@sdelsad/commodity-desk-daily 1.0.17 → 1.13.2

package/conversions.md

@@ -0,0 +1,156 @@
+# Conversion drills — one per episode, email only
+
+A daily unit drill that appears **in the email only, never in the audio**. The
+point is muscle memory: a trader converts in their head, mid-sentence, without
+reaching for a calculator.
+
+Pick the drill by episode number: **drill index = ((N − 1) mod 12) + 1**. After
+episode 12 the cycle restarts at drill 1, and so on — repetition is the
+mechanism, not a bug.
+
+Each day, reproduce the drill's **rule**, its **fast method**, and its
+**worked examples**, then invent **one fresh exercise** with different numbers
+from any previous run (vary the figures every cycle so it is never the same
+question twice). Put the exercise in the quiz area and its answer with the
+solutions.
+
+---
+
+## Drill 1 — Kilograms ↔ pounds
+
+**Rule:** 1 kg ≈ 2.2 lb · 1 lb ≈ 0.45 kg
+
+**Fast method:** kg → lb: double it, then add 10%. lb → kg: halve it, then take off 10%.
+
+- 70 kg → 140 + 14 = **154 lb** (exact 154.3)
+- 100 kg → 200 + 20 = **220 lb** (exact 220.5)
+- 200 lb → 100 − 10 = **90 kg** (exact 90.7)
+- 150 lb → 75 − 7.5 = **67.5 kg** (exact 68.0)
+
+## Drill 2 — Bushels ↔ tonnes, corn
+
+**Rule:** 1 tonne of corn ≈ 39.4 bushels (a bushel of corn is 56 lb)
+
+**Fast method:** tonnes → bushels: ×40, then shave 1.5%. Bushels → tonnes: ÷40, then add 1.5%.
+
+- 10,000 t → 400,000 − 6,000 = **394,000 bu**
+- 60,000 t Panamax → 2,400,000 − 36,000 ≈ **2.36 million bu**
+- 250,000 bu → 6,250 + 94 ≈ **6,344 t**
+
+## Drill 3 — Bushels ↔ tonnes, wheat and soybeans
+
+**Rule:** 1 tonne ≈ 36.7 bushels (a bushel of wheat or soybeans is 60 lb)
+
+**Fast method:** tonnes → bushels: ×37 is close enough, and ×37 = ×40 minus ×3. Bushels → tonnes: ÷37, i.e. ÷40 then add 8%.
+
+- 60,000 t → 2,400,000 − 180,000 = **2.22 million bu**
+- 25,000 t → 1,000,000 − 75,000 = **925,000 bu** (exact 918,000)
+- 500,000 bu → 12,500 + 1,000 ≈ **13,600 t** (exact 13,608)
+
+**Why it matters:** the corn factor and the wheat/soybean factor differ by 7%. Using the wrong one on a cargo is a six-figure error.
+
+## Drill 4 — Cents per bushel ↔ dollars per tonne
+
+**Rule:** ¢/bu × 0.394 = $/t for corn · ¢/bu × 0.367 = $/t for wheat and soybeans
+
+**Fast method:** corn: ×0.4, then shave 1.5%. Wheat/soybeans: ×0.37, i.e. take a third and add a tenth of it.
+
+- Corn at 465¢ → 186 − 3 ≈ **$183/t**
+- Soybeans at 1,182¢ → 394 + 39 ≈ **$434/t** (exact 434.0)
+- Wheat at 651¢ → 217 + 22 ≈ **$239/t** (exact 239.1)
+
+**Why it matters:** Chicago quotes cents per bushel, the rest of the world quotes dollars per tonne. Every export conversation crosses this line.
+
+## Drill 5 — Metric tonne ↔ short ton
+
+**Rule:** 1 metric tonne = 1.102 short tons · 1 short ton = 0.907 t
+
+**Fast method:** $/short ton → $/tonne: add 10%. $/tonne → $/short ton: take off 9-10%.
+
+- Soybean meal $318/short ton → 318 + 32 = **$350/t** (exact 350.5)
+- $400/short ton → **$441/t**
+- $500/t → 500 − 45 = **$455/short ton** (exact 453.6)
+
+**Why it matters:** Chicago soybean meal trades in short tons while the physical trade quotes metric. Forgetting is a 10% error.
+
+## Drill 6 — Celsius ↔ Fahrenheit
+
+**Rule:** °F = °C × 1.8 + 32
+
+**Fast method:** C → F: double it and add 30 (rough, and about 2° high in the mid range). F → C: subtract 30 and halve.
+
+- 30°C → 60 + 30 = **90°F** (exact 86)
+- 35°C → 70 + 30 = **100°F** (exact 95)
+- 95°F → 65 ÷ 2 ≈ **32°C** (exact 35)
+
+**Why it matters:** US crop weather is reported in Fahrenheit; the rest of the world thinks in Celsius. Pollination stress in corn starts around 32-35°C — know what that is on a US forecast.
+
+## Drill 7 — Millimetres ↔ inches of rain
+
+**Rule:** 1 inch = 25.4 mm
+
+**Fast method:** inches → mm: ×25 (×100 then ÷4). mm → inches: ÷25 (÷100 then ×4).
