CW Capital Wealth
Risk & The Model Books · Updated August 2, 2026

Which sleeve covers what

The Risk Atlas names eight things that can go wrong. This page maps them onto the books we actually run — tier by tier, sleeve by sleeve, including the squares where our coverage is thin.

A model portfolio is usually presented as a list of things we like. That’s the least interesting thing about it. The useful question isn’t “what do you own,” it’s “when this specific thing goes wrong, what in here is supposed to help?”

So this is the answer, in a grid. Green means a sleeve is explicitly assigned and sized for that risk. Amber means partial — something helps, but not enough to call it covered. Grey means the tier carries that risk essentially unhedged, on purpose, and you should know it.

The Coverage Map

Eight risks against the books

Five of the eight are covered the same way in every tier — so rather than print a grid of green dots, here is only what changes. Each risk, what holds the line, and which tiers carry it anyway. Every tier totals 99.5%, with an explicit Cash / Operational Reserve (SGOV settlement buffer) carrying the remaining 0.5%.

RISK 01MarketCovered

Position sizing itself, plus the 0.5% SGOV operational reserve in every tier. You don’t defeat market risk, you size for it.

RISK 02RateThin — disclosed

SGOV at 4%+. The long end has stopped catching equity drawdowns — the week the Fed held, the 30-year jumped to 5.228%, its highest since 2007, while the Dow fell 1,153 points.

RISK 03InflationCovered

The energy sleeve — CVX XOM COP KMI WMB TPL — plus equity itself, a claim on companies that raise prices.

RISK 04GeopoliticalCovered

Energy sized before the spike, plus defense (LMT RTX AVAV MSI). WTI just ran a $10 round trip in five sessions; the Pentagon just placed a $120B restock order.

RISK 05CreditCovered

The avoid list — the cheapest risk management there is. PARA and WBD refused at 6.5x leverage with bonds at 8.43%; and when Nvidia starts backstopping ~$250B of its biggest customer’s debts, we own the chipmaker’s cash flows, not the customer’s IOUs.

RISK 06ConcentrationPartial

Sleeve construction over single names, plus eight international holdings (ASML NVO SAP TM EWY PBR IBN CPNG).

RISK 07EventCovered

Refusal and position caps. No binary biotech, no merger arb, no “it’ll pop on approval.” We don’t hold coin flips in retirement money.

RISK 08BehavioralPartial

Written rules, decided in advance, when nobody is frightened. That is what a model book is — a pre-commitment device, not a stock-picking exercise.

The Sleeves

What each one is actually for

A sleeve isn’t a theme we like. It’s a job assignment.

Energy & Midstream

CVX · XOM · COP · KMI · WMB · TPL

The insurance sleeve. It exists to be paid while it hedges the two risks that arrive without warning: a barrel that just ran a $10 round trip in five sessions on strikes and talks, and an inflation print that keeps the Fed parked near 3%. Sized before the spike, never added into it.

Covers: Inflation · Geopolitical

Cash & Bills

SGOV · Cash / Operational Reserve (0.5%)

4%+, no drama, settles Tuesday. The only cushion that has never argued with us, and the liquidity source that means a client in a drawdown is never forced to sell at the bottom — the exact behavior the IPS is built to prevent.

Covers: Market · Rate · Behavioral

AI & Semis

NVDA · AVGO · TSM · MU · AMD · MRVL · LRCX · QCOM · SNDK

The growth engine — and the largest single concentration in the books at roughly 20–22% across the Aggressive tiers. We own the sleeve, not one horse, which is why July’s rotation out of the giants and into the suppliers cost us nothing: MU came off the bench, NVDA stayed held, and the sleeve absorbed the switch.

Concentration: monitored, disclosed

Power & Grid

CEG · GEV · VRT · NEE · PWR

The unglamorous half of the AI trade. Data centers need electricity before they need anything else, and the utilities selling it have contracts rather than narratives. Partially decorrelated from the semis sleeve it depends on.

Covers: Inflation · Concentration (partial)

Metals & Royalties

WPM · RGLD · IAU

The sleeve that failed this year. Gold sits roughly 24% below its January record and closed down on the Dow’s worst day of the summer — the one job it had. Held, sized, un-added-to. We don’t sell hedges at the bottom and we don’t average into failed tests. Reported here rather than buried.

Assigned: Inflation · Event — did not deliver

Financials

GS · MS · JPM · BRK.B

Paid on the plumbing regardless of which venue, which deal, or which exchange wins — Goldman traded mock symbols in the Texas Stock Exchange’s practice runs and will collect whether TXSE succeeds or joins the last three “NYSE killers.” Rate-sensitive in both directions.

Covers: Rate (partial) · Event

Defense

LMT · RTX · AVAV · MSI

Budgets have never noticed a peace trade. Roughly 10% of Tier I Aggressive and 6–16% across the family — inside the 20–25% single-sector guideline our own audit applies. The concentrated version lives in Hard Assets at about 29%, where it is the mandate, not a breach.

Covers: Geopolitical — sector cap breached

International

ASML · NVO · SAP · TM · EWY · PBR · IBN · CPNG

Eight individual names rather than a broad ETF base — a deliberate choice to own specific businesses instead of renting an index’s worst decisions. The direct answer to a book that would otherwise be ~100% US equity.

Covers: Concentration · Market (partial)
Full Disclosure

The three squares we’d rather you hear from us

1. The bond sleeve runs at about half our own audit’s recommendation

Our CFP-grounded audit found the Aggressive tiers carried zero fixed income or cash equivalents, against a CFP IN09 guideline of 5–10% even for aggressive growth — as a volatility dampener, a rebalancing source, and a liquidity supply. The recommended fix was roughly 3% TIP, 3% SGOV, 2–4% IEF.

We implemented it at roughly half that size. That is a deliberate call — with the 10-year at 4.621%, the 30-year at its highest since 2007, and bonds no longer reliably catching equity drawdowns, we would rather hold bills that pay us than duration that has stopped doing its job. But it is a choice with a cost, and the cost is that rate risk is the least-covered square on this page. If you want the audit’s full allocation in your account, say so and we’ll build it.

2. Defense concentration — corrected August 4

This flag used to say 25–30% of Tier I Aggressive was defense. That was wrong, and we are leaving the correction visible rather than quietly deleting the line. Tier I Aggressive holds roughly 10% in the defense complex (Lockheed, Palantir, Boeing) — comfortably inside the 20–25% CFP IN09 guideline, and it has been at that weight since the May audit. The 25–30% figure belongs to a different book: Hard Assets, at about 29%, where concentrated defense-and-energy exposure is the stated mandate of a Theme 1 concentrated core, not a breach of a diversified one.

Across the rest of the family the defense complex runs about 6–8% in the Tactical books and 13–16% in the larger Cores — all inside the cap. What we got wrong here was the label on the number, not the number itself. Logged on Marked to Market.

3. AI/Tech is one bet wearing several tickers

Combined AI/Tech exposure runs roughly 20–22% across the Aggressive tiers. Owning nine semiconductor names is not the same as being diversified — it is one thesis, expressed nine ways, and if the AI capex cycle stops, the sleeve stops together. We think the thesis is right. We also think you should know it’s a thesis.

None of these three are secrets, and none of them are emergencies. They’re on this page because the alternative — a model portfolio page that lists only the things we’re proud of — is marketing, and you can get that anywhere.

Want your account run through this grid?

Fifteen minutes, on the phone. I’ll show you which of the eight risks your current allocation has a sleeve for and which ones it’s carrying naked — and what the fix costs. Bring your statement; the fee audit alone usually pays for the call.

Book 15 minutes →