Thursday, 16 August 2007

Quants 'Splain Selves: Here's Why We Call Them the "Dumb Money"

More details about the whys behind the quant hedge-fund implosion. The Wall Street Journal has just published several letters to investors from funds explaining their losses - which range up to 57 percent in one case.

The most interesting came from Black Mesa Capital, which lost 7.5 percent in the first week of August. Its fund was levered 3.5 times going into July, when, starting on the 25th, it noticed an unusual pattern of liquidation in the markets.

It looked similar to the selling in Sept. 2006 when $9-billion hedge fund Amaranth needed to raise cash to meet margin calls in its disastrous natural-gas trades, which ultimately lost $6 billion, causing the firm's collapse.

Clearly, the same thing was happening again. The massive selling forced the whole market down. Quantitative strategies based on fundamentals, like Black Mesa's, failed as all stocks - well valued and not - went down at the same time.

When the selling started, it turned into a rout. "This isn't about models, this is about a strategy getting too crowded... and then suffering when too many try to get out the same door," says the letter from AQR Capital Management, which saw one fund lose 21 percent.

"We knew this was a riskfactor but, like most others, in hindsight, we underestimated the magnitude and the speed with which danger could strike."

The letters disclose that, once the "unprecedented" conditions were noticed, many funds started "delevering" - selling positions. ("Unprecedented," huh. What about Amaranth then?) Black Mesa's letter, dated Aug. 8, says the fund started selling on Aug. 6 and that its portfolio would be 50 to 100 percent cash by the time clients read it.

The company goes on to say it'll start buying again when its analyses show the liquidations have stopped for three days. How arbitrary is that! Why three days? Why not two days, or 3.14159265 days?! Some strategy. No wonder a lot of Commitments of Traders analysts call these guys the "dumb money." Now we know why the COTs data tells us to trade opposite to them in most markets.

"The question was, when will it end?" Black Mesa asks. "The answer is, we don't know. It could last two days, two weeks, two months, or two quarters.

Looks like it won't be today, in any case. VIX is above 33, my precious metals holdings are getting shot up, as are my long DJIA, Russell 2000 and TSX positions (bad calls, COTs!!), my DUG UltraShort Oil & Gas and SDS UltraShort S&P 500 ETFs are shining like diamonds (good ones, COTs), the Nikkei is going to hell and the 10-year Treasury note is up (more good calls, COTs), while the 3-Month T-Bill yield is down 17 percent, even after I tapped my computer screen (beautiful, COTs). Just checked the wires. Yup, down 17 percent!

No comments: