# Oil’s recovery has a delivery problem

Research cutoff: **October 14, 2026, 06:15 Toronto / 10:15 UTC**. This daily review covers changes since 04:07 UTC, during the European session following Asian trading. The four-month horizon remains **February 13, 2027**. U.S. September CPI has **not** been released at this cutoff.

## Decisions first

**Hold the existing one-contract March SOFR long; add zero.** The exact SR3H27 last remains 95.480. Its original assumed entry, stop and target are unchanged. No new assumed paper trade is justified this edition.

The opportunity ranking did change:

1. **Gold becomes the strongest feasible new challenger.** One April micro now fits the normal starter budget, but its rate-sensitive distribution overlaps the existing SOFR position and remains vulnerable to the imminent inflation test.
2. **Avoid new crude-normalization shorts at these prices and revised probabilities.** The IEA report is now available; “wait for its release” is obsolete. We also underwrite the opposite direction and genuine micro diesel below.
3. **Copper remains Wait, with its overnight strike uplift removed.** A cheaper price is partly offset by less supply-risk support.
4. **Avoid a new Nasdaq short based on an imminent AI-demand break.** ASML provides counterevidence, while the lower sale price removes the prior modeled edge.

These are economic decisions, not quote-precision vetoes. There are no pending orders and no real transactions.

## What changed since the overnight report

The **IEA’s October 14 report** estimates September land-based oil inventories fell **113 million barrels**, while oil afloat rose **103 million**. It projects a year-end oil deficit despite recovering production, and continued diesel constraints into 2027. These are agency estimates and forecasts, not guaranteed outcomes. The publication date and report contents are verified; its exact release minute is not. [IEA October report](https://www.iea.org/reports/oil-market-report-october-2026)

Our inference is about delivery, not a claim that all floating cargo is stranded: cargo arrivals, transit duration, refinery conversion and product availability can differ from headline wellhead output. A small change in aggregate stocks can therefore hide a much larger change in the buffers available on land. Conversely, faster arrivals and demand destruction can reverse that pressure. This makes a blanket crude short less attractive without making every oil long attractive.

**Escondida:** Reuters reported a preliminary agreement with the supervisors’ union, still awaiting member ratification. The report was available before 10:00 UTC. We remove the overnight probability uplift; we never assumed actual lost output. [Reuters, October 14](https://www.reuters.com/business/world-at-work/bhps-escondida-supervisors-union-avert-strike-with-preliminary-agreement-2026-10-14/)

**ASML:** its release at 05:00 UTC reports €11.662 billion quarterly sales and a 57.5% gross margin, with full-year sales guidance of €44–45 billion. Management says planned 2027–28 supply will not fully meet expected customer demand. That is management’s forecast, not independently verified future orders. [Company release](https://www.globenewswire.com/news-release/2026/10/14/3380319/0/en/asml-reports-11-7-billion-total-net-sales-and-5-0-billion-net-income-in-q3-2026.html)

Our inference: this challenges a near-term AI capital-spending collapse. It does not establish fair value for the Nasdaq or prevent higher yields from hurting expensive equities. The investor call is still ahead.

China’s price releases and Singapore’s MAS decision were already covered overnight. They remain dated context, not new developments in this edition. No new CFTC positioning observation is claimed.

## Price and risk conventions

All last prices in the table are exact native-exchange observations from read-only [IBKR market data](https://www.interactivebrokers.com/), dated **October 14**. Bid/ask books have no independent exchange timestamp in the returned snapshot. They are conservative study references, not claims of guaranteed executable size or synchronized spread fills.

