The index went calm the week oil went violent
The Stress Index printed Settled on 12 August with its energy score above 90. In 4,814 trading days since 2008, that pairing had never appeared before.
On 12 August the Stress Index dropped into Settled for the first time since 25 June — and its energy component crossed 90 into severe the same day. In 4,814 trading days of index history, that pairing, a severe energy score inside a headline at 40 or below, had never printed. It then printed three days running, 12 to 14 August. As of 17 August the index reads 37.6, the energy score has eased to 85.7, and nothing else inside the gauge is above 38.
Three days with no precedent in the series
The full history of the index, daily since March 2008, contains exactly three days on which the energy component stood at 90 or higher while the headline stood at 40 or lower: 12, 13 and 14 August 2026, at 38.4, 39.6 and 37.3 against energy scores of 90.2, 91.4 and 90.1. This is not an artefact of where the thresholds sit. Loosen them to energy above 85 inside a headline at or below 45 and the count grows to 26 days — every one of them in 2026, none in the eighteen years before.
The headline calm is itself deep: 37.6 is the 24.5th percentile of the index’s own past five years, and the 12 August print opened the first run of Settled readings since late June, now four prints long. The index is not merely failing to register oil. It is calmer than usual, on its own history, while one quarter of its inputs reads severe.
Severe is measuring speed, not price
The energy component does not score the oil price. It scores how violently oil is repricing: Brent’s 20-day realised volatility, now 85.2% annualised — the 95.5th percentile of its past five years — plus the size of the 63-day move (14.4%, its 71st percentile). Brent itself was last observed at $93.26 on 11 August; the spot series publishes about a week behind, and the price-percentile framing was covered here two weeks ago. What is new is not where oil is. It is how fast it keeps moving around that level.
The other volatility gauge is in its bottom quarter
Set that against the market’s general fear gauge. The VIX closed at 15.19 on 17 August, the 24th percentile of its five-year range, and the St. Louis Fed’s broad financial stress index stood at −0.77 on 7 August, its 16th — looser than normal on a composite of eighteen funding, credit and volatility measures.
Ranked the same way over the same window, the two series have spent 31 days since 1995 with Brent’s short-run volatility above its 95th percentile while the VIX sat in its bottom quarter — in six episodes: September 2000, December 2004, two runs inside the early-2015 oil-price collapse, December 2016, and the current one, which began on 6 August and was still open at the latest data. The five earlier episodes lasted 24 days between them. Each sat around an event that belonged to oil rather than to markets at large; whether that pattern holds is exactly what a sample of five cannot settle.
The calm print came from the dollar
What pushed the headline through the Settled line was not energy easing — it did not — but the dollar. The dollar-pressure component fell from 51.0 on 11 August to 38.6 on the 12th as the broad index’s 63-day pace went flat, the drift taken apart here on the 15th. Financial conditions, the heaviest weight in the index at 30% against energy’s 25%, sat becalmed throughout.
That ordering is the design, not a defect of it. The index asks whether the machinery that finances and moves trade is seizing up, and a single commodity’s turmoil is supposed to be outvoted until it shows up in funding costs, freight or the dollar. On that question the answer, for now, is no.
What the calm print cannot see
The index’s trade-activity input — a global freight-derived series — was last observed on 1 June, 77 days before the latest print. Everything the physical layer has done since is invisible to it: the Kerch Strait has gone from eleven transits a day to zero, still printing zero on 16 August; the Route Displacement Index set another record, 10.395, on the same day; the Bosporus is 35.8% below its baseline at a series low. None of that reaches the gauge before roughly November. A calm reading built on June’s freight and August’s credit is a statement about what has been measured, not about August.
What to watch
Thursday’s financial-stress print. STLFSI4 publishes weekly; the next release covers the week the record configuration printed. It is one of the two inputs to financial conditions, the component that carries 30% of the index — the first place oil’s turmoil would show if it stopped being private.
The energy score under 80. The volatility window is 20 trading days, so the violent sessions of late July and early August age out of it mechanically. If Brent goes quiet from here, the score falls on arithmetic alone by the end of the month; if it does not fall, oil is still moving.
The next trade-activity observation. The freight series updates monthly with its long lag. When the June observation rolls to July and August, the index finally votes on the month the chokepoint board has been reporting.
Correction, 13 October 2026: An earlier version counted 26 days with energy above 85 inside a headline below 45. The archived series gives 24 below 45 and 26 at or below it; the sentence now says “at or below”. It also said STLFSI4 carries 30% of the index. The 30% is the weight of financial conditions, the component STLFSI4 shares with the VIX.
Note, 13 October 2026: This piece describes version 2 of the Stress Index, which has since been replaced by version 3, built on different inputs and a different scale. The version 2 readings quoted here are archived at ttsi-v2.csv; the current index is on the Stress Index page.