+
+- "1 to 3 inches across the Midwest" → **25 to 75 mm**
+- 0.5 inch → **13 mm**
+- 40 mm → **1.6 inches**
+
+**Why it matters:** rainfall forecasts drive grain prices, and the two systems appear in the same conversation constantly.
+
+## Drill 8 — Hectares ↔ acres
+
+**Rule:** 1 hectare = 2.47 acres · 1 acre = 0.405 ha
+
+**Fast method:** ha → acres: ×2.5, then shave 1%. Acres → ha: ×0.4, then add 1%.
+
+- 1 million ha → 2.5 m − 25,000 = **2.47 m acres**
+- 90 million acres of US corn → 36 m + 0.4 m ≈ **36.4 m ha**
+- 500 ha → **1,235 acres**
+
+## Drill 9 — Bushels per acre ↔ tonnes per hectare
+
+**Rule:** corn: 1 bu/ac ≈ 0.0628 t/ha · wheat and soybeans: 1 bu/ac ≈ 0.0673 t/ha
+
+**Fast method:** corn: divide by 16. Wheat/soybeans: divide by 15.
+
+- US corn at 182 bu/ac → 182 ÷ 16 ≈ **11.4 t/ha**
+- US soybeans at 53 bu/ac → 53 ÷ 15 ≈ **3.5 t/ha**
+- French wheat at 7.5 t/ha → 7.5 × 15 ≈ **112 bu/ac**
+
+**Why it matters:** US yields are quoted in bushels per acre, European and South American in tonnes per hectare. Comparing crops requires crossing.
+
+## Drill 10 — Dollars ↔ euros per tonne
+
+**Rule:** divide by the EUR/USD rate to go from $/t to €/t
+
+**Fast method:** at 1.10, $ → €: take off 9% (÷1.1 ≈ ×0.91). € → $: add 10%. At 1.05, it is 5% each way; at 1.20, take off 17% / add 20%.
+
+- $239/t wheat at 1.10 → 239 − 21 = **€218/t**
+- €205/t Matif at 1.10 → 205 + 20 = **$225/t**
+- $350/t at 1.05 → **€333/t**
+
+**Why it matters:** Matif quotes euros per tonne, Chicago dollars per bushel. Any Matif–CBOT arb crosses both a unit and a currency.
+
+## Drill 11 — Hundredweight (cwt) ↔ tonnes
+
+**Rule:** 1 cwt = 100 lb = 45.36 kg · 1 tonne = 22.05 cwt
+
+**Fast method:** $/cwt → $/t: ×22. $/t → $/cwt: ÷22, i.e. halve it three times and add a touch (÷22 ≈ ÷20 minus 9%).
+
+- Rice at $14/cwt → **$309/t**
+- Rice at $16.50/cwt → **$364/t**
+- $400/t → 400 ÷ 20 = 20, − 9% ≈ **$18.2/cwt**
+
+**Why it matters:** US rice and cotton quote per hundredweight or per pound while the export trade quotes per tonne.
+
+## Drill 12 — Dollars per tonne → whole-cargo money
+
+**Rule:** a Panamax is roughly 60,000 t; a Supramax roughly 55,000 t; a Handysize roughly 30,000 t
+
+**Fast method:** $/t → cargo: ×6, then move the decimal four places (×60,000). For a $ per bushel figure, convert to $/t first.
+
+- $11/t margin on a Panamax → 66 → **$660,000**
+- $2/t of demurrage exposure → **$120,000**
+- 30¢/bu on soybeans → $11/t → **$660,000** on 60,000 t
+
+**Why it matters:** the instinct that turns a small per-tonne number into the size of the cheque is what makes a desk argue over a quarter of a cent.

package/curriculum.md

@@ -0,0 +1,294 @@
+# Soft Commodity Trading — curriculum (30 episodes, 6 weeks)
+
+One brief per episode. Each brief is **dense on purpose**: the listener is a
+strong quantitative graduate who has done derivatives. A concept they can grasp
+in three minutes gets three minutes — then you go somewhere they could not have
+gone alone: the second-order effects, the edge cases, what actually breaks on a
+desk.
+
+Every episode carries **either two substantive concepts, or one concept taken to
+real depth** (mechanics, failure modes, worked numbers). Never one concept
+stretched to fill ten minutes.
+
+Read `covered.md` before writing: it lists what has aired, including which market
+pulse topics were used, so you build instead of repeat.
+
+Episode 1 = Monday 10 August 2026, one per weekday.
+
+---
+
+## Week 1 — The toolkit
+
+**1. The units and the language of the desk.** The primer. Everything a
+newcomer needs to stop being lost, delivered as the vocabulary of a working
+floor rather than a lecture. Units and how things are quoted: the bushel (a
+volume measure — 60 lb soybeans and wheat, 56 lb corn, hence different tonne
+conversions), cents per bushel, the 5,000-bushel Chicago contract, dollars per
+tonne in the rest of the world, metric vs short tons, cwt for rice and cotton,
+points and ticks, and what a "lot" means when someone says they bought fifty.
