Insights
The desk
Original explainers. No earnings calls recap mill, no “ten stocks to play sentiment.” Citations welcome; wholesale copies are not.
Desk note 2026-09-14: oil, the 10-year, and the tape
Oil and the 10-year travelled together. That is the desk chain: energy tightness feeding goods inflation, then term premium, then a higher discount rate on duration. The funds rate is not the wrecking ball. The long end …
Desk note 2026-09-14: oil, the 10-year, and the tape
Oil and the 10-year travelled together. That is the desk chain: energy tightness feeding goods inflation, then term premium, then a higher discount rate on duration. The funds rate is not the wrecking ball. The long end …
The oil-tight case: depletion, fake spare capacity, and the EM car
This desk is oil-bullish for a geological and industrial reason, not a tweet. U.S. tier-one rock is aging, OPEC+ spare is mostly a press release, Russia is already near a peak, and emerging-market four-wheel demand is still climbing.
The Fed can cut. The 10-year can still wreck every asset that was duration in disguise
This desk treats the U.S. 10-year as the price of time, not a press conference. If energy inflation keeps the term premium alive, policy cuts do not save equity multiples, private credit, or housing lock-in.
AI pulled 2035’s cash flows into 2024’s multiple. That is not a tech story. It is a duration story
Like the better skeptics of this cycle, this desk treats much of the AI complex as a valuation that already spent the future. Capex can be real and still be a bubble in the equity that claims it.
The Iran war premium is a freight and insurance number, not a cable-news emotion
Energy columns on this desk start from barrels, hulls, and the Strait — then maybe from speeches. A war premium that does not show up in freight, war-risk, or the front of the oil curve is a vibe.
CrowdMood Trident: six prongs, one unpublished overlay
Trident is the desk’s named overlay on Dow, S&P 500, Nasdaq 100, WTI, gold, and euro FX. We publish stance. We do not publish the oscillator, the lookback, or the thresholds. The rest is discipline.
What consumer sentiment actually measures — and what it quietly skips
A confidence number is not a spending forecast. It is a household's self-report of conditions and expectations, filtered by survey design, sample frames, and the week the phone rang.
Michigan versus Conference Board: a field guide to the split
Two U.S. household surveys can disagree for a quarter without either being broken. The split is usually about jobs versus prices, and about how questions age in a news cycle.
The sentiment-spending gap: why gloomy surveys can still print sales
Households can feel worse and spend anyway. The gap is not a glitch. It is composition, credit, and the difference between a mood and a basket.
How retailers actually use mood data in a merchandising meeting
The valuable use of sentiment is not a heroic call on next quarter’s revenue. It is a bias on mix, promo depth, and where to take inventory risk.
Housing sentiment under rate shocks: lock-in, not just gloom
Homebuying conditions can collapse while owners report they are fine. The housing consumer is two populations sharing one street.
Social listening is not a survey with extra spice
Posts are cheap, loud, and biased toward people who perform opinions. Surveys are slow, expensive, and biased toward people who answer phones. Use both; do not launder one as the other.
Reading Asia-Pacific consumer mood without flattening it into one dragon
Korea, Japan, China, and ASEAN do not share a household. They share a time zone on your dashboard. That is not the same thing.
How to use a sentiment signal without overfitting it to last year’s crash
A tradable sentiment product fails in the usual ways: too many rules, no cost of trading, and a backtest that starts the day the vendor’s story looks good.