| Contract | Exact last | UTC last clock | Observed bid / ask |
|---|---:|---|---|
| March SOFR SR3H27 | 95.480 | 10:03:28 | 95.480 / 95.485 |
| December SOFR SR3Z26 | 95.745 | 10:05:48 | 95.745 / 95.750 |
| December Micro WTI MCLZ26 | $94.27/bbl | 10:06:50 | 94.26 / 94.28 |
| March Micro WTI MCLH27 | $90.76/bbl | 10:06:23 | 90.72 / 90.78 |
| March standard ULSD HOH27 | $4.4471/gallon | 10:06:40 | 4.4398 / 4.4425 |
| December standard ULSD HOZ26 | $4.7960/gallon | 10:10:00 | 4.7947 / 4.7961 |
| December Henry Hub NGZ26 | $3.623/MMBtu | 10:07:44 | 3.622 / 3.624 |
| March Henry Hub NGH27 | $2.819/MMBtu | 10:08:02 | 2.818 / 2.820 |
| March micro copper MHGH27 | $6.8360/lb | 10:02:20 | 6.8400 / 6.8415 |
| December micro copper MHGZ26 | $6.7330/lb | 10:07:36 | 6.7325 / 6.7335 |
| April micro gold MGCJ27 | $4,232.00/oz | 10:01:17 | 4,234.00 / 4,234.40 |
| April 1oz gold 1OZJ27 | $4,242.75/oz | 06:53:12 | 4,233.00 / 4,233.75 |
| March corn ZCH27 | 495.50 cents/bu | 10:05:09 | 495.25 / 495.50 |
| March wheat ZWH27 | 680.25 cents/bu | 10:07:29 | 680.00 / 680.50 |
| March soybean oil ZLH27 | 68.50 cents/lb | 10:06:14 | 68.50 / 68.52 |
| March soybean meal ZMH27 | $360.80/short ton | 10:04:06 | 360.80 / 360.90 |
| December S&P micro MESZ26 | 7,869.00 | 10:07:10 | 7,869.00 / 7,869.25 |
| December Nasdaq micro MNQZ26 | 30,763.75 | 10:07:01 | 30,763.25 / 30,763.75 |
| December 10-year Treasury ZNZ26 | 104.859375 | 10:06:10 | 104.859375 / 104.875000 |

One whole contract or package is sized before ranking. At NAV $1,001,826.875, the usual 0.10–0.25% starter range is approximately **$1,002–2,505**, and the 0.50% thesis ceiling is **$5,009**. Risk studies use 1.25 times price-to-invalidation loss plus disclosed costs. Round-trip friction includes $3.50 commission per contract per side and one adverse tick per side, with any additional allowance stated. Using a book side and full friction is deliberately conservative.

Scenario weights and endpoints are our judgments, not market-implied probabilities. Gap branches can pass through stops; none represents a maximum possible loss. Sensitivity is a diagnostic, not a universal requirement to survive every combined stress.

## SOFR: retain the small position, not a second entry

March’s 95.480 last implies approximately **4.520% compounded overnight funding** for its future reference quarter. December–March remains a separately observed **26.5-basis-point** implied-rate difference, unchanged from the overnight report. It is not a new dovish curve move, guaranteed carry or a literal number of Fed hikes.

The dated September Fed median policy projection and Williams’s October 13 limited-hike remarks remain support. Energy pass-through and persistent core inflation remain the strongest challenge. Policy-rate forecasts differ from compounded secured funding because of timing, repo basis and risk premia. [Fed projections](https://www.federalreserve.gov/monetarypolicy/fomcprojtabl20260916.htm) · [Williams reporting](https://www.reuters.com/world/feds-williams-currently-sees-no-need-rate-hike-2027-bloomberg-news-reports-2026-10-13/)

The existing position is **+1 SR3H27 from assumed 95.475**, stop 95.100, target 95.850. Intended exit **February 12** precedes the March 17–June 16 reference quarter and June 15 last trade: this is a repricing thesis, not a final-settlement outcome within our horizon. The contract pays $25 per basis point. [CME mechanics](https://www.cmegroup.com/education/articles-and-reports/understanding-sofr-futures)

For a hypothetical **new** unit at the 95.485 ask with $32 full friction:

| February exit/path | Weight | Reference | Net P&L |
|---|---:|---:|---:|
| Limited tightening, then plateau | 45% | 95.850 | +$880.50 |
| Additional limited tightening | 30% | 95.600 | +$255.50 |
| Persistent tightening | 20% | 95.350 | −$369.50 |
| Inflation gap through stop | 5% | 94.850 | −$1,619.50 |
| Weighted result | 100% | | **+$318.00** |

One new unit risks **$1,235.13**, 0.123% NAV. Ten probability points from favorable to ordinary stop-first leave +$130.50; ten points to the severe gap leave +$68. Both changes together give −$119.50. Doubling full friction alone leaves +$286.

