Age 65 and Older (Tax Years 2025-2028)

The One Big Beautiful Bill Act created a new above-the-line deduction of $6,000 per filer age 65 or older, in effect for tax years 2025 through 2028. For a married couple where both spouses are 65+, that is $12,000 of additional deduction stacked on top of the standard deduction and any itemized deductions they were already claiming. The deduction phases out for higher-income filers (modified AGI thresholds apply), but it fully applies to the retiree profile most likely to be reading this — a household drawing from Social Security, RMDs from traditional accounts, and modest dividend income. Unlike the SALT cap change, this rule is not itemization-dependent — it works with the standard deduction. Which means for many 65+ households, it is a straight cash-value change in their federal tax bill starting with 2025 filings.

Why it matters: This deduction is scheduled to sunset after 2028 unless extended. For Roth conversion planning and RMD-year sequencing, the 2025-2028 window is now a strictly better window to realize larger conversions or capital gains than any year outside it. Ask your tax advisor to run 2025-2028 as a distinct four-year planning block, not as "the current bracket regime."

Gold Held $4,100 Even As Airstrikes Paused

Gold opened at $4,112 on Friday morning Eastern Time, holding above $4,100 for the first sustained session in about a month. The overnight news that would ordinarily have compressed the price was the U.S. pausing airstrikes in Iran. Safe-haven premium should have unwound; instead it did not. The reason is that the buyer holding the bid is not the retail investor reacting to geopolitical headlines. It is central banks, and their pattern of accumulation has been the most consistent it has been in eight years.

The World Gold Council's most recent survey showed a record 43% of central bank respondents planning to increase their own gold holdings over the next 12 months, and 95% expected global official reserves to rise. Central bank gold now represents approximately 20% of official reserves globally, up from 15% at the end of 2023. What this trade is really telling the retiree portfolio is not "gold is going higher next month." It is that the share of your retirement allocation that touches hard assets should be evaluated against the share of central-bank reserves now sitting in them. That comparison is often uncomfortable.

🎯 Actionable Trade Setup

• Watching: GLD, IAU, GDX.

• Bias: Bullish structural; the pause tested the thesis and the thesis held. Not a day-trade thesis. An allocation-size thesis.

• Risk note: a formal ceasefire announcement (not just a pause) would compress gold short-term. The structural bid survives that compression; the tactical bid does not.

The Number You Haven't Checked Yet

The federal funds rate is 3.50%. The 10-year Treasury is 4.67%. Money market fund yields at the major brokerages — Fidelity, Vanguard, Schwab — are still north of 4%, and several of the government-only funds are quoting 4.4 to 4.7% depending on the day. Meanwhile the S&P 500 earnings yield sits around 3.4%. That is the fewest basis points between "cash in a money market fund" and "the earnings yield on the entire equity market" that many pre-retirees have seen in their working lives. The three Fed hawks dissented specifically because they do not want the market to start pricing in cuts. That protects the cash sleeve.

🎯 Actionable Trade Setup

• Watching: your money market position at your primary custodian (SPAXX, VMFXX, SWVXX).

• Bias: Sleeve consideration only. Not a directional call on cash vs equity. A "look at what you are actually being paid on the cash you already hold" call.

• Risk note: when the Fed does eventually cut, this window closes fast. The hawkish dissents this week suggest the window is not closing in September. It is closing eventually.

Trade Cheat Sheet: Fri Jul 31

For informational purposes only. Not investment advice. Past performance is not indicative of future results. Bastion Stability is a daily defensive briefing for retirees and pre-retirees and does not provide individualized tax or investment advice; consult a licensed advisor before making any decision relating to your retirement accounts.