First he was very famous, now he's very rich. But Nassim Taleb is still wrong.

While for just about everyone involved in the markets the last two years of financial history have been a massacre, they have been a long victory lap for Nassim Nicholas Taleb. Taleb is the author of The Black Swan, the book about, as the subtitle puts it, "The Impact of the Highly Improbable." It came out in 2007, just before everything that seemed highly improbable became painfully actual. As everyone else's fortunes have shrunk, Taleb's have risen. Not only have his books made him the public face of the New Catastrophism, but his insights have turned out to be extremely profitable: The Wall Street Journal reports that Universa, a hedge fund for which Taleb serves as guru and adviser, gained more than 100 percent last year and now holds $6 billion.

It's hard to argue with success. In bubble markets and bear markets, the talk always turns to new paradigms. If there was a huge crash yesterday, why shouldn't there be an even bigger one tomorrow? For a while now, though, I've been trying to explain to people why I am loath to jump on the Taleb bandwagon. The news about Universa doesn't change this. I am not surprised that Taleb's approach has made money for investors. But I will be if—assuming he doesn't change his approach—he keeps doing so.

Taleb has become the go-to philosopher of the markets with a straightforward and appealing precept: that people always underestimate the chances of improbable, out-of-the-ordinary events. This to me seems a dangerous proposition about the markets. It is also, I think, a questionable proposition about human behavior.

While Taleb has acquired a huge following in the world of business and investment, he does not present himself mainly as a "business" thinker. Little of his 2004 book, Fooled by Randomness, and even less of The Black Swan talks about investing directly. His conceit is that he helps readers see possibility. It is attractive because it separates people into the plodders—or, as Taleb calls them, nerds—and the street-smart Talebites who've learned to appreciate the unexpected. But the closer you look, the less clear it is that the plodders are as consistently wrong as Taleb thinks.

One thing to be said in Taleb's favor is that he has never lost a spectacular amount of money at once. Three times now he has made money when few others did. The first was in the 1987 Black Monday stock market crash, when he made $35 million to $40 million as a trader. The second was the tech stock crash of 2000, when Taleb's own fund, Empirica, gained 60 percent. And the third is Universa.

What you might miss in this, though, is what happened to Taleb in the in-between years. Taleb, in the pre-Universa days, said that his Black Monday windfall made up 97 percent of the money he'd made. Afterward, he moved through several trading jobs without much success. In 1999, he started Empirica, which, as the Wall Street Journal reported, followed the gains of 2000 with several lackluster years and closed in 2004. In Taleb's telling, this is part of the magic. "When you lose money steadily and then make money in lumps," Taleb told Bloomberg magazine, "people think you're crazy."

Well, maybe. Or maybe they just think that you're steadily losing money. Taleb's basic trading insight was that he could buy "deep out of the money" options that would pay off in a dramatic market fall. Most of the time, such options expire worthless. But in a market crash, they deliver a big payoff. In effect, Taleb's strategy has been to buy insurance, reasoning that folks underestimated the likelihood of catastrophic events. It's a lot like buying $200,000 of insurance on a $150,000 house. If it burns down, you've made money. Taleb's idea was that this insurance was underpriced.

The problem with catastrophism, however, is that it's very difficult for anyone in the market to wait around for the unexpected. And while Taleb buys insurance, most other market players prefer to sell it. In the New Yorker profile that first brought Taleb to prominence, Malcolm Gladwell cannily contrasts Taleb with Victor Niederhoffer, a hedge-fund manager whose stock in trade was selling the same kinds of options that Taleb bought. At the end of Gladwell's story, Niederhoffer loses a lot of money. It's a pattern that Niederhoffer repeats over and over—a later story about him christens him "The Blow-Up Artist."

By this point, what's wrong with the Niederhoffer approach has become abundantly clear. Offering insurance without knowing the odds is a strategy that is prone to blowing up in spectacular ways. It doesn't matter whether the instruments involved are options or credit-default swaps (which are literally insurance against defaults on corporate bonds) or any of the other tools of finance. In the worst case, you wind up with Niederhoffer—or AIG (AIG).

But the failures of the Niederhoffers and AIGs do not translate to a validation of Taleb-style catastrophism because these two approaches turn out to be linked. They are mirror images. In noncatastrophic times, the Niederhoffers and AIGs make money consistently and quietly and then end up losing it conspicuously and painfully. The Talebs make money rarely, amaze everyone because they do it when everybody else is getting killed—and so make it easy to forget about years of steady losses. Over the long run, the anti-catastrophists often do fairly well (if they don't get too greedy and make bets that cost them all their money in even a small market drop). But it is the catastrophists, a la Taleb, who look smarter. If you're always planning for crisis, you look like a genius when it does come.

Arguing against Taleb is a little embarrassing; who among us wants to side with the plodders when for the price of a paperback you can join the elect? But the experience of the markets here is important because it shows that neither consistently discounting the chance of unforeseen risks, as AIG did with such gusto, nor betting day after day on unforeseen catastrophes is a reliable way to make money.

In his books Taleb presents a wealth of examples of how prone we are to discount the unexpected and unlikely, but what is notably missing from The Black Swan are examples of just how likely we are to overestimate the chances of unlikely events when they are presented to us under a spotlight. Taleb is, of course, right that we fail to anticipate what we are not looking for. But we also overanticipate when we are looking too hard for the outliers. Lottery players overvalue their chances of winning $10 million, and horse bettors put too much money on 100-1 long shots. People who watch the local news too avidly believe there is a child kidnapper around every corner, and followers of Taleb assume that every time they pass a dark alley, catastrophe is about to pop out with a bloody knife.

One issue in investing that Taleb spends some time on in his book Fooled By Randomness is "survivorship bias." It's well known that fund managers as a group look much better than they should because the least successful ones go out of business. As Taleb points out, we notice the ones who do well a lot more because those who do badly simply drop out of sight. So it is also with predictions of disaster: We mostly get to hear about the few times that they turn out to be right.