# ATG futures daily monitor — September 9, 2026

## Decision and data boundary

Exit thesis / retire the unentered platinum/gold and ULSD/RBOB watches. Hold the yen research thesis without adding; Wait on sugar, wheat/soy, SOFR, gas, VIX and OAT/Bund. No new paper trade qualifies. The funded paper account remains entirely in USD cash. These are research decisions, not executed trades. No automatic reversal is authorized.

Research cutoff: September 9, 20:39:14 EDT (2026-09-10T00:39:14Z). Completion: 2026-09-10T01:23:19Z (September 9, 21:23:19 America/Toronto). The report does not use evidence arising after cutoff; pending events remain pending even if publication preparation finishes later. Prior monitor: September 8, 20:50:01.486 EDT. Horizon remains September 6, 2026–January 6, 2027.

Research status is **partial**: current execution-grade margins, matching depth and some lifecycle clocks remain unverified; several watches lack targets or hedge mechanics. Exact-contract data is available for the monitored expressions, but delayed lasts and date-only settlements are not executable orders. CME overnight observations use the September 10 Globex trade-date label while occurring on September 9 Toronto time. Vendor clocks are minute precision, not verified seconds. Current portfolio valuation is separately **cash-only**, not indicative: no risky asset mark or FX is needed.

## What changed since the prior monitor

China inflation outcome now reviewed: NBS published August CPI/PPI at September 9 09:30 Beijing, equivalent to September 8 21:30 EDT—after the prior 20:50 EDT cutoff. CPI rose 0.8% y/y and 0.4% m/m; core CPI was 1.0% y/y. The official NBS release index verifies PPI +3.8% y/y and +0.4% m/m; its full detail page remained inaccessible. These are actual pre-cutoff outcomes, not pending events. [NBS CPI release](https://www.stats.gov.cn/sj/zxfb/202609/t20260909_1965263.html), [NBS PPI release index](https://www.stats.gov.cn/sj/zxfb/).

Reuters corroborates CPI/PPI and reports July readings of +0.5%/+3.5%; August CPI matched its poll and PPI exceeded the +3.6% forecast. Its report links rising producer inflation to energy costs while domestic demand remains soft. Analytical inference: the result reinforces cost-push/stagflation caution, not a broad demand boom or a new trade. Existing decisions and gates are unchanged. [Reuters inflation report](https://www.reuters.com/world/asia-pacific/chinas-producer-inflation-regains-momentum-august-2026-09-09/).

Verification of the NBS releases occurred after the 00:39:14Z market cutoff; the releases themselves demonstrably predate that cutoff. No later futures quotes or economic outcomes were added.

The principal changes are fundamental and lifecycle decisions, not just mark updates. WPIC's actual Q2 report flips the 2026 platinum balance from a prior 297 koz deficit to a 265 koz surplus: the supportive-report first gate fails. The platinum ratio nevertheless rose, consuming more remaining value. The products spread passed its untouched $1.45 target without entry. Both old watches are retired, with no booked gain, loss, stop or fill. EIA's actual September 9 STEO supports a gas storage cushion and distillate tightness but pushes most oil normalization to Q2 2027; its forecast inputs stopped September 3 and exclude subsequent attacks. [WPIC Q2](https://platinuminvestment.com/files/354715/WPIC_Platinum_Quarterly_Q2_2026.pdf), [EIA gas](https://www.eia.gov/outlooks/steo/report/natgas.php), [EIA products](https://www.eia.gov/outlooks/steo/report/petro_prod.php), [EIA oil](https://www.eia.gov/outlooks/steo/report/global_oil.php).

The September 9 BLS release was June ECEC, not ECI: civilian compensation cost was $49.46/hour, with $33.85 wages and $15.61 benefits; private-industry total $46.89. These are cost levels, not a newly released wage-inflation surprise. The cancelled SOFR event trade stays cancelled. Masu's speech is September 10 10:30 JST, which is September 9 21:30 EDT—not the next Toronto morning. [BLS ECEC](https://www.bls.gov/news.release/ecec.nr0.htm), [BOJ schedule](https://www.boj.or.jp/en/about/calendar/index.htm).

Unless specified, changes below compare current exact-contract snapshots against the prior snapshot, a roughly 23h49m monitoring interval, not synchronized same-clock session returns. Current vendor and prior CME snapshots can differ in convention. Separate settlements are explicitly kept separate. Ratios, price changes and scenario forecasts are never paper P&L. Scores are priced-in scores from 0 to 5 (higher means more priced), not conviction, execution readiness or measured probabilities.

| Prior idea | Decision now | Exact current metric | Change versus prior observation | Priced-in score |
|---|---|---|---|---|
| Yen | Hold research thesis; no add | 6JZ26 0.0065565 | −0.0000130 / −0.1979% | 4, unchanged |
| Sugar | Wait | SBH27 vendor settle 19.39¢ | +0.30¢ / +1.5715% | 2, unchanged |
| Wheat / soy | Wait | 2×743.75/1,329.75 = 1.118631 | −1.2048% indicative ratio | 2.5, unchanged |
| Platinum / gold | Exit thesis; retire unentered watch | 1,924.70/4,475.90 = 0.430014 | +2.8386% indicative ratio | Prior 4 → final analytical 2.5 |
| SOFR | Wait; setup cancelled | 100×(95.915−95.765) = 15.0 bp | +0.5 bp adverse | 2, unchanged |
| Gas | Wait | 3.816−2.870 = $0.946/MMBtu | −$0.014 | 3.5, unchanged |
| Products | Exit thesis; retire unentered watch | 4.2680−2.7905 = $1.4775/gal | +$0.1162 | Final analytical 4 |
| VIX | Wait | 18.6248−19.4338 = −0.8090 | +0.0402 points | 3.5, unchanged |
| OAT / Bund | Wait; unrated | 113.64 / 121.34, separate settlements | −0.90 / −0.70 points | null |

Each row's direct quote source, clocks and limitations appear in the detailed sections and one-URL evidence links below.

