He proceeded to place an example on the blackboard. "Let's find the correlation between intelligence as measured by an IQ score and reading proficiency as measured by a standardized test. Let's use 50 students for our sample," He pumped the data into the correlation formula and came out with a single number with a high plus value.

"We want to find out if this number is significant or not." He used another simple formula and found that, indeed, the correlation was significant at the 0.05 level. The 0.05 level means that there are only 5 times out of 100 that the correlation was due to chance.

Now, this is the fatal flaw in all the mathematics used on Wall Street:

With such high correlation and so little due to chance, you'd think you have a good rationale at the beginning of the year to determine the outcome at the end of the year. If you knew all 50 students' IQ level on the first day of school, you'd think you could place a small wager the class would do really well on the standardized reading test at the end of the year. You'd think, but you'd be wrong, and you might just lose your wager.

Here is why. Of these 50 students, we find at the end of the year that two students were in an auto accident -- one is severely brain injured and one is in a coma. One student was diagnosed with ADHD and another with autism. One was blinded while playing with firecrackers, and finally, three of the best students moved out of state.

Now we begin to see the fatal flaw in all the mathematical formulas used on Wall Street. The variables are always shifting, changing, moving up, down, sideways. For example, there is an explosion at an oil refinery, a war breaks out between Israel and Palestine, there are new bank failures in Europe, the Dow makes a new low and on and on we go.

I want to take two specific cases where the mathematics used was essentially correct but did not consider the shifting of the underlying variables. One such formula is Long Term Capital Management (LTCM). In 1993, John Meriwether left his post as head bond trader at Salomon Brothers and founded LTCM. He recruited two brilliant mathematicians, Myron Scholes and Robert C. Merton, both of whom received the Nobel Prize for their Black Scholes model for pricing options. Meriwhether started a hedge fund because he did not want too much financial regulation (sound familiar?).

The firm grew rapidly and secured hundreds of millions of dollars. Now flush with cash, it started making bigger and bigger bets, taking on more and more risk. Trading progressed smoothly until 1998, when suddenly a Black Swan event happened and Russia defaulted on its bonds. Panicked investors sold Japanese bonds and bought U.S. Treasury bonds. LTCM was on the wrong side of these trades and lost $1.85 billion in capital.

(The term Black Swan comes from a book by Nassim Nicholas Taleb entitled The Black Swan. Black Swan refers to a once in a lifetime experience. We see white swans for our entire life and then one day, out of nowhere, we see a Black Swan.)

As losses mounted, investors pulled their money out of LTCM and LTCM was forced to liquidate positions at huge losses. The potential for untold losses loomed until the New York Federal Reserve stepped in and arranged a consortium of several large banks that agreed to supply capital to keep the firm solvent in exchange for 90% of the shares of LTCM. When the dust settled, LTCM had lost $4.6 billion and LTCM was no more.

The variables in the mathematical formulas they used shifted right out from under them in a matter of days, which brings home the point that fixed formulas do not always work in human situations. At the time, some people felt that bailing out LTCM was a mistake because Wall Street would take greater and greater risks in the future and let the government bail them out of any losses. How much more prophetic could you be?

Now we fast forward to the year 2000. Enter David X Li from rural China, after he earned advanced degrees in mathematics. He took a job at JP Morgan Chase, and instead of working with only two variables, Li worked with groups of variables called Gaussian copulas. These copulas produced a single number. Li was working on CDSs and CDOs, and his one number was used to trade these securities on Wall Street.

Wall Street had found Nirvana -- a single number on which to bet billions of dollars (CDOs are bundles of mortgages traded as a single unit). Li made his calculations on past values of CDSs (a CDS is made up of an underlying mortgage bond and the counter party insuring company). Using past prices is like trading from a chart. It tells you what happened in the past, but will not tell what will happen next.

We all know the rest. Wall Street went merrily on its way, onward and upward. The housing boom was underway, and Li's formula was tied to ever rising prices. Trading managers didn't have a clue what Li's formula was all about and never took the time to try and understand it. They just went ahead and bet billions of dollars on his formula.

Then came the Black Swan. Foreclosures rose, house prices fell and like LTCM, the house of cards fell apart. Now they say Li went back to China and Wall Street may or may not recover. Estimates run as high as $7 trillion in losses that have yet to be unwound. Good luck America.

And a final note. JP Morgan Chase is now the dominant player in the CDS market with $87.7 trillion of OTC (over the counter) contracts outstanding. Perhaps we could guess that Mr. Li's formula helped in this regard.