Some Short-term Negatives Ahead Of Historic Week

11/03/08 8:55 AM EST

Good Monday morning...We begin the day with some modest selling pressure in the pre-market futures.  In a drastic change of pace, there was little in the way of weekend news flow, and we're seeing only minor movements in foreign equities, commodities and currencies.

The focus this week will be on Friday's jobs report, where economists are predicting another disastrous releases.  And oh yeah, there is an election that a few folks might be watching as well.  I have no idea how the market may respond to the outcome, but I do know that in general markets hate uncertainty.  It's seemingly a foregone conclusion that Mr. Obama will be the victor, so the biggest surprise may lie in what happens to Congress.  Because a gridlocked Congress is the best Congress, if the Republicans manage to lose fewer seats than expected, it should be a net positive for equities.

Breadth has been a popular topic over the past month, because we've seen readings so horrid it has surpassed anything else in history.  The selling pressure has been widespread, it has been heavy, and it has been persistent.

Last week that changed, as we witnessed one of the lowest Arms Index readings in history.  As we went over in a Signpost study, that has typically had bullish intermediate-term expectations.

Even on the Nasdaq, there was unusually positive breadth last week.  The advance/decline figures on that exchange tend to run lower than they do on the NYSE, so it's a lot more unusual to see sustained positive breadth readings on the Nasdaq.  But on Thursday and Friday, the Up Issues Ratio on that exchange was over 70% both days (meaning that of all stocks moving up or down, 70% of them moved up).

Amazingly, there has been only one other time we've seen that kind of back-to-back performance in the past decade, which was September 19th of this year.  Unfortunately, that was also the end of that little short-term rally as the Nasdaq 100 (NDX) closed more than 4% lower the following day.

Even more odd was that despite exceptionally positive breadth on Friday, the NDX barely managed to squeeze into positive territory.  Normally we would expect that index to be up several percent on a day with so many advancing stocks.

In fact, I could find only one other day in the past decade when breadth was as positive as it was on Friday yet the index gained less than 1%.  That was on August 28th of this year, another point that marked a failed short-term rally as the index closed down more than -2% the next session and then just kept right on sliding.  Over the past 20 years, there were only two other instances, 09/11/87 and 04/12/90, and a week later the NDX was down more than -2% each time.

There are a number of other overbought indications, related to breadth and otherwise.  For example, odd lot traders (those trading less than 100 shares at a time) spent 57% of their volume late last week buying stock as opposed to selling, a ratio that has been a decidedly negative contrary indicator.

Many of those indicators feed into our Short-term Indicator Score, which moved into one of its most extreme overbought conditions of the year on Friday.

Since the market peak last October, there have been 32 days when the Indicator Score moved to 0.40 or worse (the dashed red horizontal line in the chart above).  Three days later, the S&P 500 was up only 6 times (a 19% win rate) with an average return of -1.6%.  The average amount it managed to gain at its best point over the next three days was +0.8%, compared to a maximum loss that averaged -2.6%.

Only four times did the S&P rise more than +1.5% during the next few days, so we shouldn't see the index hold above 985ish over the next few sessions if it's still in the same general bear-market pattern.

As we showed on Friday morning, there is a definite seasonal tailwind helping stocks this week.  The S&P 500 tracking fund has shown gains 12 out of its 13 years, while the Nasdaq 100 fund has been a perfect 9 for 9.

While recent history has been exceptionally kind to the beginning of November, on Friday we also showed that when October was very bad, then the start to November has been weak as well - maybe all those retail traders who just got their month-end statements decide to sell.

There may be something to that.  The table below shows the first-week returns in a new month when the previous month was up or down more than 5%, both over the history of the index and over the past twenty years.

From the table, we can see that when the prior month was very good, then the new month tended to start with a bang - up 68% of the time since 1950 and 77% of the time since 1987.  But when the previous month was a dud, then the next month's first week was often a rockier affair, up only 42% of the time over the past two decades.

Last week, I was inclined to place more weight on the recent string of positive seasonality during the first few sessions of November, but after more study I think the tendency to see negative returns after a poor month may out-weigh that a bit.  It's a tough call.

It is clear that we're short-term overbought, a condition that has been an absolute death knell for rallies over the past year.  We have started to see a change in character, with the multiple buying thrusts and the ability to hold in positive territory for several sessions after Tuesday's rocket ride, so I continue to believe that's another intermediate-term positive indication.  But with the multitude of short-term negatives, I don't think we're going to be able to make much short-term sustained progress from here until we settle down a bit more.