Fed hike odds fell—but the euro still closed lower: hold the short, do not add

Fed hike odds fell—but the euro still closed lower

Research cutoff: September 29, 2026, 18:06 Toronto

Four-month horizon: through January 27, 2027

Paper decision: Hold −8 December Micro Euro futures (M6EZ26); add 0, trim 0, exit 0. Make no new trade.

The afternoon produced a useful split signal. New York Fed President John Williams said there was no urgency after September's hike, and market pricing for an October increase fell to 51.5% from almost 70% earlier in the session. The two-year yield eased while the 10- and 30-year yields rose. That is not a clean, across-the-curve hawkish repricing: the front end heard patience while the long end continued to price inflation, supply and term-premium risk.

Yet the euro still finished lower. December standard Euro FX futures (ECZ26) settled at 1.1377 USD/EUR. That observation is used only as a same-expiry proxy for the Micro Euro position because a reliable exact M6EZ26 closing quote was unavailable; it is not represented as an observed micro-futures execution. The result is a disciplined Hold, not Add: the medium-term European energy, political and relative-rate pressures survive, but Williams and soft U.S. activity data are real counterevidence ahead of PCE.

Decision board

Existing euro short: re-underwritten, not defended by inertia

The position remains short eight M6EZ26 contracts from an assumed paper entry of 1.1422. Eight micros equal €100,000 of exposure. The stop remains 1.1550, the base target 1.1150, and the mandatory exit November 30; there is no automatic roll.

Using the 1.1377 ECZ26 same-expiry proxy:

The forward scenario tree is deliberately less optimistic than a simple continuation case. With 35% at 1.1150, 35% at a 1.1350 time exit and 30% at the 1.1550 stop, expected incremental P&L is about +$370 gross or +$332 after estimated exit friction. These are analyst judgments, not market-implied probabilities. The two-outcome target-before-stop breakeven probability after exit costs is about 44.2%.

The position still earns its place because the euro failed to rally on favorable U.S. news. But the expected value is only about 0.15 times buffered forward risk—enough to retain a small position, not enough to double it. Adding another eight micros would take same-thesis buffered risk toward the 0.50% thesis cap while leaning against fresh evidence from Williams and weaker U.S. activity. That would be technically possible but economically unjustified.

Positioning is also more nuanced than a one-way crowded-short story. CFTC data for September 22 show leveraged funds net short 26,694 euro contracts, but asset managers remained net long 216,483; open interest fell sharply. The weekly lag and expiry effects limit inference. The correct conclusion is mixed ownership and squeeze sensitivity—not automatic invalidation.

Invalidation: exit at 1.1550, if U.S.–European front-end expectations narrow durably while Europe's energy disadvantage normalizes, or on November 30. A softer PCE or payrolls report that sustains lower U.S. front-end pricing requires reassessment; a high 30-year yield alone would not confirm the thesis.

Sources: New York Fed—Williams speech, Reuters global market close, MarketWatch ECZ26 settlement, CFTC financial futures, September 22, CME Micro Euro specifications.

Oil: the supply-relief short is right in direction and late in timing

November WTI settled at $89.38, down $3.22, while December settled at $87.25. The November/December backwardation compressed to $2.13 from an estimated $3.59 on the prior price changes. Saudi pipeline and Yanbu operations are restoring export flexibility, and API later estimated a 1.019-million-barrel crude build versus a 1.9-million-barrel draw forecast. Exact MCLX26 then reopened at $88.63.

That is coherent evidence for normalization, but it is not permission to chase. The curve remains backwardated, distillate stocks remain well below normal, and renewed attacks could reverse the export recovery. Official EIA data arrive Wednesday at 10:30 Toronto.

Whole-contract sizing is feasible. One MCL controls 100 barrels. A three-micro short with a $6 stop distance risks $1,800 raw or $2,250 after the policy buffer before costs—about 0.225% of NAV. Four micros would exceed the normal 0.25% starter budget before costs. The package fits; the immediate edge does not. The better gate is a failed rebound after EIA confirmation and renewed curve compression.

Sources: Reuters oil close, API Weekly Statistical Bulletin, EIA weekly petroleum summary, CME Micro WTI specifications.

Natural gas: bearish structure, incomplete entry

November Henry Hub settled at $3.011/MMBtu, December at $3.338 and January at $3.704. The 32.7-cent November/December contango and a third declining session favor the bearish view. EIA's latest total storage was 3,351 Bcf, 2.9% above the five-year average.

The caveat matters: South Central salt storage remained 12.5% below its five-year average, and November has fallen almost 12% in five sessions. Eight micro contracts with a $0.150 stop would risk $1,200 raw or $1,500 buffered before costs—mechanically inside the starter band—but the immediate price path is stretched. Wait for either a failed rebound or Thursday's storage confirmation rather than convert a valid thesis into a poor entry.

Sources: EIA natural-gas storage, NOAA forecast discussion, CME Micro Henry Hub specifications.

Rates, equities, gold and agriculture

Rates. Closing yields—2-year 4.889%, 10-year 5.255% and 30-year 5.592%—show a steeper curve after Williams. A conditional two-contract ZN long above 104.71875, with 104.125 stop and 106 target, would fit at roughly $1,561 buffered risk including estimated round-trip costs. But that trigger is above Tuesday's high and remains unmet. It is a post-PCE setup, not an order.

Equities. The cash S&P 500 finished 0.17% lower but recovered from its lows. A two-contract MES short near 7,745 with 7,850 stop and 7,580 target would fit at roughly $1,332 buffered risk, but no fresh breakdown occurred. Wait.

Gold. December gold settled at $4,179.70 and exact MGCZ26 reopened at $4,214.60. A one-contract short can be sized inside the risk budget, but it duplicates the euro position's dollar/real-yield factor and is vulnerable to further Williams/oil relief. Zero contracts.

Agriculture. December corn, soybeans and wheat settled at 522.00¢, 1,297.75¢ and 692.75¢. Reported Chinese soybean inquiries and harvest delays are not confirmed additional purchases. USDA Grain Stocks and Small Grains arrive Wednesday at noon Toronto. No pre-release position has a sufficiently robust information advantage.

Next 72 hours

• Wednesday, September 30, 08:30 Toronto: U.S. PCE inflation and third-estimate GDP. This is the main immediate test for the euro short and yield curve.

• Wednesday, September 30, 10:30: EIA petroleum inventories. Confirmation or rejection of API's crude build determines whether the WTI setup improves.

• Wednesday, September 30, 12:00: USDA Grain Stocks and Small Grains. Stay flat until the surprise and price response are observable.

• Thursday, October 1, 10:00: ISM manufacturing, including prices and employment.

• Thursday, October 1, 10:30: EIA natural-gas storage.

• Friday, October 2, 08:30: U.S. payrolls, wages, unemployment and revisions.

• Friday, October 2: Euro-area flash inflation; the published date is verified, but no exact release time is assumed here.

Primary calendars: BEA, ISM, BLS, Eurostat, USDA NASS.

Bottom line

The strongest evidence is not simply that yields are high. It is that near-term Fed pricing softened while the euro still failed to recover. That resilience keeps the existing short valid, but the weaker immediate rate-divergence impulse removes the case for more size. Hold eight M6EZ26, preserve the 1.1550 stop and November 30 exit, and keep every challenger flat until scheduled data improve the entry—not merely the story.

*All trades and results are simulated. Reference prices may be delayed. The 1.1377 euro mark is a same-expiry standard-contract proxy, not an observed M6EZ26 execution. Stops are risk thresholds, not guaranteed fills.*