**Hold, add zero.** Existing exposure remains positive under the base and individual stress cases; the new information does not justify doubling the same policy-path bet. Broad sequential core pressure and guidance validating materially higher 2027 funding rates would trigger an exit review even before the numeric stop. A benign CPI also requires repricing before any add.

Original entry-based ledger risk remains **$1,191.375**; current mark-to-stop with future exit friction is $1,203.50. Original lifecycle modeled EV is +$355.50, while mark-forward modeled EV is +$346.50. These are forecasts from different starting points, not actual returns. The carried **$5,000 paper margin reserve** is a conservative assumption, not a verified current broker initial-margin requirement.

## Energy: challenge the short, then test the long

December/March WTI book-based prompt premium widened to **$3.48–3.56**, versus $3.30–3.35 overnight. This curve change favors nearer-term scarcity more than a generic four-month long. Neither spread is a synchronized execution.

The latest verified Gulf recovery survey remains **October 13**, with 329,059 barrels/day of oil shut in. No newer quantified Saudi production loss was verified; the administrative-building incident must not be multiplied into repeated outages. [MMA/BSEE survey](https://www.bsee.gov/newsroom/latest-news/statements-and-releases/press-releases/mma-monitors-gulf-response-isaias7) · [Saudi statement](https://www.spa.gov.sa/en/N2700900)

### Crude shorts: new evidence outweighs a better sale price

We reduce fast-normalization weights and increase disruption risk. March retains more normalization weight because it extends into early 2027.

| One-contract short study | Exit plan | Weights and exit prices | Net branch P&Ls | Net EV | Buffered ordinary risk |
|---|---|---|---|---:|---:|
| December MCL at 94.26; $9 costs | Nov 12 | 25/35/25/15%; 84/91/96.50/120 | +1,017 / +317 / −233 / −2,583 | **−$80.50** | $289 |
| March MCL at 90.72; $12 costs | Feb 10 | 30/35/20/15%; 78/87/95/120 | +1,260 / +360 / −440 / −2,940 | **−$25** | $547 |

The illustrated $120 gap risks are $3,226.50 and $3,672 respectively, considerably larger than ordinary stop risk. **Avoid new shorts under these distributions.** Retaining old probabilities solely to manufacture a better EV from higher entry prices would ignore the changed evidence.

### Prompt crude long: feasible size, fragile edge

One December MCL at **94.28**, structural stop **84**, $9 costs and **November 12** exit gives:

| Scenario | Weight | Exit | Net P&L |
|---|---:|---:|---:|
| Continued physical constraint | 30% | 110 | +$1,563 |
| Manageable but persistent scarcity | 35% | 98 | +$363 |
| Delivered supply normalizes | 25% | 84 | −$1,037 |
| Rapid resolution/demand shock | 10% | 75 | −$1,937 |

EV is **+$143**; ordinary buffered risk **$1,294**, or 0.129% NAV. The $75 gap study risks $2,419. One whole contract fits, but five points favorable-to-gap make EV **−$32**; ten base-to-stop leave just **+$3**. **Wait**, without tightening the structural stop merely to improve the ratio.

A March long at 90.78, stop 80, $12 costs and 25/35/30/10% endpoints 110/94/80/65 has **zero EV**, with $1,359.50 ordinary risk. Deferred exposure does not solve the problem. WTI also carries regional and crude-versus-product basis risk; it is not a diesel-crack position.