+Then the grammar of a quote: flat price vs a differential, why physical trades
+as "plus eighty" or "minus twenty" against a named futures month, month codes
+and how a month is named out loud, prompt vs deferred. Then the verbs: bid,
+offer, work, done, lifted, hit, washed out. *Use several dialogues here* — this
+is the episode where hearing it matters most. *Depth:* why the market quotes
+differentials at all instead of prices, and what that convention reveals about
+where the risk sits. Set expectations: some of this will feel foreign for a week,
+and that is the point — the glossary in the notes carries everything.
+
+**2. What a merchant does, and why basis is the whole game.** Two concepts, one
+argument. First: merchants are paid for transformation — space, time, form — not
+for prediction; physical vs paper, where paper is the hedge and never the bet;
+asset-light vs asset-heavy and what each asset unlocks; the ABCD-plus landscape
+in one pass. Second, immediately: because flat price is hedged away within
+minutes, what remains is the basis — long or short the basis, and basis as the
+price of logistics, quality and urgency. *Worked example:* one Santos→Qingdao
+Panamax carried through both halves — the direction-neutral margin, then the
+same cargo when the board moves a dollar (nets to zero) and the differential
+moves ten cents (that is the entire P&L). *Dialogue:* the same cargo quoted by
+two traders. *Depth:* the three risks that survive a "perfect" hedge, named and
+priced.
+
+**3. Futures plumbing and the shape of the curve.** Two halves. First: how a
+hedge is actually placed — lot sizes, tick values, contract months and which
+ones are liquid, the tickers shouted on a desk (ZW, ZC, ZS, ZM, ZL, KC, SB, CT),
+rolling, and margin calls as a *cash-flow* event that can kill a solvent trade.
+Second: the curve as information — cost of carry, full carry vs actual spread,
+contango/backwardation, and inverses as the market screaming for immediate
+supply. *Worked example:* a store-or-sell decision computed against full carry,
+then the same position when the curve inverts. *Depth:* why a hedger with a
+correct view can still be forced out by variation margin.
+
+**4. The physical chain, end to end.** Incoterms as a risk-allocation language
+(FOB / CFR / CIF): where title passes, where risk passes, who charters, who
+insures. Then the execution reality: laycan, nomination, load and discharge,
+laytime, demurrage and despatch, draft surveys, quality certificates and the
+documentary chain. *Worked example:* one FOB cargo from fixture to bill of
+lading, with a three-day delay priced in dollars. *Depth:* how a "small"
+execution failure becomes a contractual liability, and why execution desks are
+a profit centre, not admin.
+
+**5. Wheat: the map and the screens.** Classes and protein specs (SRW, HRW, HRS,
+durum, Black Sea milling) and why two wheats at the same flat price are not the
+same wheat. Russia's export dominance and how it redrew the map; the big
+importers and the politics of bread. Then the paper side: CBOT vs KC vs Matif —
+three exchanges pricing three different wheats — protein spreads, inter-exchange
+spreads, and the Black Sea cash market that trades on assessments with no
+futures at all. *Worked example:* a Matif–CBOT arb computed with FX and freight,
+both directions. *Depth:* hedging a Black Sea cargo with a contract that is not
+your wheat — cross-hedge risk made concrete.
+
+## Week 2 — Grains, oilseeds, and the numbers behind them
+
+**6. Corn, crop calendars and weather risk.** The US belt; feed vs ethanol
+demand and why corn is a demand story; Brazil's safrinha; Ukraine's swing role.
+Then the clock: northern and southern hemisphere cycles back to back, the
+critical windows (pollination, pod-set, flowering), and why a dry week in July
+outweighs a dry month in October. *Worked example:* corn-wheat substitution in a
+feed ration, priced. *Depth:* the anatomy of a weather premium — how it builds,
+how it decays, and why being right on the weather and wrong on the timing loses
+money.
+
+**7. WASDE and building a balance sheet.** The report that stops every grain
+desk, and the object it updates. Supply and demand line by line: production,
+imports, feed use, exports, ending stocks, and the *residual* where analysts
+earn their keep. Stocks-to-use as the tension gauge. Trading the surprise, not
+the number. *Worked example:* build a small balance sheet live, flex the yield
+by two bushels, and follow it through to stocks-to-use and price. *Depth:* why
+two competent analysts disagree on the same crop, and what that disagreement is
+worth.
+
+**8. The soybean complex and the crush.** One seed, three markets: beans, meal
+(a protein market), oil (an energy-adjacent market). Who grows, crushes,
+imports. Then the margin that runs the industry: board crush arithmetic with the
+conversion factors, oil share, how a crusher hedges (buy beans, sell products),
+and what plants do when the crush goes negative. *Worked example:* compute a
+board crush in dollars per bushel and per tonne, step by step. *Depth:* crush
+capacity as a physical constraint that caps the margin — and where the crusher's
+real optionality sits.
+
+**9. Vegetable oils and biofuels.** Palm as the volume king: BMD futures,
+Indonesian and Malaysian export policy as a price driver. Substitution economics
+across palm, soy, rape and sun oil — the complex trades as one system.
+Rapeseed/canola/sunseed specifics (Matif colza, Black Sea sunoil, EU crush).