## Regime and new opportunity assessment

The existing supply-friction regime has materially intensified toward an energy-supply shock and stagflation risk; this is an analytical inference, not a newly measured structural correlation regime. Reuters' captured market summary describes oil gains alongside weaker U.S. equities and higher Treasury yields. EIA estimates August Middle East shut-ins at 6.7 million b/d versus 5.0 million in July and forecasts 5.7 million in Q4, with most normalization only in Q2 2027. Secondary flow estimates show meaningful bypass/dark exports, but not sustained full restoration; Gulf-all-route, Hormuz-only and peak-day flow measures have different scopes. Party claims about shipping attacks are not independently verified losses. [Reuters market summary](https://www.reuters.com/world/china/global-markets-global-markets-2026-09-09/), [EIA oil](https://www.eia.gov/outlooks/steo/report/global_oil.php), [Reuters flow analysis](https://www.reuters.com/business/energy/one-third-gulf-oil-is-still-missing-despite-dark-crossings-data-shows-2026-09-09/), [Reuters shipping report](https://www.reuters.com/world/middle-east/iran-attacks-us-base-jordan-ships-near-hormuz-after-tankers-sunk-2026-09-09/).

Common shocks connect oil, distillate, LNG, ethanol-related sugar supply, inflation-sensitive rates and yen import costs. The watches are not independent diversification. Fixed 2:1 grains and 5:1 metals counts do not establish beta neutrality; dollar matching does not eliminate basis, CTD, conversion-factor or regional-delivery risk. No matched-history correlation matrix or crowding percentile was recomputed. Gap/slippage can exceed one-tick assumptions and conditional stops are not guarantees. The correct current response is no new risk, not a new trade manufactured from the more dramatic regime. No superior futures opportunity with complete supporting evidence was established. This is a daily monitor, not a new global-universe review.

## Active thesis reviews

### Japanese yen — Hold thesis, no add; score 4/5

Long 6JZ26. Retain the research-held yen thesis without adding. Price is above the original band and Masu is inside one hour; reported potential domestic-flow retention is a mechanism, not measured new repatriation.

Quote and trend: CME 10-minute-delayed last, September 9 20:25:54.020 EDT; retrieved 20:35:06.163 EDT. Feed trade date is September 10 Globex; volume 3,395 belongs to that newly opened overnight session. −0.1979% (−0.0000130 vs prior delayed snapshot; not a settlement return). [cme-yen](https://www.cmegroup.com/CmeWS/mvc/quotes/v2/69), [boj-schedule](https://www.boj.or.jp/en/about/calendar/index.htm).

Return direction and counterargument: September 6 displayed +2.5% scenario expectation remains historical. Current direction is conditionally positive, but entry asymmetry remains weak: Japan tightening is substantially anticipated, energy-import costs and U.S. inflation offset it. The small price retreat mechanically helps unchanged terminal-value sensitivity; no fresh numeric forecast is asserted. Imported-energy costs can hurt Japan; persistent U.S. inflation/rates offset BOJ normalization. A nearly anticipated hike may disappoint without further tightening guidance.

Physical, curve/carry, positioning and liquidity: Priced-in score stays 4/5, not conviction. September 1 TFF leveraged-fund net −102,188 is still the latest vintage; no fresh squeeze is measured. 6JZ26 volume 3,395, OI and bid/ask/depth unavailable. Synchronized September/December prices imply +0.0000460 deferred premium, not free carry; no fresh spot/OIS basis or stable hedge correlation measured. [reuters-yen-flow](https://www.reuters.com/world/asia-pacific/rising-japanese-yields-may-keep-more-domestic-money-home-fitch-says-2026-09-09/), [reuters-fed-poll](https://www.reuters.com/business/fed-hold-rates-steady-rest-2026-rising-number-analysts-see-least-one-hike-2026-09-09/), [cftc-tff](https://www.cftc.gov/dea/futures/financial_lf.htm).

Entry/gates: Unchanged 0.00635–0.00639 entry/add band, plus intact relative-policy and durable yen-demand evidence, executable quotes and complete specifications. Current last is above the band. Target: Unchanged 0.00670–0.00680 conditional objectives; current observation is below both. Stop: 0.00618 remains an inactive research invalidation level; no position, standing stop or reconstructed intraday path. Invalidation: Sustained adverse Japan/U.S. policy divergence or failure of durable yen demand despite anticipated BOJ tightening.

Catalyst: Masu speech September 9 at 21:30 EDT, only 50m46s after cutoff: wait for catalyst, no new risk. PPI September 10 08:30, Japan CGPI 19:50, CPI September 11 08:30 EDT remain pending. Lifecycle: Review December 7–11, before December 14 last-trade date. A December-to-March roll is needed for January 6; 6JH27 has no fresh last. Exact cessation and safe exit times remain unverified.

Evidence required: U.S. inflation and the actual BOJ conclusion must support the thesis; retained research status is not a paper position. Catalyst disappointment or adverse policy divergence after the BOJ conclusion would challenge the thesis under the existing invalidation. Coverage: Current exact delayed December price, dated positioning, calendar and policy-flow context reviewed. Current initial margin, durable demand, executable depth, March roll price and exact lifecycle safety remain incomplete.

### Sugar short — Wait; score 2/5

Short SBH27. Wait despite the retest band being touched. No new accessible sugar-share/exportable-supply confirmation or measured long liquidation completes the setup, and daily extremes cannot establish its ordered path.

Quote and trend: September 9 date-only vendor settlement 19.39; delayed last 19.35 at 12:59 EDT. Day range 19.07–19.45. Retrieved 20:37:08 EDT from exact March contracts page; OHLC does not prove a prospective failed retest. +0.30¢ (+1.5715% vs 19.09 prior vendor settlement). [sugarQuote](https://www.wsj.com/market-data/quotes/futures/SBH27/contracts), [cftcSugar](https://www.cftc.gov/dea/futures/ag_sf.htm).

Return direction and counterargument: The September 6 +2.6% scenario estimate stays dated. Higher settlement mechanically improves a hypothetical short at fixed terminal prices, but fuel/ethanol risk weakens the physical case; no recalibrated numerical return or win probability is claimed. High oil and altered fuel taxes can favor ethanol and limit exportable sugar. A larger calendar premium is not proof of future surplus; adverse crop or export-policy changes defeat the short.

