The positioning numbers are unambiguous. In the week to 11 August 2026, leveraged funds held a net short of 77,671 contracts in Euro FX, and the book grew rather than shrank: the net short deepened by 4,622 contracts over the week. That places the position at the 90th percentile of open interest and, on the three-year window, at the 2.6th percentile of net positioning, a z-score of negative 2.25. The desk's own stretch reading dates the last comparably crowded short to 28 July 2026. This is not a moderate lean. It is a book with very little room left to add.
Set that against the tape. EUR/USD trades at 1.1581 on 18 August 2026, up 1.34% over the past month and sitting at its own 20-day high. The US Dollar Index, meanwhile, is at 99.62, down 1.12% over the month and parked at its own 20-day low of 99.6. The two are moving in the direction they're supposed to: EUR/USD against the Dollar Index has run a negative 0.91 correlation over the past 20 sessions, a level shift from the negative 0.30 correlation of the prior 60. The relationship has tightened just as the dollar leg has weakened.
A euro short this stretched, sitting against a dollar this weak, has become a bet against the market's own recent direction, not merely a currency view.
The dollar side has its own supporting evidence. The broad dollar index sits at the 21st percentile of its trailing year, a z-score of negative 0.88 with a velocity reading of negative 1.05, an unusually fast move lower over the past month. One wire attributes the softness to dwindling rate-hike bets and the Iran conflict; that framing is one account among several and is treated here as colour, not mechanism. The mechanism the desk can actually price sits in the futures curve. The 12-month priced path carries a positive delta of 26 basis points, a firmer, no-cuts path rather than an easing one, alongside a 6-month delta of 14.5 basis points that reads as roughly flat. A market pricing a firmer Fed path twelve months out is not obviously a market that should also be pressuring the dollar lower on 18 August 2026, and that mismatch is part of why the euro short has not yet paid off.
The Atlanta Fed's GDPNow model cut its third-quarter growth estimate to 4.03% from 4.31% prior, a soft datapoint against an otherwise resilient growth narrative, and the FOMC Meeting Minutes land on 19 August 2026 at 18:00 UTC. Either could move the front end enough to reprice the near-term path and, with it, the dollar leg of this trade. The 2-year Treasury note offers a caution against reading positioning extremes as clean signals: leveraged funds extended that net short by 30,275 contracts in the same reporting week, taking it to the 97th percentile of its 52-week range and the 93.6th percentile of the three-year window. Proof that a crowded book can keep building rather than snap.
The desk's 6 August 2026 note argued that front-end Treasury short-covering reflected positioning ahead of the 7 August payroll print rather than a directional bet on Fed cuts, with its falsifier keyed to that print landing at or below 57,000 (with the 2-year yield falling back through 4.0%) or at or above 85,000 (with the yield holding above 4.0%). That data lies outside this pack: nothing here reports the 7 August Employment Change figure or the 2-year yield's reaction to it, so that thesis cannot be graded from 18 August evidence and stands untested rather than confirmed. What can be said on 18 August is narrower, and about a different market. The euro short is not a call this desk is making, but the crowd's own exposure is now the story. The desk is watching whether the Euro FX net short begins to shrink in the next CFTC Commitments of Traders report, or continues to deepen alongside EUR/USD holding above 1.1369.
