Part I. The Contract

Thales and the olive presses

Next video
Do this now: A new week: the chain of 4 March 2026, weekly expiry Tuesday 10 March 2026, lot 65. Row 1: the 25000 call at 98.60. Row 2: the 24000 put at 167.30. For each row write: (a) the buyer's maximum loss in rupees; (b) the seller's maximum loss, as a number or "no ceiling"; (c) each side's break-even on 10 March, before costs: the Nifty finish at which each side stops losing, and which side of it the buyer, and the seller, come out ahead. Work it in your notebook first, then check the answer in the workbook. Desk Notes, Lesson 1. Thales and the olive presses The Options Desk · Part I. The Contract · Chapter 1. The Oldest Bet Every market figure here comes from NSE's end-of-day prices for 30 September 2026 (weekly expiry Tuesday 6 October 2026), at least 30 days old on publication. Education, not advice. Neel, the presenter, is an AI. Rohan, Meera and Kabir are fictional. THE QUESTION THIS LESSON ANSWERS Rohan has a payment page open: Rs 4,999 a month for Kabir's "weekly Nifty strangle, 90% win rate". He can repeat Kabir's pitch, but he cannot yet answer two plain questions about it. What is the most Kabir can win in a week, and what is the most he can lose? This lesson gives the words and the arithmetic needed to answer both, for any row of an option chain. THE OLDEST VERSION OF THE BET Aristotle tells a story about Thales of Miletus, a Greek philosopher of the sixth century BCE (Politics, Book I, chapter 11). People mocked Thales because philosophy made no money. He studied the heavens, and while it was still winter he concluded that the coming olive harvest would be a large one. He had little money, so he did not buy olives, trees or presses. He paid small deposits for the use of every olive press in Miletus and on Chios at harvest time. In winter nobody else wanted the presses and nobody bid against him, so the deposits were cheap. The harvest was large, every grower needed a press at once, and Thales rented them out on his own terms. Two readings of the story exist. Aristotle told it as a lesson in cornering a market: Thales held every press, so he could set the price. Finance writers call it the first recorded option. Both readings are fair. What makes it an option is what Thales stood to lose. Had the harvest been poor, he would simply not have used the presses. He would have lost his deposits and nothing more. He had bought a right, not a duty. That asymmetry is the whole idea of an option. One side pays a small, known amount now for the right to act later, only if acting pays. The other side takes the small amount and accepts the duty to perform if asked. THE FIVE PARTS OF EVERY OPTION Every option since Thales has the same five parts. 1. The underlying. The thing the right is about. For Thales, the use of the presses; in this course, the Nifty 50 index. 2. The strike. The price at which the right lets you deal. 3. The expiry. The day the right ends. 4. The premium. What the buyer pays the seller for the right. 5. Two sides. The buyer (the holder, or long) owns the right. The seller (the writer, or short) has the duty if the buyer uses it. There are two kinds of right. - A call is the right to buy the underlying at the strike. It is worth something at expiry only if the underlying ends above the strike. Thales held a call on olive presses. - A put is the right to sell the underlying at the strike. It is worth something at expiry only if the underlying ends below the strike. At expiry, with the underlying at S and the strike at K, per unit: - a call pays max(S − K, 0); - a put pays max(K − S, 0). The seller pays exactly what the buyer receives. Before any costs, an option is a transfer between two people: what one gains, the other loses. A REAL ROW: THE BUYER'S SIDE The printed chain for this course is NSE's end-of-day file for 30 September 2026, weekly options expiring Tuesday 6 October 2026. Nifty 50 closed at 22,620.45. One lot is 65 units. Worked example 1. Buying the 23100 call. It closed at 19.25 a unit. - Premium for one lot: 19.25 × 65 = Rs 1,251.25. - Buyer's worst case: Nifty finishes at or below 23,100 on Tuesday, the call pays nothing, and the buyer loses the Rs 1,251.25 already paid. There is no further loss. - Buyer's break-even at expiry: strike + premium = 23,100 + 19.25 = 23,119.25. Above that level the buyer is in profit. - At, say, 23,300: the call pays 23,300 − 23,100 = 200 a unit. Net 200 − 19.25 = 180.75 a unit, or 180.75 × 65 = Rs 11,748.75 on the lot. Every chain price is quoted per unit of the index. Multiply by the lot to get