% of Time VOLUME is Up or Down  Nasdaq Composite Index 1985-2003 Up Down January 89% 11% February 21% 79% March 79% 21% April 37% 63% May 53% 47% June 58% 42% July 53% 47% August 37% 63% September 47% 53% October 95% 5% November 11% 89% December 74% 26% Average 54% 46%

% of Time PRICE is Up or Down  Nasdaq Composite Index 1985-2003 Up Down Average Gain January 78% 22% 4.45% February 63% 37% 1.30% March 68% 32% 0.30% April 63% 37% 0.70% May 63% 37% 1.76% June 68% 32% 1.56% July 53% 47% -0.19% August 58% 42% 0.12% September 53% 47% -1.52% October 58% 42% 0.71% November 63% 37% 2.12% December 63% 37% 2.87% Average 63% 37% 1.18%

The January effect. One of the most reliable seasonal predictions is that January tends to be the best month of the year for investors. Per our investigation above, January months are up an impressive 78% of the time for an average gain of 4.45%. In fact, it should probably be renamed the November/December/January effect because they are often the three strongest gainers of the year.

The spring thing. This theory contends that the January through April market performance is a reliable indicator for the balance of the year. If spring is up, the year will be up. Even if frequently correct, this theory is not very useful because by the time you know how stocks performed during spring a lot of the year's gains are already behind you.

Back-to-school is not cool. Contrary to popular belief, volume drops in 53% of September months, more than the average month (Note: due to steady market growth over the years, the average month sees a declining volume only 46% of the time. Between 1985 and 2003, the average monthly volume on the Nasdaq increased from 1.7B shares traded to almost 35B). The one September reputation that is justified and well deserved is that it is statistically the worst month of the year, dropping an average of 47% of the time for an average return of -1.52% (Note: due to the stock market's historical upward bias, the average month sees a loss only 37% of the time)

October panic. Probably because so many crashes and market bottoms occurred in a month of October, the typical investor has become fearful of them. In fact, Octobers aren't so bad with an average gain of 0.71%. Interestingly, the volume jump that was expected to happen in September materializes in October 95% of the time. Maybe it takes the money managers an extra thirty days to recover from their summer break!

Pre-Thanksgiving buy signal. The adepts of this scheme buy on the Monday before Thanksgiving and sell on the third day of January (and stay in cash for the rest of the year). Simple, and beats Buy and Hold consistently over the years according to proponents.

Election prediction. Popular wisdom has it that the stock market does better with a Republican president. Wrong! Over the last hundred years or so, the average yearly gain under Democrats is almost 30% higher than when the Republicans occupied the White House.
How about: there has never been a loss in the year preceding an election year (almost true).
This game can also be played the other way around by using the stock market as a political prognosticator. When the first half of an election year is up, the incumbent tends to get re-elected (Bush by a tiny margin). When the stock market is down in July, the incumbent more often than not gets re-elected in November (Bush by a wide margin).

Winter blues. This theory postulates that the shortness of days during winter months leads many people to depression (seasonal affective disorder or SAD), which in turn increases their risk aversion, which negatively affects stock market performance during winter months. Great, but we suggest these manic depressives meet with the advocates of the "Pre-Thanksgiving buy signal" or those of the "January effect" mentioned earlier for therapeutical assistance.

Despite our habitual tongue-in-cheek tone, we do not deny that some of the figures can seem very compelling. However, in our analysis, seasonality investing techniques experience tremendous variations from year to year. The chart below, which plots the month-to-month performance of the Nasdaq Composite over the last nineteen years, demonstrates how scattered the patterns can be. Even if some relationships exhibit favorable averages, most are not much more statistically significant than a coin flip.

Because of this, seasonality plays no part in our Trend Timing Model. Instead, we let the market tell us what it is doing and go with the predominant trend for a 70%+ winning trades ratio.

• After logging in, select the "My Profile" navigation tab and you will be able to verify and update your primary and alternate e-mail addresses in the "Personal Information" section
• In the "Preferences - Weekly Update" section we strongly recommend you elect to receive the Weekly Update notification (default setting) because it periodically tests the complete path from us to you
• Send yourself a test e-mail by clicking the "Test E-mail" button at the bottom of the "Current Signal" page

If you still do not receive our e-mails, check with your provider and make sure they do not discard TimingCube e-mails as spam. Our e-mails come from one of three possible addresses:

info@timingcube.com, sales@timingcube.com or support@timingcube.com.

Warm wishes and until next week.

The TimingCube Staff