On March 29, 1900, Louis Bachelier successfully defended his doctoral thesis at the University of Paris. It was breathtakingly innovative and complex, so much so that it would take decades for his ideas to be fully understood let alone adopted. This was the case even though his thesis advisor was none other than famed physicist Henri Poincairé and that he solved the mathematics of Brownian motion five years before Einstein. Well, he didn’t quite “solve” the problem, he was the first to model the process stochastically.

It was his research subject rather than the object for deducing it that was of great interest. Bachelier was intent upon using mathematics and really statistics to decipher financial speculation. He established what has become the dominant theory as finance is further and further quantified. He wrote, “there is no useful information contained in historical price movements of securities.” It was the birth of Wall Street’s “random walk.”

Bachelier’s theory has been behind almost all financial evolution especially during the eurodollar era where math has taken on not just central functions but even the proportions of money itself. The housing bubble was the housing bubble because Wall Street had since the late 1980’s been busy hiring every Ivy League mathematician they could find (including David X. Li and his Gaussian copula) to develop and manage complex “products” that their sales distribution networks could sell. The Finance majors were all in the front and on the phones finding new customers to buy essentially increasingly complex formulas that did nothing more than try to solve why Bachelier’s randomness always seemed to fall (catastrophically) short.

Randomness is really a cop out; it’s a short cut to use math where math doesn’t belong, or at least not to so much blind deference. We live in an incredibly complex world and operate incalculably complex systems which current technology and theory cannot actually model. There is far too much texture and granularity in individual systems than can fit in all the world’s computing power. Randomness is like an MP3 compression – that what is left out of the model isn’t really important because of subjective opinions about what is most important (independent variables).