Physical, curve/carry, positioning and liquidity: Priced-in score remains 2/5. September 1 managed-money net +233,771, with weekly longs +36,401 and shorts +647, is old accumulation, not current liquidation. Displayed SBH27 volume/OI 163,453/464,513 lacks a matched OI change or depth. March is a seasonal hump; curve carry is not guaranteed short profit. [unica](https://unica.com.br/noticias/), [iso](https://www.isosugar.org/publications/3/quarterly-market-outlook), [brazilFuel](https://www.reuters.com/business/energy/brazil-government-releases-13-billion-fuel-subsidies-2026-09-09/), [sugarSpecs](https://www.ice.com/products/23/Sugar-No-11-Futures), [sugarLifecycle](https://www.ice.com/products/23/Sugar-No-11-Futures/expiry).

Entry/gates: Prospective failed retest of 19.20–19.40¢/lb, with improving exportable supply and newly observed long liquidation. The band is touched, but the required sequence and physical gates are not established. Target: 18.00¢ first, then 17.50¢/lb, only after a qualifying future entry; no order exists. Stop: 20.15¢/lb conditional only; no current holding, fill or executed stop. Invalidation: Material upward deficit revision, persistently weaker Brazil sugar share/output, or export restrictions that prevent supply response.

Catalyst: Read actual September 11 CFTC publication at 15:30 EDT and any genuinely published UNICA update; UNICA timing remains unknown. Recheck ethanol/sugar netbacks after Brazil fuel-policy changes. Lifecycle: March covers the January 6 horizon. Official ICE calendar gives February 26 last trade and March 1 first notice; 112,000 lb, 0.01¢ tick = $11.20. Exact cessation/broker safety and current margins remain unverified.

Evidence required: The next CFTC vintage and any actually published UNICA update must support the setup; UNICA timing remains unknown. Rejection OHLC alone cannot activate the watch. Retention or demotion depends on physical evidence; adverse supply revisions cancel the setup. Coverage: Exact March vendor last/settlement, physical-source limitations, CFTC vintage and ICE lifecycle reviewed. UNICA/ISO content, actual netbacks, current margin, depth and safe execution clock remain material gaps.

### Wheat / soybeans — Wait; score 2.5/5

Long 2 ZWH27 / short 1 ZSH27. Both current delayed and separate settlement ratios are inside the numerical band. Wheat demand/basis has not strengthened and WASDE has not occurred; retain Wait and zero completed failed post-WASDE cycles.

Quote and trend: WSJ March wheat/soy both display September 9 19:21 CDT (20:21 EDT); retrieved 20:38:03.597 EDT. Ratio 2×743.75/1,329.75=1.118631322. Same-minute labels lack tick seconds and bid/ask; package remains indicative. Separate September 9 settlements 744.25/1,331.75 imply 1.117702271. −1.2048% (vs prior delayed ratio 1.132273; not package P&L). [Exact March wheat](https://www.wsj.com/market-data/quotes/futures/WH27), [Exact March soy](https://www.wsj.com/market-data/quotes/futures/SH27).

Return direction and counterargument: September 6 displayed +2.3% scenario expectation remains historical. The lower entry ratio mechanically helps long-wheat/short-soy value at unchanged terminals, but weak physical demand means the qualitative four-month edge remains conditional, not newly validated. No refreshed numeric estimate. Soy-specific crop damage can hurt the short; abundant wheat and unchanged delivered basis oppose the long. Different marketing years and unequal curve intervals cannot be treated as equal physical exposure or carry.

Physical, curve/carry, positioning and liquidity: Priced-in score remains 2.5/5. September 9 Wabash delivered SRW September basis −70U to −55U is unchanged, not March basis. Latest inspections are still September 3 week: wheat 342,733t, −20.66% weekly, MY −27.75%. September 1 legacy non-commercial net +24,703 wheat/+247,987 soy uses a different CFTC category from prior board figures, not a daily change. Evening volumes 223/449; OI 114,948/163,059, dates unspecified. No matched correlation series or depth. [USDA Illinois bids](https://www.ams.usda.gov/mnreports/ams_3192.pdf), [USDA inspections](https://www.ams.usda.gov/mnreports/wa_gr101.txt), [USDA WASDE](https://www.usda.gov/oce/commodity/wasde), [CFTC legacy CBOT](https://www.cftc.gov/dea/futures/deacbtlf.htm).

Entry/gates: Unchanged 2W/S 1.115–1.135 only after actual September 11 WASDE and stronger delivered-wheat demand/basis, with synchronized executable legs. Current numerical-band observation alone is insufficient. Target: Conditional 2W/S 1.18–1.20, not reached by the current indicative snapshot. Stop: Conditional 2W/S 1.08, not a live package stop. No stop path inferred from asynchronous daily extremes. Invalidation: Normalized Black Sea shipping/insurance without import switching, larger exportable wheat supply without demand, or soybean-specific supply damage. Demote after two unsuccessful post-WASDE export cycles.

Catalyst: September 11 WASDE at 12:00 EDT, then two actually reviewed export cycles. CFTC 15:30. Holiday-adjusted export-sales timing is unverified, and no pre-WASDE observation counts as a failed post-report cycle. Lifecycle: March expiries are beyond January 6; no horizon roll required. Exact notice/last-trade calendars, current ticks/margins and safety clocks are not freshly verified. Two 5,000-bushel wheat contracts against one soybean contract do not ensure beta neutrality.

Evidence required: Two subsequently reviewed export cycles, with dated releases, must show improved delivered demand/basis. The existing demotion rule applies after two failed cycles; no calendar due dates are inferred for unreleased or unreviewed data. Coverage: Exact month observations, separate USDA-reproduced settlements, basis, inspections, CFTC category and release calendar reviewed. Post-WASDE demand confirmation, executable depth, margin and exact lifecycle inputs remain incomplete.

### SOFR flattener — Wait; score 2/5

Long SR3H27 / short SR3Z26. Slope steepened 0.5 bp to 15.0. ECEC is a compensation-cost level release, not a new ECI inflation surprise; retain the cancelled watch pending actual PPI/CPI and complete prospective terms.

