The Drift Was Already There
John Carmean
The Strait of Hormuz closed to commercial shipping this week. European gas prices doubled in 48 hours. Twenty percent of global oil supply is at risk. Every headline is about the military conflict.
But if you were watching the energy system, the fragility was visible weeks before the first strike.
In late January, Winter Storm Fern hit the U.S. gas market hard. Henry Hub spot prices set a daily record. Storage inventories swung from a 177 billion cubic foot surplus to a 130 billion cubic foot deficit in three weeks. Production dropped 3% from freeze-offs. LNG export facilities curtailed operations. The system absorbed the hit and prices retreated. By late February, traders had moved on. Forecasters were calling for warm weather. The consensus was that the winter scare had passed.
It hadn't.
What passed was the acute symptom. The structural damage remained. U.S. storage entered March well below the five-year average. European storage sat at 30%, compared to 40% a year ago. Germany was at 21%. The system had no buffer left. And while the weather desk was relaxing, crude oil volatility was quietly doubling. The spread between international and domestic oil benchmarks was widening. Energy-specific fear, as measured by options markets, was rising nearly twice as fast as general market fear. The energy market was pricing in something the broader market had not yet acknowledged.
Then February 28 happened. And every structural weakness that January exposed became an active crisis overnight.
This is the pattern that repeats. The trigger gets the headline. The vulnerability was already in the data. Uri was the same. Summer 2022 was the same. The cause changes every time. The structure underneath does not.
What is different about this moment is the compound nature of the stress. A weather event weakened the system. A geopolitical event hit the weakened system. Two independent causes, converging on the same infrastructure, within five weeks of each other. No single-domain model connects a Texas winter storm to an Iranian military conflict. But the energy system does not care about causes. It only knows whether it has margin or it doesn't.
The organizations that read across those boundaries had weeks of lead time. Not to predict the war. To see that the system was exposed, and to position accordingly.
The ones that didn't are explaining it to their boards this week.
John Carmean developed a drift detection methodology validated across energy, transportation, and financial domains using U.S. public domain datasets with locked parameters and no domain-specific tuning.