This is our positioning letter — the one the beautiful year-end reports never write. No point target, because we graded a decade of those and they went 0-for-8. What you get instead: the stance, the dividend rotation and what pulls its trigger, the honest answer to “why aren’t you buying bonds?” — and the fixed income we would buy.
What we will do when the bad stretch comes — written on a calm Tuesday.
Left is what the market did. Right is what was already in the books before it did. None of this is a forecast. All of it is pre-committed.
There Will Be A Bad Year. We’re Not Going To Pretend To Know Which One.
Here is the most honest sentence a market letter can print: at some point — maybe this fall, maybe in 2027, maybe later — there will be a genuinely bad year, and we do not know when. Since 1950, midterm years carry the deepest average intra-year drawdown of the four-year cycle, and the Dow just paid 1,153 points in one afternoon for re-reading a Fed statement.
So here is what we will not do. No year-end index target — the industry’s went 0-for-8 since 2018, and we published the receipts. No “this is the year” calls, because drawdowns arrive on their schedule, not ours. No headline trading — the tracker moves daily; the books move monthly or on a ±trigger.
What we know is exactly what we’ll do when the bad stretch arrives, because the moves are pre-built. The calendar play reduces risk into the fall and defends the vote. The cash-and-gold reserve is sized for the worst week before it happens. The eight risks each have a sleeve assigned — and the two that aren’t fully covered are disclosed in public rather than discovered in a drawdown. When we get one wrong, it goes on the ledger in the same type size as the wins.
A plan you write while frightened is not a plan; it is a reaction with better grammar. Everything in this letter was decided on an ordinary week, which is the only kind of week in which good decisions get made.
The test is simple and we invite it: a year from now, check whether the calm-Tuesday plan did its job. That is what the ledger is for, and why the misses are printed at the same size as the wins.
“Go To Dividend Stocks” Sounds Like Panic. Done Right, It’s The Opposite.
The move isn’t “sell everything.” It is one notch down a ladder where every rung already exists as a live book with published holdings.
| The ladder, top to bottom | Realized volatility |
|---|---|
| Aggressive Growth — the full theme, tactical sleeve on | ~20% |
| Income & Quality — paid on a schedule | ~13% |
| Midterm Dividend — the defensive calendar books | ~12% |
| Conservative / Theme 4 — staples, healthcare, gold, T-bills | 6–9% |
The discipline was written before anyone was frightened: two weeks of hotter inflation prints tilt new money toward the dividend books first. Existing positions wait for the rebalance date or a band-crossing — a 4–5% drift from target — because rotating a portfolio on a Tuesday headline is how returns get donated to faster traders.
Why dividends carry the bad stretch: the payment doesn’t care about the day’s headlines. A book of payers keeps writing checks through the drawdown. That is cash to redeploy at the bottom — and, just as important, a reason not to sell. The behavioural rung matters as much as the financial one.
The rotation is a staircase, not a trapdoor. Every rung is a real book you could look up today, with its volatility published beside it — so “going defensive” means moving to a named place, not selling into cash and hoping to guess the way back.
And note which half of the rule does the work. New money rotates immediately; old money waits for a band. That asymmetry is deliberate: it lets the stance change without the portfolio churning.
Why Aren’t We Buying Bonds? Because We Watched What They Did On The Worst Day Of The Summer.
The classic case for bonds is two promises: they pay you, and they zig when stocks zag. In a sticky-inflation regime the second promise keeps failing — when rates are the thing frightening the market, stocks and bonds fall together: one bet, two hats.
July 29 was the cleanest demonstration in a year. The Fed did nothing. Equities sold off 2.19%. And the 30-year Treasury had its worst day in over a year, to 5.228% — the highest since 2007. The long bond sold off harder than the stocks it was supposed to hedge.
So our aggressive tiers run the bond sleeve at roughly half our own audit’s recommendation — a deliberate, disclosed call, not an accident. We would rather hold bills that pay us now than duration that has stopped doing its job. That choice has a cost — rate risk is our thinnest square — and we’d rather you hear it from us than from a drawdown.
So what fixed income would we buy? Three rungs, each with a job.
| What we own, and why | Weight |
|---|---|
| Bills (SGOV) — 4%+, zero duration, settles in days. The cushion, and the dry powder for the re-risk leg | ~3%+ |
| Inflation-protected Treasuries (TIP) — principal rises with inflation; the one bond paid to agree with our theme | ~3% |
| Intermediate Treasuries (IEF) — the recession seat. Insurance, sized like insurance | 2–4% |
| The aggregate (AGG) — ~6 years of duration in one ticker: the exact risk that just failed, bundled and labelled “core” | refused |
| IG corporates (LQD) — spreads priced for a world where nothing goes wrong, while the AI trade grows a $250 billion credit desk | refused |
| High-yield & EM debt — 8–14% menu yields on B-grade paper, one-year rows as low as −8.8% | refused |
The test for every rung: does it still do its job on the day stocks fall? Bills and TIPS said yes on July 29. The refused column said no.
We’re not anti-bond. We’re anti-pretending. Duration is a bet that inflation is finished, and the market’s own evidence — three dissents up, 3%-plus prints, a 2007-high long bond — says that bet isn’t ours to make.
So we own the fixed income that works without a forecast: bills that pay now, TIPS that pay more if we’re right about prices, and a small honest insurance seat for the road where we’re wrong. When the regime changes, the ladder changes — on a trigger, in public, logged on the ledger.
The August Rebalance, In Plain English
Why this week, and not a headline week: every midterm year since 1950 saw a market drop before election day, averaging about −16% (First Trust, 1950–2024). The quarter we are in now — the one before the vote — is historically the weakest of the entire four-year cycle, averaging +1.1%. Volatility peaks October to December, then falls to its cycle low the following spring. And the payoff sits on the other side: November of a midterm year averages +3.0%, and the eleven months after average +17.0%.
1 · Check every book against its targets. Anything 4–5 points away from where it should be gets brought back. The rules decide, not moods — that is the whole point of having targets.
2 · Take defense profits down to our own limit. Defense has run past the 20–25% sector cap we set for ourselves, a flag public on our coverage grid since July. A $120 billion Pentagon restock rally is when you take winnings off the table, not when you add.
3 · Fill the cash reserve. T-bills back to full weight in every book. Not scared money — the money we’ll spend when good companies go on sale.
4 · New money goes to the dividend books first. The written rule when inflation runs hot. Money already invested moves only when a limit is crossed.
5 · Gold goes back to target. Not sold in disappointment, not doubled in hope. Both easy moves are wrong, and we wrote that rule before this year tested it.
6 · The avoid list stays avoided. And the alarm we actually listen to is AI credit spreads — not the day’s equity headlines.
7 · We write it down. This rebalance gets a dated row on Marked to Market — so a year from now you can check whether the calm-Tuesday plan did its job.
Read that list again and notice what is missing: not one of the seven moves required us to know what happens next. Every one is a rule meeting a number — a cap breached, a band crossed, a reserve below weight.
That is the difference between positioning and predicting. A prediction needs the world to cooperate. A position only needs you to follow your own rules on the day it stops being fun.
Which rung is your cash actually on?
Most statements we review are holding the refused column — the aggregate, tight-spread corporates, a bank money market paying under half a percent. Fifteen minutes: we’ll walk your fixed income through the ladder and the July 29 test.