+Then biofuels: how a mandate converts a political decision into a standing bid
+for a crop; ethanol, biodiesel, renewable diesel/HVO. *Worked example:* trace a
+mandate change through to implied demand and price. *Depth:* why agricultural
+balance sheets now contain an energy term, and what breaks when crude moves.
+
+**10. Freight: dry bulk and chartering.** Vessel classes (Handysize to
+Capesize), draft as a hard constraint, the Baltic indices. Voyage charter vs
+time charter, FFAs to hedge freight, bunkers. Freight as a *component of the
+arb*, not an afterthought. *Worked example:* the same cargo on two vessel
+classes, then an arb that dies when freight rallies twenty dollars before the
+vessel is fixed. *Depth:* why merchants charter rather than own, and what the
+freight position does to a book that looks flat.
+
+## Week 3 — Softs
+
+**11. Storage, elevation and trade flows.** Silos, elevators, terminals;
+elevation margin as a business of its own; blending to spec — turning two
+off-spec lots into one sellable cargo. Then flows: origin vs destination
+pricing, arb windows, replacement value, and how flows reroute when one leg
+moves. *Worked example:* a blend that creates value from discount material, and
+two origins competing for the same import tender. *Depth:* why owning the
+bottleneck asset in a congested port is worth more than any trading view.
+
+**12. Coffee: the market.** Arabica vs robusta — different plants, altitudes,
+markets. Washed vs natural processing; the origin map (Brazil, Vietnam,
+Colombia, Ethiopia, Honduras). KC and RC futures, certified stocks as the
+market's visible inventory, deliverable origins and grading. *Worked example:* a
+certified-stock drawdown and the front-month reaction. *Depth:* why the same
+"coffee" trades at two very different prices, and what certification actually
+guarantees.
+
+**13. Coffee: differentials, PTBF and volatility.** Origin differentials as the
+real price. Price-to-be-fixed contracts: the physical is agreed, the flat price
+fixed later — who carries fixing risk and when it bites. The chain from farmgate
+to roaster and where value sits at each hand-off. Then volatility: Brazilian
+frost and drought, Vietnamese robusta cycles, funds in a market too thin to
+absorb them. *Worked example:* a PTBF contract fixed at the wrong moment, cost
+quantified. *Depth:* the anatomy of a frost rally and its retracement — why the
+physical desk and the fund lose money in opposite ways.
+
+**14. Sugar.** Raws vs whites (No. 11 and No. 5). Brazil's Center-South as the
+swing supplier and the ethanol parity that lets mills choose between sugar and
+fuel — the single most important switch in the market. The white premium and
+refining economics; India's export policy as a recurring shock; tolling and
+destination refineries. *Worked example:* the sugar–ethanol switch computed in
+ATR terms, then a refiner's margin. *Depth:* sugar as a market with two demand
+curves (food and fuel) and what that does to the price floor.
+
+**15. Cotton, rice and juice.** Cotton: ICE CT, mill demand as the real driver,
+and on-call sales — the mechanism that forces mills to buy into a rally. Rice: a
+thin, policy-driven staple where futures matter less than government decisions.
+Juice: FCOJ, Brazil vs Florida, citrus greening as a structural supply story.
+*Worked example:* an on-call position that turns a rally into a squeeze; then an
+export ban's price path in rice. *Depth:* what makes a market "thin", and how
+position size must change when liquidity does.
+
+## Week 4 — The instruments and the craft
+
+**16. Spreads: calendar, inter-commodity, inter-exchange.** The spread as the
+market's own supply-and-demand thermometer — more honest than flat price. Old
+crop vs new crop; rolling. Wheat-corn, meal-oil, Matif-CBOT. Full arb math with
+FX and freight loaded in. *Worked example:* read three curves and say what each
+is telling you; then compute one inter-exchange arb both directions. *Depth:*
+why spreads are where physical desks express views, and how a spread position
+can carry more risk than the outright it replaced.
+
+**17. Options where hedgers actually use them.** Not pricing theory — usage.
+Fences and collars around a physical position; protecting a tender bid; skew as
+information about who is scared of what; volatility into report days and its
+collapse afterwards. *Worked example:* collar a long physical position into
+WASDE, with the P&L in three scenarios. *Depth:* why a hedger sells the wing
+they think is "free", and what that costs when the market gaps.
+
+**18. EFP, delivery and squeezes.** Exchange for physical: swapping a futures
+position for a cash position off-exchange, and why the physical world trades
+this way (EFP/EFS/AA). Then expiry: what actually happens at delivery, the
+short's options — deliver, roll, or buy back — load-out constraints, and how
+deliverable supply becomes the whole story. *Worked example:* an EFP that
+transfers a hedge with the cargo; then the arithmetic of a squeeze. *Depth:*
+famous corners and the structural conditions that make one possible.
+
+**19. Basis deep dive and origination.** What actually moves basis: freight,
+farmer selling, export demand, storage capacity, quality. Separating basis P&L
+from flat-price P&L. Then the buying side: farmers, co-ops, elevators, posted
+bids, deferred pricing, HTA contracts — and farmer selling psychology, why sales
+cluster and what that does to basis. *Worked example:* decompose a finished
+trade into flat, basis and spread; then a harvest basis collapse and recovery.