Quote and trend: CME 10-minute-delayed H27/Z26 share September 9 20:25:44.934 EDT; retrieved 20:35:05.494 EDT. 100×(95.915−95.765)=15.0 bp implied-rate slope, not a meeting probability or executable spread. +0.5 bp (Adverse steepening vs 14.5 bp prior monitor). [cme-sofr](https://www.cmegroup.com/CmeWS/mvc/quotes/v2/8462), [ecec](https://www.bls.gov/news.release/ecec.nr0.htm).

Return direction and counterargument: No numerical scenario return was validated. Current direction is neutral-to-negative for the retained flattener under oil/inflation risk; a wider slope does not itself revive the cancelled policy edge. Restrictive policy can sustain the positive slope. Quarterly compounded SOFR and repo/basis dynamics are not one-for-one meeting probabilities.

Physical, curve/carry, positioning and liquidity: Priced-in score remains 2/5. H27/Z26 fell 0.025/0.020 points versus prior marks; long H/short Z moved adversely. Overnight volumes 4,600/6,687, OI and depth unavailable. September 1 leveraged-fund net −2,605,980 is all-month context. Convexity, repo and quarter/meeting allocation prevent literal policy-probability interpretation. [bls-calendar](https://www.bls.gov/schedule/2026/09_sched.htm), [reuters-fed-poll](https://www.reuters.com/business/fed-hold-rates-steady-rest-2026-rising-number-analysts-see-least-one-hike-2026-09-09/), [sofr-rulebook](https://www.cmegroup.com/rulebook/CME/IV/400/460.pdf), [cftc-tff](https://www.cftc.gov/dea/futures/financial_lf.htm).

Entry/gates: Cancelled prior event-entry setup stays cancelled. Wait for actual PPI/CPI policy and flattening evidence; no numeric entry band is currently approved. Target: No validated target; a prospective instruction remains incomplete. Stop: 16 bp is an inactive conditional adverse threshold, 1 bp above the current slope; not a standing stop or executed risk exit. Invalidation: Persistent inflation/policy evidence against flattening. An overshoot alone cannot activate this watch.

Catalyst: September 9 ECEC outcome reviewed, not ECI. PPI September 10 and CPI September 11 at 08:30 EDT remain pending; no anticipatory entry edge is established. Lifecycle: SR3Z26 reference quarter December 16–March 17, last trades March 16, 2027; H27 reference quarter March 17–June 16, last trades June 15. Named month starts accrual, not expiry. Contract-specific ticks and exact cessation/safety remain unverified.

Evidence required: PPI/CPI must support relative flattening and complete prospective trade terms. Demotion from entry consideration is warranted if post-release terms or the policy edge remain unsubstantiated. Coverage: Exact delayed legs, BLS release/calendar, CFTC vintage and SOFR reference-quarter rule reviewed. Entry, target, adverse leg scenario, execution-grade margin/depth and exact safety clocks remain missing.

### Natural gas calendar — Wait; score 3.5/5

Short NGF27 / long NGH27. Published STEO reinforces a storage cushion, but the entry premium remains below $1.10 and the target is still absent. No trade qualifies; winter and LNG confirmation remain unavailable.

Quote and trend: Vendor delayed lasts September 9 20:16/19:47 EDT, a 29-minute skew; retrieved 20:36:36 EDT. Jan−Mar=$0.946. Separate date-only settlements 3.822/2.873 give $0.949; do not mix these vintages. −$0.014 (vs prior Jan−Mar premium $0.960). [gasQuote](https://www.wsj.com/market-data/quotes/futures/NGF27/contracts), [steoGas](https://www.eia.gov/outlooks/steo/report/natgas.php).

Return direction and counterargument: No numerical four-month model was validated. Smaller premium reduces prospective compression payoff; the new storage/production forecast helps the thesis conditionally, but LNG/winter risks and missing target prevent positive expectancy from being established. Record LNG demand, freeze-offs and regional deliverability can overwhelm aggregate storage comfort. A record-production forecast is not a guarantee of a mild January or a fair-value estimate for this spread.

Physical, curve/carry, positioning and liquidity: Priced-in score stays 3.5/5. STEO forecasts record production and winter-start inventories 5% above average, but record LNG demand too. Latest actual stock remains August 28: 3,214 Bcf, +30 weekly, +160 vs five-year, −50 y/y. September 1 managed-money net −89,481 is not Jan−Mar crowding. Evening volumes 97/15 and OI 181,605/165,949; no matching depth. Seasonal January hump and carry are path-dependent. [weeklyGas](https://ir.eia.gov/ngs/ngs.html), [gasSchedule](https://ir.eia.gov/ngs/schedule.html), [cftcGas](https://www.cftc.gov/dea/futures/nat_gas_sf.htm), [gasSpecs](https://www.cmegroup.com/rulebook/NYMEX/2/220.pdf).

Entry/gates: Unchanged Jan−Mar premium ≥$1.10/MMBtu plus supportive winter/deliverability/LNG evidence and complete execution conditions. Current $0.946 fails the numerical gate. Target: Numeric target remains unavailable; a complete prospective trade cannot be specified. Stop: $1.35/MMBtu is conditional only, not an active stop; no holding or fill exists. Invalidation: Persistent January scarcity from severe winter weather, regional deliverability, freeze-offs or LNG demand defeats the fade.

Catalyst: September 9 STEO reviewed; September 10 storage at 10:30 EDT still pending. Actual terminal-level LNG feedgas and fresh winter evidence must be checked, not substituted with European gas headlines. Lifecycle: Rule-derived NGF27 last trade December 29, 2026; NGH27 February 24, 2027, notice next business day. Review December 18–23; January leg cannot cover January 6 unchanged. 10,000 MMBtu; $0.001 tick=$10. Exact dated calendars, cessation/safety clocks and margins remain unverified.

Evidence required: STEO/storage, the January premium against the original 1.10 gate and regional deliverability must support the trade. Demotion from entry consideration is warranted if a target or adverse scenario remains unavailable after the storage release. Coverage: Exact delayed vendor legs and separate settlements, STEO, actual older inventory vintage, CFTC and rulebooks reviewed. 29-minute skew, current margin/depth, winter/LNG detail, target and lifecycle clocks remain material blockers.

