Harry Domash's Winning Investing


Ignore the Experts -- Figure It Out Yourself

Can anyone predict the market? Apparently not!

In my last column, I described data that showed that stocks rated “buy” by analysts were just as likely to go down as up.

Dismayed by the analysts’ track record for individual stocks, I wondered about the value of the overall stock market predictions from so-called ‘experts’ that we see on TV and read on the Internet. It turns out that they are about as good as the stock analysts—they have a 50/50 chance of being right.

I learned that from CXO Advisory Group (www.cxoadvisory.com), which tracks the market forecasts given by 48 market commentators, and posts the results in its Guru Grades report. CXO only tracks gurus whose prognostications are available free on the Web (the report actually listed 49 gurus, but one, David Nassar, has been inactive since May 2006). The report lists the total number of market forecasts, as well as the number of forecasts that were essentially right or essentially wrong, for each guru.

Like Flipping a Coin
According to CXO, only two of the 48 gurus managed to get it right least two-thirds of the time. They were money managers Ken Fisher, who writes monthly columns for Forbes Magazine (www.forbes.com) and James Oberweis, whose musings can be found on Zacks site (www.zacks.com). Regular readers might recall that Fisher also led the list when I last mentioned the Guru Report in a July 2006 column.

But, even the numbers for Fisher and Oberweis paint an overoptimistic picture. For instance, for each of the last three years, Fisher forecast much stronger markets than actually occurred. To be fair, for 2007, he accurately predicted that foreign stocks would have a much better year than U.S. stocks. Oberweis correctly forecast 2005’s lackluster results, but his calls for 2006, when the market gained 14%, were ambiguous, and he expected “above average” returns last year, when the market struggled and only managed a sub-par 4% return.

Only four other gurus tracked by CXO (David Dreman, Jack Schannep, Louis Navellier, and Jason Kelly) got it right at least 60% of the time. Some of the other well-known gurus tracked by CXO include Bob Brinker (59% batting average), Tobin Smith (49%), and James Cramer (48%). Select Guru Grades on CXO’s homepage to see the entire list.

But, Guru Grades aside, CXO provides considerable information that could help you make better investing decisions on your own. These take the form of reports summarizing research conducted by CXO itself, as well as summaries of academic studies.

Here are some that I found interesting. You can access the Trading Calendar mentioned next from CXO’s home page. Otherwise, you can get to everything that I describe from the “Blog-Investing Notes” section.

Best/Worst Months
CXO’s Trading Calendar displays the average S&P 500 index returns by month since 1990. According the CXO, the best months to be in the market are December (up 2.4%), May (up 2.0%), October (up 1.9%) and November (up 1.8%). The worst months are July (down 0.6%) and August (down 0.2%).

Best/Worst Days
Zeroing in on individual days, since 1990, the best single day to own a stock was October 28, and the worst was August 5.

Turn of the Month
According to CXO, the best time to own stocks is during the last six trading days of one month through the first five trading days of the next month. By owning the S&P 500 index on those days only, from December 21, 1989 through August 7, 2006, CXO says you could have turned $100,000 into $375,000. In a later study, CXO found that you would have ended up with slightly more than $400,000 instead of $375,000 if you had refrained from executing the strategy at the end of July, August, and September of each year.

Time of Day
CXO quotes a recent research study that found that stocks register most of their gains overnight, that is from one day’s closing price to the next day’s open. Thus, the best time to buy is just before the market closes, and the best time to sell is at the open. These conclusions, based on data from 1993 through 1996, hold for individual stocks, stock indexes, and exchange-traded funds. Further, according to CXO, the overnight outperformance persists regardless of the day of the week, day of the month, or month of the year. ”Before you quit your day job, you need to know that we’re talking about very small numbers. The overnight gains range from 0.03 to 0.05%, on average. 

January Barometer
According to a research paper quoted by CXO, the market adage that says, “As goes January, so goes the rest of the year,” is true. According to the data, from 1940 through 2006, the market gained 14% or so in years when it was up in January, and more or less broke even in years with down Januarys. CXO says that a portfolio fully invested in U.S. stocks for the balance of the year after a positive January, and invested in Treasury bills in other years, beats a buy-and-hold strategy by 3% to 4% per year. However, the January Barometer only works for U.S. stocks, not elsewhere.

Super Bowl
On average, the market drops slightly in the week before the Super Bowl and gains around one percent in the week after. However, these results were based on data ending with the 2005 Super Bowl, when it was still played in January.

I only had room to describe a tiny fraction of the worthwhile information available on CXO. Everything on the site, which is run by Steve LeCompte, a retired engineer, is free. It’s a great resource.
published 2/3/08

Dividend Detective: If you like dividends, you'll LOVE Dividend Detective

Questions or comments about this site:

Winning Investing   199 Quail Run Road Aptos, CA 95003

(Aptos is 'the beach' for Silicon Valley)

(800) 276-7721 • (831) 685-1932