+*Depth:* relationships as infrastructure — why the same cargo has a different
+price depending on who is asking.
+
+**20. Destination, tenders and storage economics.** How large importers buy:
+tenders, terms, counterparties, import premiums, and pricing a bid backwards
+from the destination. Then store-or-sell decided with numbers: cash-and-carry
+with financing costs, Matif's volume storage regime and equivalents. *Worked
+example:* price a tender bid backwards including freight, FX and financing; then
+a full carry trade. *Depth:* the winner's curse in tenders — being cheapest is
+not the same as being right.
+
+## Week 5 — Risk, money, and information
+
+**21. The book and P&L attribution.** Position keeping in practice: physical
+long, paper short, by month and by location; what a position sheet looks like
+and how errors surface. Mark-to-market discipline. Then attribution: decompose a
+finished trade into flat price, basis, spread, freight, FX and financing.
+*Worked example:* read a small position sheet, find the unhedged month, then
+attribute a completed trade line by line. *Depth:* why attribution is how desks
+actually learn, and how a "good" P&L can hide a broken process.
+
+**22. Risk management and why hedges are never perfect.** Position limits, VaR
+and its limits, stress tests; who says no and why risk sits outside trading.
+Then every reason a hedge leaks: basis, quantity, quality, timing, FX,
+cross-hedge. *Worked example:* a "fully hedged" trade that loses money,
+dissected line by line; then a stress scenario applied to a realistic book.
+*Depth:* the difference between a risk you chose and a risk you inherited.
+
+**23. Trade finance, contracts and counterparty risk.** Letters of credit,
+borrowing base, working capital as the real constraint on a merchant's size —
+and why financing capacity is an edge competitors cannot copy quickly. Then the
+legal layer: GAFTA and FOSFA terms, defaults, washouts, string trades and
+circles, arbitration. Counterparty and country risk: sanctions, export bans,
+force majeure. *Worked example:* the cash-flow timeline of one cargo; then a
+washout settlement computed. *Depth:* what happens when the person on the other
+side simply does not perform.
+
+**24. Data sources and positioning.** Export inspections, vessel lineups,
+satellite imagery, crop tours, customs data — what is timely, what is noisy,
+what is late, and how to weight them. Then the COT report: managed money vs
+commercials, crowded trades, and positioning as a risk indicator rather than a
+signal. *Worked example:* one week of data and what it implied; then a crowded
+long unwinding. *Depth:* the information edge is mostly speed and
+interpretation, not secrets.
+
+**25. China, and reading market color.** Soybean imports, state reserves and
+auctions, COFCO — how one buyer's decisions move world balance sheets, and what
+destination crush margins tell you about the next purchase. Then the language of
+the market: how bids and offers are phrased, "trades at, done", replacement
+value, workable, indication — and what broker talk conveys beyond the number.
+*Worked example:* a reserve auction's ripple; then decode a realistic broker
+message line by line. *Depth:* what a desk hears in silence — when nobody is
+offering, that *is* the information.
+
+## Week 6 — Macro, and putting it together
+
+**26. The Black Sea and policy shocks.** Russian and Ukrainian export
+machinery, corridors, export taxes and quotas, war-risk premium — why one region
+reprices global wheat. Then policy as a market force in general: export bans,
+tariffs, mandate changes, retaliation, with case studies. *Worked example:* a
+corridor disruption priced through, day by day. *Depth:* how desks position for
+a decision that is binary and unforecastable — and why the answer is usually
+optionality, not a view.
+
+**27. Weather, ENSO and FX.** El Niño / La Niña playbooks by commodity — who
+wins, who loses, and how reliable the pattern actually is versus how confidently
+it is quoted. Then FX for commodity traders: the Brazilian real and farmer
+selling (the single most useful relationship in agriculture), the dollar and
+import demand, hedging FX on a physical book. *Worked example:* one ENSO cycle
+across three commodities; then a BRL move that unlocks farmer selling and
+collapses the basis. *Depth:* correlations that hold until they matter, and how
+to size for that.
+
+**28. The energy link.** Agriculture as an energy derivative: crude → ethanol →
+sugar; natural gas → fertiliser → planted acres and grain cost curves; bunkers →
+freight → the arb. The cost curve as a soft price floor. *Worked example:* trace
+a crude move all the way into a sugar price, then into next season's acreage.
+*Depth:* why the ag desk that ignores energy is trading a lagged version of
+someone else's market.
+
+**29. Life of a trade, and the desk around it.** One cargo from origination to
+discharge: every desk it touches, every risk it picks up and sheds, with dates
+and dollars. Then the ecosystem: traders, execution, analysts, risk, operations,
+finance, legal — who does what, and why the trader is the visible tip of a large
+machine. *Worked example:* the full timeline of one trade; then one problem
+cargo routed through five functions. *Depth:* where trades actually go wrong —
+almost never on the price view.
+
+**30. The houses, and the trader's checklist.** The arc from 1851 grain
+merchants to today's ABCD-plus landscape: consolidation, COFCO's rise,
+Bunge-Viterra, the shift toward processing and traceability — even-handed across
+firms. Then the capstone: the questions a good desk asks about any opportunity,
+in order, and where to go deeper (reports, data sources, habits worth keeping).