rupees. THE SELLER'S SIDE, AND THE NUMBER NOBODY PRINTS Kabir sits on the other side of rows like these. His strategy sells, and on this chain one of his two legs is the 22100 put at 15.55. Worked example 2. Selling the 22100 put. - Premium received: 15.55 × 65 = Rs 1,010.75. - Best case: Nifty finishes at or above 22,100. The put expires worthless and the seller keeps Rs 1,010.75. Nothing can make it more. - Worst case: the put lets its buyer sell Nifty to the seller at 22,100. The lowest an index can go is zero. Then the seller owes 22,100 a unit, less the 15.55 collected. - Maximum loss: (22,100 − 15.55) × 65 = 22,084.45 × 65 = Rs 14,35,489.25, about Rs 14.35 lakh. - Break-even: strike − premium = 22,100 − 15.55 = 22,084.45. Nobody expects Nifty at zero; that figure is an edge of the arithmetic, not a forecast. The point is the shape. The seller's gain stops at Rs 1,011. The loss stops only where the index stops. Worked example 3. Selling the 23100 call. The seller collects Rs 1,251.25. The call pays its buyer everything above 23,100, and an index has no upper limit. A short call's maximum loss therefore has no ceiling. At 23,600, for example, the seller pays 500 − 19.25 = 480.75 a unit, or Rs 31,248.75 a lot. These three results hold for every option, so learn them as rules. Position | Most it can gain (per unit) | Most it can lose (per unit) Long call | No ceiling | The premium Short call | The premium | No ceiling Long put | Strike − premium | The premium Short put | The premium | Strike − premium THE TRAP: "I COLLECT PREMIUM, SO I CAN'T LOSE MUCH" The premium tells a seller the most they can win. It says nothing about the most they can lose. A seller has two separate numbers, and the second one decides whether an account survives. Kabir's reel speaks only of the first. Throughout this course, every position is described by both numbers before anything else is said about it. KABIR'S STRANGLE, AS IT STOOD THAT WEEK Kabir's rule, which this course follows from lesson to lesson: on the first trading day after a weekly expiry, at 9:30, sell one Nifty call and one Nifty put about 2% from spot on the next weekly expiry, and hold to expiry with no stop. On the product, "about 2%" is ten strike steps of 50 points from at-the-money spot: 500 points each side. On the 30 September chain those legs are the 23100 call at 19.25 and the 22100 put at 15.55. - Premium collected: 19.25 + 15.55 = 34.80 points, or 34.80 × 65 = Rs 2,262 a lot. - Best case: Nifty finishes between 22,100 and 23,100; both options expire worthless; Kabir keeps Rs 2,262 before costs. - Worst case on the call side: no ceiling. - Worst case on the put side: (22,100 − 34.80) × 65 = Rs 14,34,238 if Nifty went to zero. Lesson 2 draws this. The whole premium is credited against the put side, because both legs cannot pay out at once. What happened. On Tuesday 6 October 2026 Nifty closed at 22,776.10, up 0.69%. Neither strike was reached and both options expired worthless. The seller kept Rs 2,262 before costs. Lesson 5 shows that holding to expiry cost about Rs 52 in charges under the course's stated brokerage assumption. These are end-of-day prices, not a backtest; Part IV runs the backtest. Most weeks look like this, which is why the strategy is easy to sell. The course is about the other weeks. THE NEXT-OLDEST STORY: DUTCH TULIPS During the Dutch tulip mania of 1636 to 1637, bulbs were traded by forward contracts: promises to buy at a fixed price later, binding on both sides. Prices collapsed in February 1637. The States of Holland then suspended the contracts. In May 1638 the city council of Haarlem, which other cities followed, let a buyer walk away from a contract by paying 3.5% of its price (Garber, 1989, working from Posthumus). After the fact, each contract had become an option, with a premium of 3.5%. A duty was turned into a right, and the price of the right was set more than a year after the crash, by a city council, not by a market. Some historians, Thompson (2007) among them, argue that the February 1637 decision itself had already made the contracts options; that remains their reading, not settled fact. GLOSSARY - Option: a contract giving its buyer a right, not a duty, to buy or sell an underlying at a set price up to or on a set date. - Call / put: the right to buy / the right to sell at the strike. - Underlying: the asset or index the option is written on (here, the Nifty 50). - Strike (K): the price at which the option lets its holder deal. - Expiry: the last day of the option's life. Nifty weekly options expire on Tuesday (Lesson 4). - Premium: the price of the option, quoted per unit; times the lot for rupees. - Lot: the number of units in one contract. 