Back when I was in college, a friend and I were having a discussion over a case of beers about how you could defraud lots of people. We weren't really planning to try and do it, but we were just tossing around the ideas of Ponzi schemes, multi-level marketing schemes like Amway, and basically any other white-collar equivalent of the pigeon drop.

My friend old me about an age-old con: you acquire a very large spam list of email addresses and start sending them periodic emails. You pick an event in the stock market that in which two things can occur; in his example, it was picking whether a particular stock will go up in a given month. Let's say its Apple (NASDAQ: AAPL). You split the list of emails in half the first month. To the first half, you tell them that Apple's stock will increase in valuation per share. To the second half, you tell them the opposite: that it will go down in value.

At the end of that first month, you have (in theory) half of that spam list believing that you predicted the stock's movement in valuation accurately. You scrap the half to whom you sent the incorrect prediction, and repeat the process with the other half in the following month.

If you start with 40,000 emails, and do this for six months, you end up with 625 people who've received emails from that made accurate "predictions" about the movement of Apple's stock valuation in the past six months. At that point, you try and convince the 625 of them that you're an experienced money manager that is able to make accurate predictions (instead of random guesses) about stocks and try to get investment money from them.

The logic behind this con struck a chord with me. It was a terrible idea to try and execute for all kinds of reasons, (again, the intent was never to actually do it. We were just bored and partly drunk college kids) but it turned on some lights in my head. I started to think about the guys on Wall Street who are responsible for analyzing stocks and making similar such "predictions" with other people's money.

I'm not saying that all financial investors are dishonest schemers, or even gamblers. But what I started to think about was the pool of investors as a whole. They are, doubtless, a large group of diverse people with a common goal: to make lots of money in a short period of time. Let's say you start with 40,000 of them, and turn them loose to invest in the stock market with lots of money. Imagine they have little else to go on besides guessing, at random, at what time and in which stocks they should invest in to yield a positive return.

If you do that, and the pool of investors is large enough, then some portion of them are bound to guess correctly. They'll make money. But the problem comes with two things. The first is visibilitiy. Most likely, we'll only hear about the winning investors while the rest of them are quietly dismissed (this is called survivorship bias.)

The second problem is attribution: if you perceive the success of those few as being a function of their intelligence as investors, instead of good luck, then as a manager responsible for those investors, you might start making bad decisions. You fire those that lost money, for example, because you interpret their bad luck as incompetence. But more importantly, you might have acquired a false sense of trust about the investors who guessed correctly...and give them more money to manage.

In my younger days, I wrote off this thought as being something that was possible or even likely, but difficult for me to verify. I wasn't (and still am not) trained as an investor. I assumed that people who managed investment funds had studied economics and financial markets and generally had a pretty good grasp on what they were doing. (Warren Buffet can't just be the luckiest guy on the planet.)

In the book Fooled by Randomness, Nassim Taleb explores this idea in great detail. As an investor himself, and someone who is very familiar with probability and statistics, Taleb expresses his concern for the fact that a lot of otherwise bright people seem to make this error in attribution. The lucky investors, who made good decisions or avoided making catastrophically bad decisions, come out on top, and we hear about them and their (apparent) brilliance in predicting how the stock market will work.

Using the texts of Nostrodomous to pick stock tips is also a bad idea. The direction stocks take doesn't follow the patterns set by astrology. Numerology won't crack tomorrow's valuation of Apple. But as humans, we're pretty good at finding patterns in things, especially when we want to. This would a tough sell to a Wall Street speculator. As Upton Sinclair once said: "It is difficult to get a man to understand something, when his salary depends upon his not understanding it!"

No wonder we're in such a mess.