+*Depth:* what separates a trader who survives a decade from one who has a great
+year.

package/ep01.md

@@ -0,0 +1,144 @@
+# Commodity Desk Daily — Episode 1: The Units and the Language of the Desk
+
+*Monday, August 10, 2026 · ~10 min listen*
+
+## Key takeaways
+
+- The **bushel is a volume measure, not a weight** — a basket of roughly 35 liters, inherited from the English grain trade. Weight per bushel differs by commodity: **60 lb for soybeans and wheat, 56 lb for corn**, so the tonne conversions differ too: **≈36.7 bu/t** for beans and wheat, **≈39.4 bu/t** for corn.
+- Chicago quotes grain in **cents per bushel**; one futures contract — a **lot** — is **5,000 bushels** (≈136 t of wheat/beans, ≈127 t of corn). The rest of the world trades **dollars per metric tonne**: $6.40/bu wheat ≈ **$235/t**, which is what makes a $223/t Russian offer readable at a glance.
+- Two ton traps: the **metric tonne (2,204.6 lb)** vs the US **short ton (2,000 lb)** — 10% apart. Rice quotes in **$/cwt** (hundredweight = 100 lb); coffee, sugar and cotton in **cents per pound**.
+- The **tick** is the minimum move: ¼¢/bu in Chicago grains = **$12.50 per lot**. Softs count in **points** — 1/100 of a cent. A 10¢ move on 50 lots of corn = 40 ticks = **$25,000**.
+- Physical grain almost never trades at a full price. It trades as a **differential** against a named futures month — "**plus 70 over Z**" — using one-letter **month codes** (H March, K May, N July, U September, X November, Z December). **Prompt** = nearby shipment; **deferred** = further out.
+- The verbs are precise and binding: you are **bid** or you **offer** (commitments, not moods), you **work** a resting order, an offer gets **lifted**, a bid gets **hit**, offsetting contracts get **washed out** — and "**done**" creates a contract, voice first, paperwork later.
+- Why differentials at all? Because a grain price is **two risks glued together**: the futures leg is world risk anyone can shed on the screen in one click; the plus is the local part — port, window, quality — that no screen will take off your hands. The grammar mirrors the risk, and it lets an offer stay **firm** for ten minutes while the board moves.
+
+## Vocabulary
+
+| Term | Desk meaning |
+|---|---|
+| Bushel | Volume measure (~35 L): 60 lb of soybeans/wheat, 56 lb of corn |
+| Lot | One futures contract — 5,000 bu in Chicago grains |
+| Bushels per tonne | ≈36.7 for beans/wheat, ≈39.4 for corn |
+| Metric tonne / short ton | 2,204.6 lb vs 2,000 lb — 10% apart |
+| Cwt (hundredweight) | 100 lb; the quoting unit for US rice |
+| Tick | Minimum price move: ¼¢/bu = $12.50 per lot in Chicago grains |
+| Point | 1/100 of a cent per pound — how softs desks count moves |
+| Month codes | H=Mar, K=May, N=Jul, U=Sep, X=Nov, Z=Dec — said out loud ("plus 70 Z") |
+| Flat price | The full outright price level |
+| Differential / basis | The premium or discount to a named futures month ("plus 80") |
+| Prompt / deferred | Nearby shipment vs further down the curve |
+| Bid / offer | Firm prices to buy / sell — commitments, not moods |
+| "At" | Introduces the offer side: "462 bid, at 462½" |
+| Work | Leave an order resting with a broker |
+| Done | The word that makes a trade a binding contract |
+| Lifted / hit | The buyer took the offer / the seller sold into the bid |
+| Firm | A tradable quote that binds if accepted — often with a time limit |
+| Washout | Cancelling offsetting physical contracts by settling the difference |
+| Bag (coffee) | 60 kg — how the coffee trade counts volume |
+| WASDE | USDA's monthly World Agricultural Supply and Demand Estimates |
+
+## Market pulse (Monday Aug 10)
+
+Softs led. ICE **certified arabica stocks fell to 244,172 bags — a 2½-year low** — after slower Brazilian shipments following June rains; arabica jumped over 4% Friday and December held that range Monday. Robusta is the mirror image: certified stocks at a **4½-month high**. **October raw sugar printed new highs for the move** as Brazil's Center-South crushed sharply less cane in June and the new season runs behind last year. **December cotton rallied to its highest since mid-May** on a hot, dry US Cotton Belt and a softer dollar. Grains marked time ahead of **Wednesday's WASDE**; China booked 238,000 t of US soybeans and unknown buyers took 105,000 t of corn.
+
+---
+
+## QUIZ
+
+*Episode 1 — today's block only (no earlier episodes to draw on).*
+
+**Q1 — Done for fifty thousand tonnes.** December corn futures stand at $4.62½. An exporter offers Gulf corn at "plus 70 over Z" and a buyer says done for 50,000 metric tonnes. (a) Compute the full price per bushel and per metric tonne (corn ≈ 39.37 bu/t). (b) How many bushels did the buyer just commit to, and how many lots would hedge it one-to-one? (c) A colleague runs the same numbers using 36.7 bu/t "because that's the factor from the episode". What goes wrong, in bushels, lots and tonnes of exposure?