### Micro diesel: better mechanism, too much structural risk

There is a real smaller contract: **Micro NY Harbor ULSD, MHO, 4,200 gallons**, with a $0.0001/gallon tick worth $0.42. It is cash-settled and ends trading one business day before its corresponding standard contract. [CME rulebook](https://www.cmegroup.com/content/dam/cmegroup/rulebook/NYMEX/4/434.pdf)

March MHO’s own returned close lacked a usable observation date. That does **not** veto a paper trade: same-month standard HO is a defensible labeled proxy. At the observed standard ask **4.4425**, with structural stop **3.80**, February 12 exit, and 30/40/25/5% endpoints **5.50/4.60/3.80/3.00**, one micro gives net **+$4,413.50 / +$633.50 / −$2,726.50 / −$6,086.50**, EV **+$591.50**.

The $28 study friction includes $7.84 basic commissions/ticks plus a modeled $20.16 proxy/liquidity allowance—not an exchange fee. Ordinary buffered risk is **$3,401.13, 0.3395% NAV**, above the usual starter band; the $3.00 gap study is **$7,601.13**, above the thesis budget as a stress comparison. Five points favorable-to-gap leave +$66.50; combining that with ten base-to-stop makes −$269.50.

**Wait; no starter-size exception.** The 3.80 invalidation allows meaningful supply normalization. Moving it to 4.00 merely to fit would not remove that risk. A standard HO is ten times larger. The March micro’s February 25 last trade permits a February 12 exit without a roll.

### Gas: winter repricing is already occurring

December–March Henry Hub last-price premium widened from a dated $0.701 to **$0.804/MMBtu**. The October 8 EIA report, for October 2, still shows total stocks **68 Bcf above** the five-year average; South Central stocks were **2.3% below** their average. Those regional differences are old data newly emphasized, not new releases. [EIA storage](https://ir.eia.gov/ngs/ngs.html)

A December 1,000-MMBtu micro long using the standard **3.624 ask proxy**, 2.90 stop and $9 costs, with 25/40/25/10% outcomes 4.50/3.65/2.90/2.50, yields **−$73 EV**, risk **$914**. **Wait** for October 15 storage and weather-adjusted demand/LNG evidence. A December trade exits or is re-underwritten in November; no free roll to February is assumed.

## Metals: cheaper gold improves more than cheaper copper

### Copper loses its strike uplift

The latest retrieved LME figures remain **October 13**: cash premium $115.50/t, down from $164.50, and stocks 229,650 tonnes, down 1,625. No October 14 inventory figure was verified. COMEX March-over-December book bounds are **10.65–10.90 cents/lb**. London prompt tightness and deferred U.S. premium are different location/tenor signals. [LME data compilation](https://www.westmetall.com/en/markdaten.php?action=table&field=LME_Cu_cash)

Remove the overnight five-point favorable-case uplift after the preliminary deal report; restore **35/40/25%**. One March micro at **6.8415 ask**, 2,500 lb, 6.35 structural stop and $9.50 costs:

| February 12 outcome | Weight | Price | Net P&L |
|---|---:|---:|---:|
| Wider supply constraint persists | 35% | 7.60 | +$1,886.75 |
| Tight but manageable | 40% | 7.00 | +$386.75 |
| Demand/location-basis reversal with gap risk | 25% | 5.80 | −$2,613.25 |

EV **+$161.75**, risk **$1,545.44**, 0.154% NAV. Five points favorable-to-adverse give **−$63.25**. Keeping the overnight weights would show +$236.75, but that would ignore new evidence. **Wait** for ratification, operating information and a new physical update—not an old entry-band test. February 12 exit precedes February 24 last trade.

