Citadel Investors Shouldn't be Surprised It Struck Out

If the giants of U.S finance are too big to fail, then many of its investors appear too dumb to succeed.

The most striking thing in the WSJ piece this morning about hedge fund manager Ken Griffin's struggles raising money for his Citadel Investment Group is the impression it gives of investors peeved at the firm's audacity to lose money during the financial crisis (subscription required):

Most of all, Citadel wants to leave 2008 in the past.

"But I don't think anybody else has," Jon Gans, president of Ironwood Capital Management, told other Citadel investors this past summer in the firm's downtown San Francisco offices, according to people familiar with the matter. Ironwood and other longtime Citadel investors say they intend to reduce investments with the firm, the people say.

This isn't to defend Citadel's -- or any hedge fund's -- calamitous loss, nor to celebrate its gains. Rather, it's to observe a fairly mundane truth about these investment pools: Sometimes you win, sometimes you lose. An obvious corollary is that doubling down can speed either of these outcomes.

Citadel's algorithmic trading unit, which basically uses big computers and nifty equations to make rapid-fire investments, in 2004 had returns of roughly $3 million. The following year it made $75 million. In 2007, that ballooned to $892 million, while last year the winnings rose even higher to $1 billion.

What does this suggest -- that Citadel is so smart, that its equations arrange Greek symbols more artfully than those of other investors? If so, it's hard to fathom why the firm fared so much worse in 2008 than other hedge funds, which on average fell 19 percent.

No, such fast gains and high volatility indicate a firm that fell in love with swinging for the fences. When it connects, the ball sails far into the night sky, drawing appreciative oohs and aahs from investors. When it whiffs, like all sluggers do, the fans wonder where the magic has gone.

Of course, there is no magic (although Warren Buffet comes pretty close to mastering the conjurer's arts). What does exist is risk. Griffin is now said to be dialing back on that.

Citadel has cut his wagers on corporate bonds and other holdings that hurt him badly last year, he says, and has been putting more cash into plain-vanilla stocks. In a letter to investors late last month, he told them not to expect huge bets like the E*Trade investment, which saddled the funds with a "significant concentration of risk."

Now they're having second thoughts. Fair enough. Commendable, even. But if you don't like the rules of the game, don't play.