When the Model Breaks
Fat tails, regime shifts, Taleb's critique, and what professional books actually do about it
On October 19 1987 the Dow Jones Industrial Average closed at 1738.74, down 22.61% from the prior session — the largest single-day percentage decline in the index's history.
Under the lognormal-returns assumption that underpins the Black-Scholes pricing model from lessons o1_l1 through o1_l3, a one-day move of that magnitude was an event of probability so small that, taken at face value, it should not have been observed in the entire history of the universe at then-prevailing volatility levels. The model said the move could not happen. The move happened.
In the days that followed, equity option chains around the world repriced. Pre-October 1987, SPX option implied volatilities across strikes were close to flat — an empirical pattern consistent with the constant-volatility, lognormal-returns assumption of Black-Scholes. After October 1987, the equity volatility skew became — and has remained — a permanent feature of the option market.
That is the opening. Finishing a lesson is where it stops being interesting and starts being useful: the full lesson runs to 9 sections and ends with 6 practice questions. A free account is what opens the rest, and the other 255 lessons in the Academy with it. No card.
What this lesson covers
- 1The five Black-Scholes assumptions and their empirical violations
- 2Derman's framing — models as maps, not territory
- 3Model-Failure Stress Tester
- 4Quantifying the model-failure gap — the empirical evidence on tail probabilities
- 5Specific historical regime-shift events and their model-failure profiles
- 6Practitioner responses to BSM failures — what professional desks actually do
- 7The 1987 vol skew — how a single day permanently restructured equity option markets
- 8Where to see this on the platform
- 9Summary — module o1 capstone