As Stretched As We Ever See

01/05/09 9:05 AM EST

Good Monday morning...we begin "real" first week of the year with a slightly soft performance in the pre-market futures which have been surprisingly quiet as traders begin to filter back to their desks after a well-deserved extended absence.

First, a quick interruption.

Many of you have asked for updates on our daughter, who suffered a seizure a couple of weeks ago.  We were hoping for the best, that she wouldn't have another episode for months, maybe years, perhaps ever.

Unfortunately that wasn't to be, and she had another yesterday morning.  That makes three episodes in about six weeks, the kind of cluster that we did not want to see as it pretty much rules out the chance that this is "just one of those things".

She will be undergoing neurological testing on Tuesday, and again a couple of weeks after, as we begin the search for what may be causing these.

I know it helps to associate a face with a name, so I've included a family picture below.  My beautiful (and kind, and funny, and smart...) wife Lisa is the one who handles subscription issues, etc., so please understand that responses to those inquiries may be a bit slow as we do what we can for Emmerson.

We all very much appreciate your kind thoughts and prayers, now more than ever.

On to less important matters...

My friend Peter Eliades, who does some great work with cycles and breadth, is a particularly strong fan of the McClellan Oscillator and McClellan Summation Index (they're named after the parents of another friend who does great breadth work, Tom McClellan).

I've touched on these measures before - essentially they reflect the momentum of the breadth of the market.  They look at how far, and how fast, the underlying advance/decline figures are changing.  Because they are impacted by the total number of stocks trading on an exchange, I only use ratio-adjusted figures which make accurate historical comparisons possible.

In the words of the McClellans themselves, The Oscillator is a shorter-term indicator which subtracts a 39-day exponential moving average (ema) of the daily advance/decline number from a 19-day ema of same.  The Summation Index simply tracks a cumulative total of the daily Oscillator readings

In his latest letter, Peter noted the exceptionally high Oscillator reading after Friday's session, which has occurred in the context of a negative Summation index.

To come up with the Summation Index I use, I started with a value of 0 in 1940 and let it cumulate 'til today.  As of Friday, the Oscillator was +105 while the Summation Index was -122.

Peter's point is a good one - what's happened before when we become exceptionally overbought (an Oscillator reading over +100) when the market had been showing a long-term negative trend (a Summation Index under 0)?

Since 1940, this combination has occurred on 45 trading days, covering 19 distinct time periods.  The last instance wasn't the long ago, November 4th of this year, on a day the S&P shot up +4% and many of us thought we were seeing the beginnings of an intermediate-term rally out of those nasty October oversold readings.  Not so, as the market topped out that very day.

Prior to that, we hadn't witnessed this McClellan combination since 1984 as the market was exploding higher out of a year-long funk.  And prior to that, the last reading was in 1982 under similar circumstances.

The table below shows all occurrences since 1940, along with the performance in the S&P 500 going forward:

Overall, the S&P wasn't able to sustain its short-term upward momentum, throwing off a negative return over the next several sessions about 68% of the time.  But in the intermediate- to long-term, these breadth thrusts were most often positive, leading to good returns around 70% of the time.

As now, we saw this setup at the very beginning of the year in 1942, 1974 and 1975.  The two former instances saw the market head almost straight down over the next week or so following these readings; the latter saw the S&P chop around for a couple of weeks, making no progress either way, before resuming its uptrend.

The short-term prospects now are similarly questionable after Friday's session.  We have a whole host of readings that are exceptionally stretched, some historically so.  The STEM.MR Model is at its all-time extreme as of Friday's close, and our Short-term Indicator Score has never been more stretched than it is now.

I could point to several others, like the Cumulative TICK or the Down Pressure or the Price Oscillators or Put/Call Ratios, but it becomes redundant.  Suffice it to say that right now, the short-term rubber band is about as stretched as it ever gets in terms of the aggregate condition of our indicators.

This was the first time since 2002 that the S&P 500 has enjoyed three straight 1% daily gains.  Since 1940, when this has occurred (71 times), the next day was up 40% of the time.  There have been 45 times when it rallied +3% or more on a Friday, and the following Monday was up 58% of the time...however, Turnaround Tuesday usually took effect as that day was up only 36% of the time.

There have been four times when the S&P rallied +3% of more on the first trading day of the year since 1928.  The other three instances (1931, 1988 and 2003) all led to some short-term upside follow through (each gaining at least +2%), but topped out between 2 to 5 trading days later.  By three weeks later, all showed losses, averaging -2.4%.

There have been 13 times the S&P was up 1% or more on the first trading day.  The next day, it was up 9 times averaging +0.5%, but in the weeks following that positive bias eased.  Over the next three weeks, it was positive only 30% of the time averaging -1.7% with a maximum gain during the three weeks (+2.1%) that was half the average maximum loss (-4.2%).

So short-term, given all those historical facts, the undeniably stretched short-term indicators and a host of negative Signposts, I suspect we may get a little more upside follow through over the next session or two (there is a tendency for the market to follow through after a day like yesterday, and the 2nd trading day of the year is one of the most consistently positive), but that should be all given back and then some over the coming week(s).

There is really only one situation in which that would not come to pass, and that's if we're in the early stages of a major intermediate-term rally that's about to last longer than I've been expecting.

There is some good evidence that may be the case - given the money market data we discussed in the latest Data Brief and the AAII data from Friday, in addition to the number of positive indications we've discussed over the past few weeks (here and here and here), I'm still looking for generally higher prices in an intermediate-term time frame (one to three months).

As for what would have me backing off that idea, a move under 850ish on the S&P 500 would be the first sign to take a step back, and a break of 820ish would probably see me completely abandon it.  Also, a continued deterioration in things like the Intermediate-term Indicator Score and the Smart Money / Dumb Money Confidence would have me looking to sell into strength towards the 1000 level or so.

With many thanks,