Trader’s Narrative

This is a guest post by Wayne Whaley, CTA:

Before presenting my latest seasonality study - which may be the most statistically significant - here is a short summary of the previous three studies that I have shared with you. Enjoy.

End of Year Positive Bias
Since 1950, the average annual return on the S&P 500 index has been 8.05%. Over half of that (4.25%) annual return has on average occurred in the three months, November to January.

(Abnormally) Positive Septembers
September is usually the weakest month of the year. But when September has a positive return, the last 3 months of the year have a strong positive bias (21-4) and an average gain of 4.84%. For full details, see: When September Flexes Its Muscle

Positive Septembers - Negative Octobers
If the a positive September is followed by a negative October, then the odds for the last two months of the year go to 10-1 with an average gain of 4.71%.

Seasonality When Jan-Oct Returns Are +10%
If the first ten months of the year provide a 10% or more return, then the final two months of the year are 21-3, with an average gain of 4.99% (see table to the left for details). Perhaps more interestingly, there was not a one percent loss in the the 24 observations.

If the end of the year rally doesn’t materialize with the tail wind of the above statistics at its back, it should be viewed as very bearish for the first quarter of 2010.

This is a guest post by Wayne Whaley (CTA):

I’m not sure whether this market reminds me more of my grandfather’s beagle puppies or the current Secretary of State’s husband, but either way he (the market) doesn’t seem to be inclined to hang around the house for long. It looked like this market might show signs of mean reversion in the second half of October, but then yesterday, it caught wind of the third quarter GDP report and wondered off again, allowing the S&P 500 to post it’s 8th straight up month since the bottom was established in March 2009.

So what’s ahead? A couple of considerations from a seasonal vantage point:

First, since 1950, the average annual return on the S&P 500 index is 8.05%. Over half of that annual return (4.25%) has arrived in the three months November to January.

Second, some respected market technicians have argued that since we didn’t get our usual autumnal sell off, that it may come later in the year. I do agree that the market tends to try to confuse as many of us as possible, but it is very possible the 5% sell off in late October was all we needed to rattle a few cages and I am still inclined to believe that the market’s ability to prevail against traditional seasonal headwinds is a sign of forward strength.

Last month, I posted an article “When September Flexes Its Muscle“, that showed that if the market can manage to post a gain in the seasonally weak September month, the market was has a very high probability of finishing the last quarter positive with an average gain of 4.84%.

Below is a table, with an update of those results for the final two months of the quarter, when both September and October are positive.

Since 1950, the November & December time frame, following up September & October , was 13-2, with a median +6.69% return. If you can allow yourself to consider 1968 to be an unchanged data point (-0.18%), then 2007 was the only noticeable loser in the 15 data points. Under the theory, when the market goes against the seasonal trend - go with the market, the 2007 data point provided strong clues as to what was to come in 2008.

There is a meaningful pullback coming. And staying long for the well known, traditionally strong, year end rally almost seems too logical. But I think that given it has been eight in a row with the holidays to go (rhyme intended), we are well advised to stay with the trend at least through January of 2010 or until the tape shares some information with us to the contrary.

Those of you who follow my commentary know that I am also influenced by the three strong breadth thrust “The Mother of All Momentum Thrust Years” we had this year, the last coming in September. However, if the year end rally doesn’t materialize (ala 2007), one would be well advised to take some defensive measures.

Good luck to us all.

This is a guest post by Wayne Whaley (CTA):

Men Who Can Be Right and Sit Tight Are Uncommon.
Reminiscences of a Stock Operator

The S&P 500 has officially closed the month of September up by 3.68%. A few weeks ago I shared some statistics on the significance of a positive September: When September Flexes It’s Muscle.

To summarize my previous comments: September has a deservedly bad reputation. Since 1950 it has been the weakest month of the year, averaging a return of -0.61% with losses occurring 57% of the time.

Therefore, the market’s ability to buck this bearish seasonal trend in September should be respected and viewed as a sign of continuing strength. When September is positive, fourth quarter returns are positive 84% of the time with an an average gain of 4.84% over that time frame.

Only two of those 25 fourth quarters saw the market down more than 1%. Even “The Mother of all Crash Months” October was positive after positive Septembers 60% of the time for an average gain of 1%.

So now that we have a rare positive September, what now?

This is especially interesting since this September was the 7th consecutive positive month for the S&P 500 index. The inclination is to play contrarian and view this move as a sign of overextension. But, as Lee Corso would say, “Not so fast my friend”. The odds for an eighth positive month are 12-2.

Here is the full data for similar past instances of such “overextension”:

As you can see from the historical data, there is no mean reversion. Instead, stock prices continue their trajectory and tack on even more gains. The sweet spot is 6 months ahead, which from here would take us to February 2010, with a gain of about 8% (and 93% positive returns).

From City Slickers

This is a guest post by Wayne Whaley (CTA):

Before I share some statistics with you on the impact that positive September’s have on the last quarter, I’m reminded of a scene in one of my favorite comedies, “City Slickers”. Billy Crystal’s character Mitch fancies himself a cowboy and in an attempt to bond with Jack Parlance’ character Curly he pulls up next to him on the cattle run and whimsically wishes Curly a good day and ask him if he’s killed anybody today. Curly stares at Mitch, unamused and responds, “Nope but the day ain’t over yet”. Mitch fades back into the pack of his wannabe cowboy buddies, fearing that he might not live through the day.

So with the fact that September is up 4.4% and still has 8 trading days remaining, I hesitantly share the following statistics with you, hoping that I haven’t jinxed the rest of the month. I preface the table with the comment that September is historically the weakest calendar month of the year and the ability for the market to buck the trend is a sign of strength for the rest of the year. The table shows all Up Septembers since 1950 followed by the Pct Change in both October and the total 4th quarter.

Positive Septembers vs. October & 4th Quarter (for the S&P500):

There were only two significant fourth quarter losses following a positive September: in 1973 & 2007. Using the logic that “if the market can buck the seasonal trend, go with the trend”, since most November-Decembers are up, the losses in 2007 and 1973 were a very timely harbinger of worse things to come.

While we’re on the subject of seasonality and how September is the albatross around the stock market’s neck these days, I’d be remiss to not point out that what has been historically negative for equities has been a boon for gold and gold stocks.

Last year, I also pointed out the positive seasonality of gold in September (and the remaining months left in the year). But, as measured by the Philadelphia Gold Bugs Index (HUI) gold stocks didn’t follow their historical path and finished the month pretty much unchanged. However, in mid-September there was a rally that took the Gold Bugs index 35% higher within the month.

In any case, here is a chart of the monthly performance of gold for the past 5 and 15 years:

If you compare this chart with a longer duration seasonality chart you’ll notice that during this current super-bull market for gold (which started in 2000) seasonality has shifted slightly. For these most current years (red line) there are really two big waves of positive seasonality for gold and gold stocks. The first is about to start while the second comes after a correction in October and lasts from mid-October to February of the following year.

Here’s a chart of the Gold Bugs index with the past two September’s and this year’s marked by a green arrow:

I didn’t bother marking the obvious triangle pattern that has formed in the gold stocks index. Prices are getting coiled into a spring and will potentially break out. However, over-head resistance is just 100 points higher at 475-500 - where the Gold Bugs index hit a wall early last year.

Remember, seasonality, while having a powerful and undeniable influence, is a secondary driver of prices. It is more helpful to think of it as context for the actual analysis of trend.