Some market timing services base their signals solely on nonhuman factors. Timing Cube in Austin, Texas, generates between three to five trading signals a year from market-driven computer models aimed at removing emotion from the equation. "People always tell us that no one can predict the market," says Timing Cube cofounder Serge Dacic. "We're just listening to what the market is telling us, and we follow the trend and invest with it."

Believing that a traditional buy-and-hold approach increases risk by inducing investors to gut it out during significant market downturns, Timing Cube follows prevailing long-term trends mainly based on price and volume action on the Nasdaq Composite index. Depending on the trend, Timing Cube suggests investors should either jump completely into the market, jump ship entirely or hedge their portfolios by taking long and short positions at the same time. Their vehicles of choice are exchange-traded funds and mutual funds that track the Nasdaq 100, Russell 2000 and S&P 500 indexes, which Dacic says provides enough variable performance to create diversified trading opportunities.

Since it began in June 2001, Timing Cube has generated 10 trades, with the average signal lasting 105 days. The shortest signal was just three days. The longest signal was a 393-day buy sign that ended this past April, when Timing Cube issued a sell call in anticipation of a correction based on several factors- indications of institutional selling after prices fell on higher volume following strong earnings reports, interest rate concerns and geopolitical fears.

During its three-year existence, Timing Cube has generated profits in all four of its trading strategies, which entail a mix of long-only and long-and-short positions. Returns range from a 43.73% cumulative gain on a long-only position on the S&P 500 (versus a 6.12% loss on a buy-and-hold position on that index) to a 712.81% cumulative return on its long-and-short-with-margin position on the Nasdaq 100 (versus a 11.41% loss with buy-and-hold). To date, Timing Cube's signals have been profitable 73% of the time across its various trading strategies.

Timing Cube's infrequent signals and overall track record appeal to Dave Garrett, who licenses the service for his $20 million money management firm in Salt Lake City. In addition to Garrett Capital Management, Garrett also oversees TimerTrac.com, a Web site that tracks roughly 400 timing signals so subscribers (financial advisors are a target market) can pick and choose services for their clients' portfolios.

Garrett, an unabashed timer who manages accounts for ten advisors, licenses as many as eight timing services at a time to help manage his clients' money. If one of his partner advisors likes a particular timing signal, Garrett will license it to use for his firm. He's more concerned with avoiding market tumbles than handily beating the market indexes, although his roughly 11% average annualized returns since 1999 beat the averages during that time.

Garrett admits that market timing is a tough sell to advisors because of its very nature: it works best in down markets, but markets historically are up about two-thirds of the time. He explains that timing tends to perform better in bear markets because investors run to cash, the market sinks for a prolonged period, and timers can be heroes by avoiding huge downturns. The trick is getting back in before prices get too high and exceed the initial selling prices.

Conversely, downturns during bull markets are generally short-lived, creating situations where investors sell and then get back in at higher prices than where they cashed out. Considering that bullish markets typically last longer, that presents more opportunities for timers to get head faked into wrong positions.

It doesn't take much to set back a portfolio's long-term performance with a wrong call. According to Ibbotson Associates in Chicago, the hypothetical value of $1 invested in the S&P 500 from year-end 1983 through 2003 earned $11.50 (versus $2.71 for the S&P 500 minus the best 17 months during that timeframe). That same dollar invested in the S&P 500 in 1925 would've gained $2,285 (versus $17.42 minus the best 37 months).

That's a powerful argument for anti-timers, though pro-timers tend to brush off such figures. "Timing is an emotional topic," says Garrett. "It evokes almost religious-like responses on both sides of the issue."