Bad Timing

Over a one-week period in mid-July, the S&P 500 advanced by more than 1% on a single day -- three times. I learned about this via a post by the folks at Bespoke Investment Group, at seekingalpha.com. They mused that while it may seem like an unusual occurrence, it's really not much of one. During the past year's market slide, it has happened at least several times. Their cautionary conclusion? Don't assume that this means the market is turning around. It has done this before, only to resume sliding.

Right.

This reminded me of various studies of market timing, which point out how badly your returns can suffer if you're out of the market on a handful of key days. The week in July contained some such days. Even individual stocks make significant advances on single days. Check out the following one-day gains, for example:

Company	% move
Sunoco (NYSE: SUN  )	6%
BioCryst Pharmaceuticals (Nasdaq: BCRX  )	20%
Kraft Foods (NYSE: KFT  )	5%
Amgen (Nasdaq: AMGN  )	12%
TurboChef (Nasdaq: OVEN  )	15%
Gencor Industries (Nasdaq: GENC  )	12%
OpenWave Systems (Nasdaq: OPWV  )	10%

They were made on July 28, a day when the Dow dropped 200-plus points. If you were trying to time the market with one of those stocks, you'd have missed out on its gain. That day, BioCryst officials announced that a flu drug had shown good results in a trial, while it was revealed that Philip Falcone's Harbinger Capital hedge fund had acquired a 6.6% share of Sunoco. These are not the kinds of announcements you can anticipate -- at least, you aren't likely to be able to pinpoint them.

Some might argue that other kinds of news can be anticipated, such as earnings reports. Well, yes, you can definitely know when a company will be reporting its latest earnings results -- but you still aren't likely to know whether they'll inspire or disappoint the investing world, and whether the stock will be driven up or down.

What to do
So what should you do with this information? Well, some investors will try to figure out what to do by looking for patterns in past market performance. As someone asked about the Bespoke article: "Any idea what the weekly, monthly performance after a '3 1% up days in 5' is?" This worries me because even if patterns and trends emerge, they'll never be 100% reliable.

That's why, despite all this market uncertainty, what most investors should do is simply remain invested in the market, through downturns and upturns.