### Gold is now a genuine small-contract choice

Dated World Gold Council evidence illustrates both sides: strong September ETF buying coexisted with lower prices and futures liquidation as yields and the dollar rose. Flows alone did not dominate opportunity cost. No new ETF observation was verified this session. [WGC, October 7](https://www.gold.org/goldhub/research/gold-market-commentary-september-2026)

Preserve February 12 endpoints **4,600/4,320/3,920**, weights **30/45/25%**, and structural stop **4,050**:

| April expression | Net branch P&Ls | EV | Buffered stop risk |
|---|---|---:|---:|
| One 10oz MGC at 4,234.40; $9 costs | +3,647 / +847 / −3,153 | **+$687** | **$2,314; 0.231% NAV** |
| Eight 1oz contracts at 4,233.75; $60 costs | +2,870 / +630 / −2,570 | **+$502** | **$1,897.50; 0.189% NAV** |

One MGC **now fits**; the prior size objection is obsolete. It also avoids eight sets of commissions. Ten points favorable-to-gap leave MGC +$7, or −$2 with doubled costs. A ten-point favorable-to-ordinary-stop-first stress instead leaves +$137.

**Wait, with a ranking upgrade.** The economic uncertainty is whether energy-led inflation lifts real yields or instead weakens growth enough to support gold. Our unchanged 30/45/25 distribution is not sufficiently reliable to add a second rate-sensitive exposure immediately before the discriminating release. This is not a blanket CPI ban; a new price and real-rate response can justify entry. The 1oz book’s limited displayed size is disclosed but is not an execution gate under the assumed-fill convention. April covers the February exit without a December roll.

## Agriculture: smaller contracts solve size, not weak economics

No fresh agricultural fundamental release supersedes October 13 Crop Progress. Corn harvest at 33% versus 39% normal remains context; winter-wheat planting delays are not proof of permanent yield loss. [USDA report](https://esmis.nal.usda.gov/sites/default/release-files/796104/prog4126.txt)

All following studies exit February 12, before relevant delivery or micro expiry boundaries.

| Expression | Side-aware reference; stop | Net endpoints; weights | EV | Buffered risk |
|---|---|---|---:|---:|
| One March corn short | 495.25; 535; $32 costs | −2,769.50 / +730.50 / +2,230.50 at 550/480/450; 30/40/30% | +$130.50 | $2,516.38 |
| Five March micro corn shorts, labeled standard proxy | Conservatively rounded 495.0; 535; $60 costs | −1,435 / +315 / +1,065; same endpoints/weights | **+$15** | **$1,310** |
| One March wheat long/corn short | 680.50−495.25 = 185.25; spread stop 170; $64 costs | +2,173.50 / +423.50 / −1,826.50 at 230/195/150; 25/45/30% | +$186 | $1,017.13 |

One standard corn is now about $12 above the preferred starter ceiling. We actively tested **500-bushel micro corn**, $5 per cent and a half-cent tick worth $2.50. Five micros fit, but provide little net expected return; they are not 1,000-bushel contracts. [CME micro specifications](https://www.cmegroup.com/markets/agriculture/micro-ag-futures)

**Wait corn** for marketed harvest pressure and export/basis evidence. Only 2.61 probability points favorable-to-adverse erase the standard study’s edge. **Wait wheat/corn** for relative export/fieldwork evidence: five points favorable-to-adverse make EV −$14. Two packages fit but do not improve the distribution.

**Avoid soy oil/meal re-entry.** Using standard March oil ask 68.52 and meal bid 360.80 as micro-compatible proxies gives metric 60×oil−10×meal = $503.20 per pair. Three pairs, 35/40/25% endpoints 870/480/53.60 and $61.20 friction yield **−$41.10 EV** and $1,747.20 buffered adverse-endpoint risk. The result improves from overnight but remains negative. October 15 NOPA crush and oil stocks are the next joint test.