### VIX calendar — Wait; score 3.5/5

Long VXV26 / short VXZ26. Both official settlement legs rose and October outperformed modestly, but the spread is still outside the entry band and has no validated target or distinct expiry-relative edge.

Quote and trend: Cboe September 9 official date-only settlements: October 18.6248, December 19.4338; difference −0.8090. Retrieved 20:33:26 EDT. Exact settlement/publication clocks absent; these observations cannot execute. +0.0402 (Favorable narrowing vs −0.8492; no package P&L). [cboe-settlement](https://www.cboe.com/markets/us/futures/market-statistics/settlement/futures/daily), [cboe-statistics](https://www.cboe.com/markets/us/futures/market-statistics/daily/futures).

Return direction and counterargument: No numerical expected return is established. Favorable 0.0402-point narrowing does not prove prospective edge and moves farther from the preferred cheap-entry band; a rise in both maturities is not the thesis. December can rise faster than October during a volatility shock. Roll, carry and two-leg slippage can absorb the small spread edge.

Physical, curve/carry, positioning and liquidity: Priced-in score stays 3.5/5. October/December rose 0.2228/0.1826 points; September/October/November/December curve 17.052/18.6248/19.2547/19.4338 remains in contango. Current contract volume/OI/depth unavailable; CFE statistics returned September 4. September 1 leveraged net −26,258 is all-month positioning. Carry and correlation are path-dependent. [cftc-tff](https://www.cftc.gov/dea/futures/financial_lf.htm), [bls-calendar](https://www.bls.gov/schedule/2026/09_sched.htm).

Entry/gates: Unchanged −1.20 to −1.05 points plus a named October-relative-to-December catalyst edge, target and exit/roll plan. Current spread is outside the band. Target: No validated numeric target; the calendar trade remains incomplete. Stop: −2.00 points conditional only; no holding or observed executed stop. Invalidation: No calendar-specific catalyst edge or inability to define an appropriate holding/roll horizon.

Catalyst: PPI/ECB September 10 and CPI September 11 remain pending. Review their differential effect on October versus December; a generalized volatility rise is not a calendar edge. Lifecycle: October 21 and December 16 expiries both precede January 6. No unchanged four-month holding; require a named expiry-relative edge and explicit pre-expiry exit/roll with verified safety clocks.

Evidence required: A specific reason for October to outperform December after policy data; a rise in both legs is insufficient. Retirement of the calendar watch is warranted if the inflation release leaves no distinct edge and complete target. Coverage: Exact official date-only settlements, expiry table, stale positioning vintage and data-availability caveats reviewed. No execution clocks, target, current margins/depth or horizon-compatible package plan.

### OAT / Bund — Wait; score unrated

FOATZ26 / FGBLZ26 · hedge TBD. Official December settlement coverage is current, but the economic hedge remains undefined. ECB is inside 12 hours; stay unrated and Wait.

Quote and trend: Eurex September 9 official dSettle for December maturity 20261208; retrieved 20:35:05.777/20:35:06.235 EDT. Date-only, with no invented publication clock. Raw difference −7.70 is not a yield spread or duration-neutral metric. −0.90 / −0.70 (Separate official settlement changes; not a hedge return). [eurex-oat](https://www.eurex.com/api/v1/overallstatistics/34684?busdate=20260909&filtertype=overview), [eurex-bund](https://www.eurex.com/api/v1/overallstatistics/34656?busdate=20260909&filtertype=overview).

Return direction and counterargument: Expected direction and numerical return remain indeterminate until a hedge is specified. Falling separate settlements cannot establish the performance of a France/Germany tightening package. CTD switches, delivery baskets, special repo and country-specific basis can overwhelm a sovereign spread view. EUR 100,000 face values do not mean equal duration.

Physical, curve/carry, positioning and liquidity: Unrated: price coverage does not make the hedge scorable. September 9 outright volumes 401,116/1,676,615 and OI 862,212/2,072,658 do not establish package depth. Raw difference −7.70 vs −7.50 is not a sovereign yield change. CTD switches, repo specialness, conversion factors and basis dominate naïve correlations; futures DV01 requires CTD cash DV01 divided by its conversion factor. [eurex-deliverables](https://www.eurex.com/ex-en/data/clearing-files/notified-deliverable-bonds-conversion-factors), [eurex-ctd](https://www.eurex.com/ex-en/markets/int/financing-of-futures-ctds), [ecb-webcast](https://www.ecb.europa.eu/press/press_conference/html/index.en.html), [ecb-weekly](https://www.ecb.europa.eu/press/calendars/weekly/html/index.en.html).

Entry/gates: Verify CTD bonds, conversion factors, futures-adjusted DV01 and repo/basis; then define signed integer hedge counts and a futures-native entry, target and stop. No nominal 1:1 shortcut. Target: Undefined pending a valid futures hedge and metric. Stop: No approved futures-package stop; an old cash yield-spread limit cannot be tested with two raw futures prices. Invalidation: Failure of a properly specified France/Germany tightening thesis; exact test awaits hedge verification.

Catalyst: ECB decision September 10 08:15 EDT, 11h35m46s after cutoff; press conference 08:45 and projections 09:45. Wait for catalyst with no pre-event edge or new risk. Lifecycle: Both December API rows identify December 8 last-trade date, before January 6. The present source extraction does not verify exact cessation or broker safety clocks; any future position needs a sourced pre-delivery close/roll plan, not automatic delivery.

Evidence required: The ECB outcome and a December OAT/Bund tightening package with sourced hedge mathematics. Demotion to context-only research is warranted if hedge inputs remain unavailable after the ECB decision. Coverage: Exact official settlements, outright volume/OI, ECB calendar and deliverables reviewed. No validated CTD selection, conversion factors, DV01, repo/basis, integer ratio, target/stop, fresh FX or margins.

Wheat benchmark reconciliation: current delayed legs fell 12.75¢ wheat and 6.50¢ soy versus the prior 756.50/1,336.25. Separately, USDA-reproduced September 9 closes 744.25/1,331.75 yield 1.117702, down 1.7699% from the separately retained September 8 USDA pair 761.50/1,338.50 (1.137841). These separate settlement comparisons do not replace the prior delayed benchmark. September 9 local contango is 15.50¢ December-to-March wheat and 6.50¢ January-to-March soy: unequal intervals, not a net package carry estimate. [USDA Illinois Grain Bids](https://www.ams.usda.gov/mnreports/ams_3192.pdf).

