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

Physics Hits the Street 
•
129
this, financial modeling is much like mathematical modeling in engi-
neering and science more generally. Models fail. Sometimes we can 
anticipate when they will fail, as Greenbaum and Struve did; in other 
cases, we figure out what went wrong only as we are trying to put the 
pieces back together. this simple fact should urge caution as we de-
velop and implement new modeling techniques, and as we continue 
to apply older ones. Still, if we have learned anything in the last three 
hundred years, it’s that the basic methodological principles of scien-
tific progress are the best ones we’ve got — and it would be foolish to 
abandon them just because they aren’t always perfect.
What’s more, since mathematical modeling in finance is an evolving 
process, we should fully expect that new methods can be developed 
that will begin to solve the problems that have plagued the models 
that have gotten us to where we are today. one part of this process 
has involved modifying the ideas that Black and Scholes introduced to 
financial practice to better accommodate Mandelbrot’s observations 
about extreme events. But that’s only the beginning. the final part of 
the book will show how models have continued to evolve outside of 
mainstream finance, as physicists have imported newer and more so-
phisticated ideas to finance and economics, identifying the problems 
with our current models and figuring out how to improve them. Black 
was instrumental in producing a new status quo on Wall Street, but his 
ideas were just the beginning of the era of financial innovation.

---

W
hen the santa fe trail was first pioneered in 1822, it 
stretched from the westernmost edge of the United States
— Independence, Missouri — through comanche terri-
tory and into the then-Mexican state of nuevo Mexico. from there it 
passed over the high plains of what is now eastern colorado and then 
took the Glorieta Pass through the Sangre de cristo Mountains, the 
southernmost subrange of the rockies. to the southwest was the foot 
of the trail, the Palace of the Governors in the city of Santa fe, the seat 
of Mexican power north of the rio Grande. In front of the palace was 
the city’s central market square, where traders from the United States 
displayed their goods. twenty years after the first American trailblaz-
ers arrived in the city, the U.S. Army followed, battling through the 
Glorieta Pass and claiming the city and all of its surrounding territory 
as part of the newly annexed state of texas.
A century and a half later, two men in their late thirties sat in a 
saloon at the end of the trail, long paved over and replaced by an in-
terstate highway, sipping tequila. they were surrounded by younger 
men, chatting furiously. outside, the park in the bustling market 
The Prediction Company
c H A P t e r 6

---

The Prediction Company 
•
131
square was green from late-summer rains. Across the way, the Palace 
of the Governors sat as it always had, the oldest continuously used 
public building in north America. the square was surrounded by 
low-slung buildings, reddish brown and in the pueblo style, much 
as it was when the American army arrived in 1846. the men in the 
saloon were the newest traders to hang their sign in Santa fe’s his-
toric market district. down the road from the square, in a one-story 
adobe house on Griffin Street, a bank of state-of-the-art computers 
was humming, following the instructions set by the men before they 
left for their evening drink. the year was 1991. the men were in the 
prediction business.
the two graybeards — at least by the standards of the new field 
of nonlinear dynamics and chaos, which they had spent the last fif-
teen years helping to create — were James doyne farmer and nor-
man Packard. Until recently, farmer had been head of the complex 
Systems group at Los Alamos national Laboratory, the government 
lab most famous for having been the headquarters of the Manhattan 
Project. Packard, meanwhile, had just left a tenured position as associ-
ate professor of physics at the University of Illinois’s flagship campus. 
Among the other men at the bar were former graduate students and 
recent Phds, adventurers looking to follow farmer and Packard as 
they blazed a new trail.
the new venture was a company, soon to be called the Prediction 
company (though as they sat that evening on the Santa fe market 
square, the company was still nameless). their goal was to do the im-
possible: to predict the behavior of financial markets. If anyone could 
do it, it was this group. Between them, farmer and Packard had three 
decades of experience in a subject known as nonlinear forecasting, an 
area of physics and applied mathematics (and increasingly other fields 
as well) that sought to identify predictive patterns in apparently ran-
dom phenomena. In Packard’s words, it involved identifying the order 
at “the edge of chaos,” the small windows of time in which there was 
enough structure in a chaotic process to predict where a system would 
go next. the tools they used had been developed to predict things like 
how a turbulent fluid would behave in a narrow pipe. But farmer and 

---

Packard, and the half-dozen acolytes who had followed them to Sante 
fe, believed they could predict far more than that.
As head of the Manhattan Project, J. robert oppenheimer was cer-
tainly the most important member of his family at Los Alamos. But 
he wasn’t the only one. His kid brother, frank, was also a physicist
— and when the elder oppenheimer took over work on the bomb, 
frank pitched in, first at Lawrence Berkeley lab in california, and 
then at oak ridge in tennessee, before finally joining his brother in 
new Mexico. eight years younger than his famous brother, frank ar-
rived at Los Alamos just in time to help coordinate the trinity test, 
the world’s first nuclear detonation, which was staged in the middle 
of the tularosa Basin in new Mexico on July 16, 1945. After the war, 
robert appeared on the covers of Time and Life. He was the public 
spokesman for cold War science in the United States, and for military 
restraint regarding the use of the nuclear technology he had helped 
developed. frank was not quite so prominent, but even so, his mili-
tary research landed him a job in the physics department at the Uni-
versity of Minnesota.
In 1947, J. robert oppenheimer was appointed director of both the 
Institute for Advanced Study in Princeton — possibly the most presti-
gious scientific research institute in the world — and the newly formed 
Atomic energy commission. the same year, the Washington Times-
Herald reported that frank oppenheimer had been a member of the 
American communist Party from 1937 to 1939. eager as he was to con-
tinue in his brother’s footsteps, 1947 was not a good year for a would-be 
nuclear physicist to be outed as a communist. frank initially denied 
the charges and appeared to have escaped with his reputation intact. 
But two years later, amid mass fear about Soviet nuclear research and 
the mishandling of the “atomic secret,” frank was called before the 
infamous House Un-American Activities committee. Under oath and 
before congress, he admitted that he and his wife had been members 
of the party for about three and a half years, pushed to political ex-
tremes during the Great depression.
the confession was a newspaperman’s dream. frank oppenheimer, 
brother of the American scientist-savior, was an admitted commu-
132 
•
t h e p h y s i c s o f wa l l s t r e e t

---