65 for Nifty 50 from the January 2026 series. - Long / short: having bought / having sold. - Writer: another word for the seller of an option. - Break-even (at expiry): the underlying level at which the position neither gains nor loses. For a long call, K + premium; for a long put, K − premium. - Maximum loss: the worst result the payoff allows, in rupees for one lot. - Payoff: what the option pays at expiry, before subtracting the premium. - Strangle: a call and a put on the same expiry at different strikes, usually both out of the money. Kabir sells one. FOR THE CURIOUS: WHAT A WIN RATE CANNOT TELL YOU 1. At a fair price, the seller's expected profit is zero, whatever the win rate. Price an option the way a desk does: the premium is the discounted expected payoff under the risk-neutral measure Q (Part III builds this; Hull's chapters on binomial trees and on Black-Scholes derive it). For a short put, P = e^(−rT) E_Q[(K − S_T)⁺]. The seller's P&L at expiry is P e^(rT) − (K − S_T)⁺, so its Q-expectation is exactly zero. Split that expectation into wins and losses: if w is the Q-chance of keeping something and Ā_w, Ā_l the average win and loss, then w·Ā_w = (1 − w)·Ā_l. A seller whose strike gives a 90% Q-chance of winning must, at the fair price, have an average loss nine times the average win. A high win rate is just how the fair price of a far out-of-the-money option looks, not evidence of an edge. Any edge must come from the gap between the market's Q and the real-world P. Part V measures that gap for Nifty (the variance risk premium; Carr and Wu, 2009). 2. The market's own rough odds for Kabir's week. The course sheet (Black-76 on the chain's forward, r = 6%) gives the 23100 call a delta of +0.111 and the 22100 put a delta of −0.083. Delta is only a rough stand-in for the risk-neutral chance of finishing in the money (Lesson 27 gives the exact relation, N(d₁) against N(d₂)). The two events cannot both happen, so the rough Q-chance that one leg finishes in the money is about 0.111 + 0.083 ≈ 19%. That leaves about 4 weeks in 5 in which both expire worthless. Kabir advertises 9 in 10. These are rough, risk-neutral figures for one week; they say nothing yet about what really happened. Part IV counts that on 104 weeks. 3. Convexity is what the buyer pays for. Write the call payoff as f(S) = (S − K)⁺, a convex function of S. By Jensen's inequality, E[f(S_T)] ≥ f(E[S_T]). More generally, a mean-preserving spread of S_T, more dispersion with the same mean, can only raise E[f(S_T)] (Rothschild and Stiglitz, 1970). So an option's value rises with the dispersion of outcomes, not with any view of direction. The buyer pays for convexity; the seller is paid to hold concavity. Lesson 3 shows this convexity premium as time value, and Part III shows why its price needs no forecast of direction. 4. The zero bound is a support bound, not a risk number. In the lognormal model behind Black-Scholes, S_T > 0 almost surely, so the Rs 14.35 lakh edge sits in a region of zero probability. The economically relevant number for a short put is a stress loss: the loss at a plausible extreme move, sized against the account. Lesson 6 computes it for a three-sigma week, and Part V builds a stress sheet. Taleb's Dynamic Hedging is the classic practitioner's treatment of why a seller's book is judged by its tails, not its average week. 5. Thales again, more precisely. In modern terms Thales bought calls on the rental price of presses at harvest, with the deposits as premiums. But he bought every call there was, so after exercise he also held the market, which is Aristotle's point. Option value and market power are different sources of profit. Exchange-listed options separate them on purpose: standard contracts, many buyers and sellers, and position limits. FURTHER READING - Aristotle, Politics, Book I, chapter 11 (1259a), for Thales. - Peter M. Garber, "Tulipmania", Journal of Political Economy 97(3), 1989; and Famous First Bubbles (2000). - Peter L. Bernstein, Capital Ideas (1992), for the long road from these stories to modern option theory. - John C. Hull, Options, Futures, and Other Derivatives, the chapters on the mechanics of options markets. - Sheldon Natenberg, Option Volatility and Pricing, the opening chapters on contract terms and payoffs. - Nassim Nicholas Taleb, Dynamic Hedging (1997), on the asymmetry of short option positions. - Peter Carr and Liuren Wu, "Variance Risk Premiums", Review of Financial Studies 22(3), 2009. Next: Lesson 2. Drawing the payoff.