## Newly retired unentered watches

### Platinum / gold — Exit thesis, retire

September 9 delayed PLF27 $1,924.70 at 20:22 EDT / GCG27 $4,475.90 at 20:12 EDT = 0.430014075, +2.8386% vs prior 0.418144796; ten-minute skew, indicative. Separate vendor settlements $1,940.70/$4,496.50 = 0.431602358. Prior priced-in score 4/5; final analytical score 2.5/5 before retirement, not conviction or a buy signal. The higher ratio further consumes old remaining value. The watch is retired on September 9 after the failed WPIC gate; no defended retest, automatic reverse, fill or P&L. Thin stocks and an H2 deficit remain context only. Old .410–.414 retest band, .390 invalidation and .445–.460 targets are archival. January physical delivery requires verified December pre-notice safety if ever reconsidered. [PLF27](https://www.wsj.com/market-data/quotes/futures/PLF27), [GCG27](https://www.wsj.com/market-data/quotes/futures/GCG27).

The required sequence was supportive WPIC → verified close above .414 → defended .410–.414 retest. The first condition has now failed; a price print above .414 cannot repair it, and no defended retest is established. WPIC reports Q2 surplus 244 koz, H1 surplus 548 koz and forecast H2 deficit 283 koz; investment-demand expectations were cut 601 koz, although projected year-end stocks still cover only 3.4 months. That is enough to preserve metals context, not the original deficit-led entry watch. The failed first condition warrants retirement of the entry watch. The lower analytical priced-in score does not mean improved entry value: the higher price ratio and weaker deficit premise both argue against this old expression. Its September 6 return row is removed, not refreshed or recast as performance. [WPIC Q2 report](https://platinuminvestment.com/files/354715/WPIC_Platinum_Quarterly_Q2_2026.pdf), [WPIC latest summary](https://platinuminvestment.com/supply-and-demand/platinum-quarterly).

Liquidity/crowding: evening PL/GC volume 92/161 and OI 24,275/32,480 (OI date unspecified) do not provide executable depth. September 1 legacy non-commercial net longs were 15,000 platinum and 228,124 gold, not post-WPIC positioning. Delayed January platinum above October and February gold above December are not synchronous physical premiums; historical H1 lease-rate easing is not a current lease quote. Gold may outperform on growth/liquidity stress. PL is 50 oz, GC 100 oz, each $0.10/oz tick; both physically deliverable. Jan 27/Feb 24 last-trade dates are rule-derived, not exact dated-calendar verification; January delivery requires an earlier verified December pre-notice exit if ever reconsidered. [CFTC platinum](https://www.cftc.gov/dea/futures/deanymesf.htm), [CFTC gold](https://www.cftc.gov/dea/futures/deacmxsf.htm), [PL rulebook](https://www.cmegroup.com/rulebook/NYMEX/1a/105.pdf), [GC rulebook](https://www.cmegroup.com/rulebook/COMEX/1a/113.pdf).

The platinum/gold watch is retired following the September 9 WPIC release. Reinstatement would require a newly underwritten prospective thesis, a supportive physical basis and complete price/liquidity/lifecycle terms; there is no automatic short or reverse.

### ULSD / RBOB — Exit thesis, retire

September 9 HOF27 $4.2680 / RBF27 $2.7905 per gallon, both vendor delayed labels 20:26 EDT = $1.4775 indicative, +$0.1162 vs prior $1.3613. Same minute is not synchronized ticks. Separate vendor settlements 4.2984/2.7975 = $1.5009. Final priced-in score 4/5 remains context, not conviction. Original $1.12–$1.18 entry, $1.45 target, $1.03 stop are archival; no target-profit, fill, P&L or automatic reverse. STEO supports distillate tightness but extreme energy/geopolitical repricing overtook the plan. The unentered watch is retired before the September 10 petroleum release; that data informs market context only. January last trade December 31 is rule-derived and precedes the horizon end. [HOF27](https://www.wsj.com/market-data/quotes/futures/HOF27/contracts), [RBF27](https://www.wsj.com/market-data/quotes/futures/RBF27/contracts).

The fundamental case for tight distillate remains strong: STEO forecasts U.S. distillate below 100 million barrels in September and below its five-year low through end-2026 and much of 2027, reflecting export pull, lower foreign refinery supply, maintenance and seasonal demand. This forecast is not a new inventory observation. Older August 28 data show a 0.8 million barrel build but inventories 14% below average, refinery utilization 98%, and four-week distillate demand down 6% y/y. The weak-demand and supply-recovery counterarguments still matter. The former expression already prices beyond its untouched target, so an old profitable outcome cannot be awarded to a trade never entered. [EIA products](https://www.eia.gov/outlooks/steo/report/petro_prod.php), [EIA weekly summary](https://ir.eia.gov/wpsr/wpsrsummary.pdf).

Evening HO/RB volumes 89/164 and OI 21,758/39,645 are not full-day depth. September 1 managed-money net longs 20,344 ULSD and 89,215 RBOB are not spread positioning. Equal gallons do not remove crude, freight, product-specification or refinery basis. Each leg is 42,000 gallons with $0.0001/gal tick = $4.20; physically deliverable. December 31 last trade, January 4 intentions and January 5 exchange notice are rule-derived, not execution-certified clocks. Both January legs terminate before the horizon ends. [CFTC petroleum](https://www.cftc.gov/dea/futures/petroleum_sf.htm), [ULSD rules](https://www.cmegroup.com/rulebook/NYMEX/1a/150.pdf), [RBOB rules](https://www.cmegroup.com/rulebook/NYMEX/1a/191.pdf).

The products plan is retired because it has already been overtaken. The September 10 noon petroleum release is still pending and becomes context-only. Reinstatement needs a newly underwritten prospective plan or fresh qualifying pullback with all gates; no automatic reuse, target-raising or reverse.

