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

Epilogue: Send Physics, Math, and Money! 
•
209
using methodological insights that are commonplace in physics (and 
engineering) and that are useful in studying virtually anything. the 
stories in this book show the methodology in action: one uses simpli-
fying assumptions to make a problem tractable and solve it. then, once 
you see how your solution works, you can double back and begin ask-
ing what happens when you play with your assumptions. Sometimes 
you realize that your original solution is no good, because it depends 
too heavily on assumptions that never really apply; other times, you 
discover that the solution is pretty good but can be improved in simple 
ways; and other times still, you realize that your solution is great under 
certain circumstances, but you need to think about what to do when 
those circumstances don’t apply.
obviously, physicists aren’t the only people who have thought about 
understanding the world in this way. this kind of model building is 
ubiquitous in economics and in other sciences. Unsurprisingly, most 
advances in economics have been made by economists. But physicists 
are very good — perhaps especially good — at thinking like this. And 
they are usually trained in a way that helps them solve certain kinds 
of problems in economics, without the political or intellectual baggage 
that sometimes hampers economists. Plus, physicists have often come 
to these problems with different knowledge and backgrounds from 
people who are trained as economists, which has meant that in some 
cases, physicists have been able to look at problems in a fresh way.
However, when I say that science is a process, and particularly that 
financial modeling should be understood as an example of that pro-
cess, I do not mean to say that financial modelers are somehow march-
ing along the path of scientific progress, inexorably approaching some 
“final theory” of finance. the goal isn’t to find the final theory that will 
give the right answer in every market setting. It’s much more modest. 
You’re trying to find some equations that give you the right answer 
some of the time, and to understand when they can be relied on.
derman and Wilmott, in their Manifesto, make this point quite 
clearly. We should never mistake a good model for the “truth” about 
financial markets. the most important reason for this is that markets 
are themselves evolving, in response to changing economic realities, 
new regulations, and, perhaps most importantly, innovation. for in-

---

210 
•
t h e p h y s i c s o f wa l l s t r e e t
stance, the Black-Scholes model forever changed how options markets 
operate — which meant that the markets the model was designed to 
describe were revolutionized by the increasing use of the model. this 
led to a feedback loop that wasn’t fully recognized until after the 1987 
crash. As sociologist donald MacKenzie has observed, financial mod-
els are as much the engine behind markets as they are a camera capable 
of describing them. this means that the markets financial models are 
trying to capture are a moving target.
far from undermining the usefulness of models in understanding 
markets, the fact that markets are constantly evolving only makes the 
iterative process I have emphasized more important. Suppose that Sor-
nette’s model of market crashes is perfect for current markets. even 
then, we have to remain ever vigilant. What would happen if investors 
around the world started using his methods to predict crashes? Would 
this prevent crashes from occurring? or would it simply make them 
bigger, or harder to predict? I don’t think anyone knows the answer to 
this question, which means that it is just the kind of thing we should 
be studying. the biggest danger facing mathematical modelers is the 
belief that today’s models are the last word on markets.
Weinstein and Malaney’s proposal is different from the other ideas dis-
cussed in this book. every other chapter concerns, in one way or an-
other, finance and financial modeling. the other physicists I discussed 
were looking at a bunch of statistics — stock prices, market moves, an-
nual returns — and trying to make predictions about how the num-
bers would change in the future. the details of how markets work is 
of course relevant to such predictions, but it is not so hard to see how, 
as osborne observed, a person trained as a physicist is well suited to 
interpret statistical data. Weinstein and Malaney, however, have pro-
posed a new theory of welfare economics, inspired by ideas developed 
in physics. this is a far more ambitious project, and one that is more 
difficult to wrap one’s head around.
nonetheless, if one understands the connection between physics 
and finance in the right way, there is nothing weird about using phys-
ics as a way of making progress in economics more broadly. It isn’t that 
financial markets bear some special connection to the subject matter 

---