+
+**Q2 — Read the tape.** Your broker's morning, October shipment: *9:02 — "I show plus 90 offered, best bid plus 82." 9:40 — "Buyer lifted the 90 for 25,000 tonnes. Done." 11:15 — futures have fallen 15¢; the same seller re-offers October at plus 90.* (a) At 9:40, who moved — buyer or seller — and what exactly did the word "done" create? (b) At 11:15, is the new plus-90 offer cheaper, dearer, or identical compared with the 9:40 trade? Answer twice: once in flat-price language, once in differential language. (c) Why can a differential offer sit "firm for ten minutes" while a flat-price offer can't?
+
+**Q3 — The unit traps.** (a) A US seller quotes SRW wheat at "$228 per ton", meaning short tons. A buyer signs for 30,000 metric tonnes assuming metric. What price per metric tonne did the seller actually mean, and what is the total dollar gap on the contract? (b) Rough rice is quoted at $18.50/cwt. Convert to dollars per metric tonne. (c) Coffee "falls one-eighty on the day". How much is that in cents per pound, and what counting convention are you using?
+
+---
+
+ 
+
+ 
+
+ 
+
+## ▼ SOLUTIONS (spoilers) ▼
+
+**S1.** (a) Full price = 4.625 + 0.70 = **$5.325/bu**. Per tonne: 5.325 × 39.37 ≈ **$209.6/t**. (b) 50,000 t × 39.37 = **1,968,500 bu**; ÷ 5,000 = 393.7 → **≈394 lots**. (c) 36.7 bu/t is the **soybean/wheat** factor (60 lb bushels); corn bushels weigh 56 lb, so a tonne holds more of them. Using 36.7 gives 1,835,000 bu → 367 lots — an under-hedge of **~27 lots ≈ 133,500 bu ≈ 3,400 t** of corn left exposed to flat price. The trap: the bushel isn't one unit — it's one unit *per commodity*, which is exactly why the episode gave two conversion factors.
+
+**S2.** (a) The **buyer** moved: "lifted" means the buyer took the standing offer. "Done" created a **binding contract** for 25,000 t at October futures + 90¢ — voice first, confirmations later. (b) In **flat-price** terms the new offer is **15¢ cheaper**: the board fell 15¢ and the offer is board + 90, so the all-in dollar price fell with it. In **differential** terms it is **identical**: plus 90 before, plus 90 now — in the language of the desk, the seller "hasn't moved". Both statements are true at once; which one matters depends on which risk you're carrying (Episode 2 takes exactly this up). (c) A differential offer only exposes the seller to movement in the *local* component — the board leg reprices itself continuously and both sides can hedge it in one click. A flat-price offer silently bets that futures won't move while it sits: in a falling market it's instantly stale (or adversely selected in a rally). The differential convention is what lets physical offers survive minutes, not seconds, next to a live screen.
+
+**S3.** (a) A short ton is 2,000 lb = 0.90718 metric tonnes, so $228/short ton = 228 ÷ 0.90718 ≈ **$251.3/metric tonne**. Gap ≈ $23.3/t × 30,000 t ≈ **$700,000** on the contract — from one three-letter word ("ton") left unconfirmed. (b) $18.50/cwt × 22.046 cwt/t ≈ **$408/t**. (c) Softs count in **points**, 1/100 of a cent per pound: "one-eighty" = 180 points = **1.80¢/lb**. The habit to build: every quote comes with a unit convention attached, and the desk never says it out loud — you're expected to know.
+
+---
+
+## The episode, in writing
+
+### Eleven words, ten million dollars
+
+"Plus seventy over Z. Fifty thousand tonnes. Firm for ten minutes." Somewhere this morning, a sentence like that bought roughly ten million dollars of corn — and to anyone new to the business, not one of its eleven words means anything. This opening episode is the decoder: the units, the grammar of a quote, and the small set of verbs that do the industry's business. By the end, that offer reads itself.
+
+### A basket, not a weight
+
+Start with the strangest unit on the floor. The **bushel** is a *volume* measure — a basket of roughly 35 liters — inherited from the English grain trade, which measured grain by the basket long before anyone weighed it. America kept the bushel; the rest of the world moved to tonnes.
+
+Because it's a volume, its weight depends on what's in it: **60 lb** for soybeans and wheat, **56 lb** for corn. That single fact drives the two conversion factors a grain trader uses all day:
+
+| | lb/bu | bu per metric tonne | 5,000-bu lot |
+|---|---|---|---|
+| Soybeans, wheat | 60 | ≈36.7 | ≈136 t |
+| Corn | 56 | ≈39.4 | ≈127 t |
+
+Chicago quotes in **cents per bushel**, and one futures contract — one **lot** — is 5,000 bushels. So "we bought fifty lots of corn" means a quarter of a million bushels, about 6,300 tonnes: counted in lots, moved in tonnes, quoted in cents. Three languages for one pile of grain.