## Previously retired contexts

### Crude normalization — remain retired

Keep normalization retired: sustained transport/production restoration is not established within the horizon; no automatic reverse. September 9 delayed CLV26 $96.75 at 20:15 EDT / CLZ26 $89.40 at 20:23 EDT = $7.35/bbl indicative, +$0.91 versus prior $6.44 and $1.10 above old $6.25 invalidation. Vendor settlements 96.05/89.32 imply $6.73. No position, stop execution or profit/loss. STEO puts most normalization in Q2 2027, outside January 6. Rule-derived last trades September 22/November 20; no inherited roll. [Direct source](https://www.eia.gov/outlooks/steo/report/global_oil.php).

### Dow / Nasdaq rotation — remain retired

Remain retired: no newly established durable rotation or economically hedged reinstatement edge. Exact delayed YMU26 52,537 at 20:25:24.710 EDT / NQU26 29,458.25 at 20:25:56.069 EDT; indicative raw ratio 1.783439274, −0.2573% vs 1.788040 prior. Legs are 31.359 seconds apart. Not beta/notional neutral or package P&L. September 18 last-trade dates; no automatic roll or reverse. [Direct source](https://www.cmegroup.com/CmeWS/mvc/quotes/v2/318).

### Cocoa long — remain retired

Remain retired: a modest rebound and provisional logistics headline do not re-establish a physical/catalyst edge. CCZ26 September 9 date-only vendor settlement $5,951/t vs $5,913 prior: +$38, +0.6427%. Separate delayed last $5,937 at 13:29 EDT. Vendor volume/OI 14,298/75,444, not verified exchange-final flow; Dec–Mar delayed curve +$116/t. Traceability difficulties are search-summary-only context, not verified scarcity or demand recovery. December physical delivery lies within horizon; exact notice/margins unavailable. No reinstatement or reverse. [Direct source](https://www.wsj.com/market-data/quotes/futures/CCZ26).

### CAD / EUR — remain retired

Remain retired/unavailable: no new exact 6CZ26/6EZ26 pair, valid hedge or reinstatement edge verified. September 6 historical reference only: 6CZ26 0.72585 / 6EZ26 1.16545, from the September 6 monthly review. No current quote, September 9 move, new ratio, expected return or reversal is asserted. Missing current data is not evidence of zero change. [Direct source](https://www.cftc.gov/dea/futures/financial_lf.htm).

Crude's September 1 NYMEX physical WTI managed-money net long 94,281 (weekly +10,261) is not current October–December flow. Current overnight CLV/CLZ volumes 7,262/3,671 and OI 220,770/210,494 likewise do not validate current package depth. One CL contract is 1,000 barrels, $0.01 tick = $10. [CFTC petroleum](https://www.cftc.gov/dea/futures/petroleum_sf.htm), [WTI rules](https://www.cmegroup.com/rulebook/NYMEX/2/200.pdf), [October WTI](https://www.wsj.com/market-data/quotes/futures/CLV26), [December WTI](https://www.wsj.com/market-data/quotes/futures/CLZ26). Cocoa's traceability headline is search-summary-only, not an independently verified scarcity outcome; 10-tonne physical contracts and December notice risk need fresh calendars before any new plan. [Reuters cocoa](https://www.reuters.com/world/africa/ivory-coast-cocoa-sellers-struggle-use-new-traceability-system-2026-09-09/), [ICE cocoa](https://www.ice.com/products/7/Cocoa-Futures).

## Dated scenario expectations, not performance

The September 6 full-review tree is retained: persistent friction 40%, renewed supply shock 25%, transport normalization 20%, growth break 10%, AI/liquidity disappointment 5%. These subjective historical weights sum to 100% but are not market-implied odds, current forecasts, asset allocations or evidence of a stable regime. The intensified energy-shock assessment above is qualitative; this daily monitor does not manufacture a numerical recalibration. Horizon remains January 6.

September 6 displayed expected returns remain sugar +2.6% and yen +2.5% on outright notional, and wheat/soy +2.3% on absolute gross leg notional, with old displayed asymmetries 2.1:1, 2.0:1 and approximately 1.6:1 respectively. Those are archived assumptions, not current entry offers, leveraged margin returns, probabilities or paper returns. Platinum's old +1.2% row is removed on retirement; products never had a validated numerical row. SOFR, gas, VIX and bonds retain null expected returns and unestimable asymmetry. Higher sugar and lower wheat/soy entry marks would mechanically help unchanged terminal-price sensitivity; yen's small retreat also helps mechanically. That does not override the qualitative gates. No fresh numeric mark sensitivity or return forecast was calculated today, and no old forecast is relabeled current.

## Next 72 hours and event reconciliation

All times below are Toronto EDT. Window ends September 12 at 20:39:14 EDT. Reviewed China CPI/PPI, WPIC, STEO and June ECEC are reported above; event-specific outcomes are read, never inferred merely from a passed clock.