## Remaining liquid challengers

December index and Treasury studies exit or are re-underwritten **November 20**, not held without costs/roll analysis through February.

| Study | Reference / stop; full friction | One-contract risk | Endpoints; weights | EV |
|---|---|---:|---|---:|
| S&P micro long | 7,869.25 / 7,650; $9.50 | $1,379.81 | 8,100/7,900/7,650/7,350; 40/30/25/5% | +$94.25 |
| Nasdaq micro short, revised | 30,763.25 / 31,650; $8 | $2,224.88 | 29,600/30,500/31,650/33,000; 25/30/40/5% | **−$201.50** |
| 10-year Treasury short | 104.859375 / 105.265625; $38.25 | $546.06 | 103.875/104.750/105.265625/106; 30/40/25/5% | +$142.22 |

**Avoid new Nasdaq short:** the unchanged earlier tree now gives only +$3.50; the lower bid alone removed $228.50 of EV. ASML justifies transferring five points from the strongest selloff to the ordinary adverse branch. **Wait MES:** ten points favorable-to-stop make −$130.75. **Wait ZN:** five points favorable-to-gap leave +$35.97; adding doubled friction makes −$2.28. The old Treasury position remains closed.

**Euro short — Wait.** No refreshed exact native euro future was established. October 5’s 1.1242 standard December proxy remains historical, not a current sale. More importantly, policy divergence remains unconvincing against the limited-Fed-tightening thesis. Reassess relative policy after CPI; do not turn spot into an exact futures quote.

**INR — context watch only.** The dated dollar-window intervention mechanism remains relevant; no new exchange-listed implementation or policy change is claimed. **Retired March MES, directional soybeans and silver long — Avoid new entry.** No fresh evidence warrants reviving these old plans; no new quote or hypothetical profit is invented. This is a daily six-asset-class challenger review, not a new monthly universe scan.

## Next 72 hours

| Event | UTC time | Decision relevance |
|---|---|---|
| U.S. CPI and real earnings, Oct 14 | 12:30 | Underlying breadth, real rates and SOFR/gold distributions |
| ASML investor call, Oct 14 | 13:00 | Demand visibility versus near-term equipment deliveries |
| Fed Beige Book, Oct 14 | 18:00 | Pricing power, wages and demand |
| U.S. PPI, Oct 15 | 12:30 | Input pressure and pass-through |
| EIA gas storage, Oct 15 | 14:30 | Weather-adjusted balance and regional inventories |
| EIA petroleum inventories, Oct 15 | 16:00 | Holiday-delayed release: products, runs and delivered stocks |
| NOPA crush and oil stocks, Oct 15 | 16:00 | Processing and oil inventories together |
| Import/export prices, Oct 16 | 12:30 | External price transmission |
| Industrial production, Oct 16 | 13:15 | Industrial demand and capacity |

[BLS](https://www.bls.gov/schedule/2026/10_sched.htm) · [Federal Reserve](https://www.federalreserve.gov/newsevents/2026-october.htm) · [ASML](https://www.investor.asml.com/quarterly-results) · [EIA oil schedule](https://www.eia.gov/petroleum/supply/weekly/schedule.php) · [Gas schedule](https://ir.eia.gov/ngs/schedule.html) · [NOPA](https://www.nopa.org/wp-content/uploads/2026/02/NOPA-ONLY-Crush-Reporting-Release-Dates-2026-FINAL.pdf)

Escondida ratification and Gulf operating updates are event-driven; no guaranteed publication time is assumed. Every Wait has a specific evidence question and next reassessment, not an indefinite requirement for perfect confirmation.

## Simulated portfolio

The sole position remains **+1 SR3H27**, assumed bought October 13 at 22:16:53 UTC at **95.475**, using the exact 95.470 observation at 22:06:51 plus one adverse tick. No new fill, cancellation, addition or real order occurs this edition.

The mark-only ledger update at **October 14 10:16:59 UTC** uses the exact **95.480 at October 14 10:03:28 UTC**. Cash remains **$1,001,814.375**, open gross P&L **+$12.50**, NAV **$1,001,826.875**, and cumulative closed-trade net P&L **+$1,817.875**. Original entry-based open risk is $1,191.375; modeled margin reserve $5,000. Valuation time is distinct from the quote’s observation time. No book reset, backdated execution or inferred intraday trigger is used.

Research and an illustrative paper portfolio only. Futures losses can exceed initial margin; stop orders and scenario stress levels do not cap all possible losses.