4. What you need before touching a derivative

Derivatives 6: When it goes wrong

This closes the block of fundamentals, six of the course's nine levels, before moving on to the three advanced ones. It is the place to take stock without decoration: what you are carrying, what you are not, and what would need checking before opening the first position, if it should be opened at all.

What you are carrying so far

• Futures (level 1): what a standardized forward contract is, what the clearing house exists for, how margin and daily settlement work, and why contango turns a correct idea about a commodity into a slow loss.
• Options (level 2): call and put, premium, intrinsic and time value, and the parity relationship that stops the two instruments trading a life of their own.
• Volatility (level 3): the difference between historical and implied, why implied changes with the strike, what the VIX really measures and why a volatility risk premium exists.
• Sensitivities (level 4): delta, gamma, theta, vega and rho, that is, the breakdown of why an option moves the way it moves.
• Strategies (level 5): protective put, covered call, spreads, and the question almost nobody asks, which is when it is better not to hedge at all.
• And this level: the seller's asymmetry, the mechanism of margin and forced liquidation, and four real cases that share a skeleton.

Note what does not appear on the list: any strategy that makes money. That is not a failing of the course. A derivative is not a source of return, it is an instrument for shifting a risk from one balance sheet to another, and being paid to accept it is a legitimate business only for someone who can survive the day the bill falls due.

The four checks to run first

Before opening a derivative, and however convincing the idea sounds, you need four answers in writing:

• The worst-case loss profile. Not the likely loss, nor the expected one: the number you arrive at if the underlying does the worst it can do. If that number does not exist, as with a naked call, you already have your answer.
• How much can genuinely be lost. Buying a put caps you at the premium. Selling that same put does not. And the possibility of ending up with a debit balance, that is, of losing more than you deposited, is not a theoretical case: it happened in April 2020 with oil and in November 2018 with gas.
• Capital to withstand topping up margin. Not today's initial margin, but the margin that would be demanded with the underlying against you and volatility multiplied. If that figure does not fit comfortably, the position is too big, however right the analysis may be.
• What would have to happen for the idea to be wrong. Written before opening, not improvised afterwards, once money already lost is influencing your judgment.

Why most investors do not need derivatives

It is worth saying plainly. A private investor with a diversified portfolio and a long horizon can meet their goals without touching a single derivative in their whole life, and statistically will do better that way. Derivatives add value when there is a specific risk you want to shift: a farmer locking in the price of a harvest, a company securing an exchange rate, a manager capping an exposure it cannot afford.

Outside those cases, what is usually being sought is leverage dressed up as sophistication. Level 5 already flagged it with the covered call: it looks like regular income and it is the sale of the good stretch of your own portfolio in exchange for a modest premium. It is not that it is forbidden; it is that you have to know what you are selling. If the answer to "why do I need this derivative" is "because I want to make more", the right instrument is not a derivative, it is a review of the size and composition of the portfolio.

What comes next

Up to now, volatility has been a parameter that goes into the price of an option and positions have been made of a single piece. Over the three remaining levels that stops being the case.

• Level 7, volatility as an asset: an underlying does not have an implied, it has a whole curve with a figure for every expiry; how to back out the volatility the market is pricing between two future dates, which is the only figure that tells you anything ahead of an earnings report; what it means to buy and sell volatility with a thesis of your own; and why the VIX cannot be bought and the products claiming to replicate it bleed out month after month.
• Level 8, combined positions: straddles and strangles, calendars, ratios and butterflies. Why a long straddle can lose money even though the market moves in the direction you expected, at exactly what moment a capped position turns into one with an open tail nobody has measured, and which specific executioner belongs to each structure when the bad day arrives.
• Level 9, derivatives in a real portfolio: the statement of an actual account. The size of an options position is not the premium paid but the notional it controls; how the margin demanded, the spread and the counterparty all three get worse on the same day and for the same reason; the full bill for a strategy that rolls every month; and the question the whole course prepares you for, which is when a derivative is the right tool and when it is leverage told a different way.

There is also a gap that none of the nine levels fills: knowing whether the track record of an options strategy says anything or is selection. That is the territory of math and statistics, the intermediate track where the warnings of this level stop being prose: fat tails explain why the Gaussian bell always underestimates episodes such as February 5, 2018, multiple testing puts a number on the suspicion that a twenty-four-month run proves nothing, and out-of-sample validation turns into a procedure the idea that a short record does not validate a selling strategy. The other intermediate track, Risk and portfolio management, answers the question of size, which here has been taken for granted.

Below sit the three easy code-free tracks — Market theory, Macro and Costs — and above them the three hard ones: Technical and Python, Data and production, and Expectations and edge, which contains not a single line of code and is hard because of what it demands, not because of programming. The full map, with a recommended order, is at the end of level 9.