| Event | Toronto time | Cutoff status / implication |
|---|---|---|
| [BOJ Masu speech — wait for catalyst](https://www.boj.or.jp/en/about/calendar/index.htm) | Sep 9 21:30 | Sep 10 10:30 JST = Sep 9 21:30 EDT, 50m46s after cutoff. Pending; no pre-event edge or new risk. |
| [ECB policy decision](https://www.ecb.europa.eu/press/press_conference/html/index.en.html) | Sep 10 08:15 | 11h35m46s after cutoff. No rate outcome inferred; Wait. |
| [U.S. August PPI](https://www.bls.gov/schedule/2026/09_sched.htm) | Sep 10 08:30 | Pending policy/inflation evidence; a surprise must be read, not presumed. |
| [ECB press conference](https://www.ecb.europa.eu/press/press_conference/html/index.en.html) | Sep 10 08:45 | Pending guidance and country-spread implications. |
| [ECB staff projections publication](https://www.ecb.europa.eu/press/press_conference/html/index.en.html) | Sep 10 09:45 | Pending growth/inflation assumptions; European local labels converted using September daylight time. |
| [EIA natural-gas storage](https://ir.eia.gov/ngs/schedule.html) | Sep 10 10:30 | Pending weekly storage outcome. |
| [EIA petroleum headline / remaining report](https://www.eia.gov/petroleum/supply/weekly/) | Sep 10 12:00 / 14:00 | Holiday shift: headline noon EDT, remaining content 14:00 EDT. Products/crude are retired; outcome informs context, not automatic reinstatement. |
| [Japan August CGPI](https://www.boj.or.jp/en/about/calendar/index.htm) | Sep 10 19:50 | Sep 11 08:50 JST = Sep 10 19:50 EDT, pending. |
| [U.S. CPI / real earnings](https://www.bls.gov/schedule/2026/09_sched.htm) | Sep 11 08:30 | Pending. Reassess policy and calendar edge after actual release. |
| [Michigan preliminary sentiment](https://www.sca.isr.umich.edu/) | Sep 11 10:00 | Pending inflation expectations and growth context. |
| [Lagarde Paris interview](https://www.ecb.europa.eu/press/calendars/weekly/html/index.en.html) | Sep 11 10:00 | Pending communication after the policy meeting. |
| [USDA WASDE](https://www.usda.gov/oce/commodity/wasde) | Sep 11 12:00 | Read actual report, then delivered basis/demand; the band alone cannot activate the watch. |
| [CFTC Commitments of Traders](https://www.cftc.gov/MarketReports/CommitmentsofTraders/ReleaseSchedule/index.htm) | Sep 11 15:30 | Expected September 8 positions; confirm actual release and report category before claiming liquidation or crowding change. |
| [Treasury buyback — announced, not completed](https://www.reuters.com/world/us-treasury-buy-up-6-billion-sept-10-buyback-operation-2026-09-09/) | Sep 10 Time unverified | Up to $6bn announced; exact operation/result time unverified. No completed result claimed. |
| [Lagarde September 9 dinner speech — content unverified](https://www.ecb.europa.eu/press/calendars/weekly/html/index.en.html) | Sep 9 Time unverified | Scheduled time passed before cutoff, but content was not verified. Elapsed time is not a reviewed outcome. |

USDA holiday-adjusted export-sales timing and weekly claims exact-date confirmation remain unverified, so no invented clock is shown. China CPI/PPI is now explicitly reviewed from NBS and corroborating Reuters material; its September 8 21:30 EDT publication was after the prior baseline. September 9 Lagarde speech content remains unreviewed. Treasury buyback is announced, not completed. FOMC September 15–16 and BOJ September 18 are later catalysts outside this 72-hour window, not today's outcomes.

## Urgency, invalidations and what changes next

- Immediate: retire the unentered platinum and products watches and their obsolete entry triggers. Record no fills or profits; never reverse automatically.
- Before new yen risk: read Masu at 21:30 EDT; price must return to the unchanged 0.00635–0.00639 band with intact policy and durable-demand evidence. No pre-event edge exists.
- After ECB/PPI/storage/CPI: bonds need CTD/conversion-factor/DV01/repo/basis and integer hedge verification. SOFR needs a real post-data edge plus entry/target/adverse-leg terms. Gas needs ≥1.10 premium, target, winter/LNG/deliverability and safe specs. VIX needs its cheap band plus named expiry-relative edge and target. Unsupported watches warrant demotion or retirement under the evidence conditions above.
- After WASDE: wheat needs stronger delivered basis/demand and two actually reviewed export cycles to test its failure rule; completed failed cycles remain zero before the event. Sugar needs an actual new CFTC vintage, verified physical supply, and a prospective failed-retest sequence, with unknown UNICA timing preserved.
- Every future candidate: verify current exact-contract initial margins, quote-time volume, executable depth, bid/ask, adverse leg-level loss scenario, currencies and verified safe exit dates. A sourced band touch does not complete physical/catalyst/liquidity gates. No simulated fill is inferred from these research conditions.

No latest active quote breaches a conditional stop, but snapshot monitoring cannot reconstruct an unobserved path. The products target crossing and retired crude invalidation breach are research facts, not executed risk events. The strongest adverse shared scenario remains prolonged supply disruption with higher inflation/yields, weather/LNG or refinery stress, and leg-correlation/basis breaks. No existing risk exit is needed because the paper book is empty.

## Simulated paper portfolio — separate from research

Initial simulated funding remains $1,000,000 USD. This is a cash-only no-trade review, with no positions, pending instructions or fills. No risky-asset or FX marks are required for the all-USD cash balance. Research-held yen does not mean a simulated long. Retiring an unentered watch closes no paper lot.

| Metric | Starting | Ending / current review |
|---|---:|---:|
| NAV | $1,000,000 | $1,000,000 |
| Economic USD cash / available capital | $1,000,000 | $1,000,000 |
| Realized gross / net P&L | $0 / $0 | $0 / $0 |
| Open P&L / net total P&L | $0 / $0 | $0 / $0 |
| Commissions / modeled slippage | $0 / $0 | $0 / $0 |
| Positions / contracts / pending intentions | 0 / 0 / 0 | 0 / 0 / 0 |
| Gross notional / margin reserve | $0 / $0 | $0 / $0 |
| Gross leverage / planned open risk | 0× / $0 | 0× / $0 |
| Theme risk / unavailable held marks | $0 / none | $0 / none |
| Total / benchmark / excess return | 0% / 0% / 0% | 0% / 0% / 0% |
| Observed drawdown / max drawdown | 0% / 0% | 0% / 0% |

There were no deposits, withdrawals, orders, fills, rolls or costs in this review. Deployed risk and margin usage are zero. Simulation cost assumptions remain $3.50 per contract per side and one adverse outright tick per side; observed session volume does not guarantee executable depth. Cash and benchmark earn 0%. Zero P&L is not evidence of alpha; win rate, profit factor, annualized return and Sharpe remain unavailable. No held lot faces an imminent roll or delivery; the research lifecycle limitations above remain material to any future trade.

No new paper trade is proposed because substantive price, physical and catalyst conditions remain unsatisfied alongside missing execution inputs. Holding cash is an affirmative no-trade decision.

## Source dates and limitations

Current research cutoffs, forecast input dates, positioning vintages, quote observation times and retrieval times are separate. The NBS CPI page, official NBS PPI index and corroborating Reuters report were verified after the market cutoff, but the underlying releases predate it. The failed full PPI-page retrieval is not used to substantiate a numerical claim. No broker, personal account, real order or external execution was accessed.