+
+The conversion is worth practicing once: Chicago wheat at $6.40/bu × 36.7 ≈ **$235/t**. Suddenly a Russian export offer at $223/t means something to your eye. (Different wheat, different port — comparing them *properly* is a later episode — but the numbers now live on the same axis.)
+
+Two remaining ton traps: the world's **metric tonne** is 2,204.6 lb, while US domestic markets often use the **short ton** of 2,000 lb — 10% apart, and a costly thing to leave ambiguous. Rice quotes in dollars per **hundredweight** (cwt = 100 lb); coffee, sugar and cotton in cents per pound.
+
+Finally the **tick**, the minimum price move: a quarter-cent per bushel in Chicago grains, worth **$12.50 per lot**. Softs desks count in **points** — hundredths of a cent ("coffee up one-twenty" = 1.2¢/lb). Tick math turns squawk into money: a 10¢ move against fifty lots of corn is 40 ticks × $12.50 × 50 = **$25,000**.
+
+### How a futures order sounds
+
+> **TRADER:** Where's Dec corn?
+> **BROKER:** Four sixty-two bid, at four sixty-two and a half.
+> **TRADER:** Buy me fifty at the half.
+> **BROKER:** Done. Fifty lots at four sixty-two and a half.
+
+Eight seconds of business. "Dec" is the December contract. "Bid" is where buyers are paying, and the small word "**at**" introduces the offer: $4.62 bid, offered at $4.62½. And "**done**" is not filler — done means the trade exists. On a desk, done is a contract; the paperwork comes later.
+
+### The grammar of a physical quote
+
+Physical grain — real tonnes in a real port — speaks a second dialect. It almost never trades at a full price. It trades as a *distance from the futures board*: plus eighty, minus twenty, always against a named month. That distance is the **differential**, or the **basis**.
+
+The months carry one-letter codes, said out loud: **H** March, **K** May, **N** July, **U** September, **Z** December — and **X** November, the soybean harvest month. Hence "Gulf corn plus seventy Z", "beans plus eighty X". Add two words of tense: **prompt** (shipment now or near it) and **deferred** (further down the curve). The same corn can be plus 70 prompt and plus 50 deferred — and that gap is information.
+
+> **BUYER:** Gulf corn, first half September, where are you?
+> **SELLER:** I can offer plus seventy-two over Z.
+> **BUYER:** I'm a buyer at plus sixty-eight.
+> **SELLER:** Can't get there. Plus seventy, fifty thousand tonnes, firm for ten minutes.
+> **BUYER:** Done. Seventy over Z, first half September.
+
+That seller's line is the sentence the episode opened with. Notice what never got said: a price. Both parties carry the board in their heads; the only number they argued about was the plus. And "firm for ten minutes" is an offer with a fuse — a **firm** offer is tradable, binding the moment someone says done, which is exactly why sellers put a clock on it.
+
+### The verbs
+
+There are about seven, and they are precise. You are never "interested in buying" — you are **bid**. You never "consider selling" — you **offer**. Both are commitments, not moods. To **work** an order is to leave it resting with a broker. When a buyer takes an offer, it was **lifted**; when a seller sells into a bid, it was **hit** — direction is baked into the verb. And when two offsetting physical contracts between the same parties are cancelled against each other, settled for the money difference instead of shipping grain both ways, the trade was **washed out**.
+
+> **BROKER:** Your plus eighty offer — I've got a buyer indicating seventy-five.
+> **DESK:** Work seventy-eight. Firm.
+> **BROKER:** He lifts it. Done at seventy-eight for twenty-five thousand tonnes.
+> **DESK:** Confirmed. What's on the bid side?
+> **BROKER:** Someone hit a seventy-two bid this morning. I'd call your market seventy-two, seventy-eight.
+
+Two trades, and the desk knows where its market lives: 72 bid, 78 offered. Nobody published that number — it exists only in the traffic. That is why desks pay brokers, and why the language matters.
+
+### Why a plus, and never a price
+
+The question underneath the whole episode: why does physical grain quote differentials at all? Because a grain price is really **two risks glued together — and only one of them is yours**. The futures leg is world risk: weather, funds, a WASDE print — public, violent, and sheddable on the screen in one click, so the conversation doesn't waste breath on it. The plus is the local part: this port, this shipment window, this quality — the part no screen will take off your hands, and therefore the part professionals negotiate.
+
+The grammar mirrors the risk: the board carries the world; the plus carries your problem. It's also what lets a physical offer survive — plus 70 can stay firm for ten minutes while the board moves a nickel, where a flat-price offer would be stale in seconds. The convention is what makes trading real grain next to a live screen possible at all.
+
+### Market pulse recap
+
+Softs led Monday: ICE certified arabica stocks at 244,172 bags (a 2½-year low) against robusta stocks at a 4½-month high — one market draining, the other refilling; October raw sugar at new highs for the move on Brazil's weak June crush; December cotton at its best since mid-May on a hot, dry Cotton Belt and a softer dollar. Grains waited on Wednesday's WASDE while China booked 238,000 t of beans and unknown buyers took 105,000 t of corn.
+
+*Next — Episode 2: Flat price vs basis. The screen says one number; a cargo is worth another. What a desk actually does with the plus you just learned to hear.*