The Physics of Wall Street: a brief History of Predicting the Unpredictable

part and parcel of what I have described as thinking like a physicist —
it amounts to resisting complacency in model building. And indeed, 
trying to figure out how to predict the kinds of events that might have 
seemed like black swans from the perspective of (say) osborne’s ran-
dom walk model is precisely what led Sornette to start thinking about 
dragon kings. Surely not every black swan is really a dragon king in 
disguise. But that shouldn’t stop us from figuring out how to predict 
and understand as many kinds of would-be black swans as possible.
taleb, though, wants to go further than this. He believes that black 
swans show that mathematical modeling, in finance and elsewhere, is 
fundamentally unreliable. figuring out how to predict dragon kings, 
or using fat-tailed distributions to address the fact that extreme events 
occur more often than normal distributions indicate, isn’t enough. It 
seems to me that one can argue successfully that any particular model 
is flawed — albeit usually in ways that a responsible model builder 
would recognize from the start. But taking this to the next level and 
arguing that the model-building enterprise as a whole is doomed is a 
different matter.
Just consider: the process of building and revising models that I 
have described here is the basic methodology underlying all of science 
and engineering. It’s the best basic tool we have for understanding the 
world. We use mathematical models cut from the same cloth to build 
bridges and to design airplane engines, to plan the electric grid and 
to launch spacecraft. What does it mean to say that the methodology 
behind these models is flawed — that since it cannot be used to predict 
everything that could ever happen, it should be abandoned altogether? 
If taleb is right about mathematical models, then you should never 
drive over the George Washington Bridge or the Hoover dam. After 
all, at any moment an unprecedented earthquake could occur that the 
bridge builders’ models didn’t account for, and the bridge could col-
lapse under the weight of the cars. You should never build a skyscraper 
because it might be hit by a meteor. don’t fly in an airplane, lest a black 
swan collide with one of its engines.

---

216 
•
t h e p h y s i c s o f wa l l s t r e e t
taleb would have it that finance is a different kettle of fish from 
civil engineering or rocket science, that extreme events are more un-
predictable or more dangerous there. But it’s hard to see why. cata-
strophic events, when they occur, usually come without warning. this 
is true in all walks of life. And yet, it doesn’t follow that we shouldn’t 
do our very best to understand what risks we can, to domesticate as 
many unknown unknowns as possible. It’s important to distinguish 
between the impossible and the merely very difficult. there’s little 
doubt that mastering financial risk is extremely difficult — much more 
difficult, as Sornette would say, than solving problems in physics. But 
the process that I have described in this book is the best way we have 
ever come up with for addressing our biggest challenges. We shouldn’t 
abandon it here.
there’s a third criticism of financial modeling that one sometimes 
hears. this one is a little deeper. It has been made most influentially by 
Warren Buffett, who has famously warned of “geeks bearing formulas.” 
this view has it that financial innovation is a dangerous thing because 
it makes financial markets inherently riskier. the excesses of the 2000s 
that led to the recent crash were enabled by physicists and mathemati-
cians who didn’t understand the real-world consequences of what they 
were doing, and by profit-hungry banks that let these quants run wild.
there is much that is right in this criticism. the idea that deriva-
tives, including options, are a manufactured “financial product” has 
proved extremely powerful — and profitable. over the past forty years, 
financial engineers have come up with ever more creative, and often 
convoluted, derivatives, engineered to make money in a wide variety 
of different circumstances. dynamic hedging — the idea behind the 
Black-Scholes model — is the basic tool used in this new kind of bank-
ing, since it allows banks to sell such products with apparent impunity. 
As the banking sector has evolved to put more and more emphasis 
on new financial products, the impact of a failure of the mathemati-
cal models undergirding these products has become ever larger. And 
indeed, some of these creative new financial products were at the heart 
of the 2008 crisis. So it is certainly true that physicists and mathemati-

---