Short-term Clues For Intermediate-Term Trend

10/13/08 2:50 PM EST

As of: SPX 1075 HELP  ARCHIVE

It's been a steady march higher today as the indices spent the bulk of the day hitting higher intraday highs while satisfying most of the requirements for a trend day.

A trend day is one of those formerly rare days that opens at one extreme and closes at the other, with minimal retracements during the course of the day.  When we get into the mid-morning hours with most sectors participating, and buying pressure coming in every time the NYSE TICK approaches the zero line, then we usually close at or near the high of the day.

The steady trend has started to get a little sloppy over the past hour, as none of the major indices has been able to hit a new intraday high, and the small-cap Russell 2000 has backed off significantly.  We're also seeing the TICK become a little "loose", venturing further below the zero line than we usually see, but so far I don't see anything too terribly out of sorts.

Also notable is that our most sensitive indicators have moved firmly into overbought territory for the first time since October 1st.  Overbought conditions have been deadly to this market since last October, so as always this will prove to be a good test of the potential bottoming process.

Virtually every time the market forms an intermediate-term bottom, it is kicked off by a burst of buying pressure that triggers short-term overbought conditions.  When we see such insistent buying interest that prices hold up or even move higher in the face of those overbought conditions, then it's a sign of latent demand that usually continues to take us higher over the coming weeks.

Breadth is exceptional today, with more than 90% of all issues on the NYSE showing gains from Friday's close.  I could find only three other times that the S&P hit a new yearly low one day, then carved out a 90% day the following day (05/27/70, 07/12/74 and 03/11/08).  Each time the S&P managed to continue higher over the next one to two weeks, at least.  We're also seeing the number of stocks on the NYSE hitting new 52-week lows drop to 59, from an astronomical and by far all-time record of 2,524 on Friday. 

Since last October, and especially over the past couple of months, the market has crumbled almost immediately when we hit short-term overbought readings.  So I'm keeping a close eye on how we behave over the next day or so - as I noted earlier, given everything we went over last week, I do believe that Friday was the exhaustion point, and we shouldn't crack too hard under these overbought readings.  My general rule of thumb is that we shouldn't see more than about 50% of the gains be given back.  We've had an abnormally large range since Friday's low, so that would be quite a loss in point terms, but again I think we have room before becoming too concerned about falling to new lows.

Post-Crash Expectations

10/13/08 8:50 AM EST

Good Monday morning...We begin the new week with significant buying pressure in the pre-market futures, with most domestic indices up 4% or more from Friday's closes, and most overseas markets rising 5% - 10%.  Trading in the cash market of US dollar-denominated fixed income is closed today.

In a Data Brief after the first in a series of mini-crashes that we've witnessed over the past two weeks, we looked at price behavior in the Dow Jones Industrial Average during the past 100 years when it had experienced similar wipeouts.

A common thread among those precedents was the market most often saw a one- to three-day wicked upside rally which then failed and tested (often exceeding) the panic low from a few days prior.  In every case, the index formed a more sustainable low within 5 to 10 trading sessions.  In only one case (1937) it did not lead to at least a one- to three-month rally.

Assuming Friday's low holds, we've held fairly close to that template, though the selling pressure last week was far beyond the norm.  If we take a look at the three instances where the Dow violated its first panic low by a significant degree, another pattern emerges:

In 1929, it violated the panic low by 14%, bottomed 9 days later then rallied +25% over the next 6 days before taking a breather.  In 1931, it violated the low by 20%, bottomed 9 days later, then bounced 22% over the next 5 sessions.  In 2001, it dropped by an additional 10% over 4 days, then jumped 14% over the next 8 days before pausing.

The 2001 instance is probably the least comparable to our current instance, not only because of the tragedy of 9/11, but also because the violation of the low was the least severe.  Friday was the 9th trading day following the mini-crash on September 29th, with a total loss of about 18% on a closing basis from the low on the 29th, so we're pretty much in line with the precedents from 1929 and 1931, both of which saw rallies of more than 20% over about a week's time.

We spent a good deal of time last week going over various extremes.  They went from becoming interesting, to notable, to historical, to downright unprecedented as the week progressed.  By Friday, we were seeing a multitude of signs that we've just never seen before (an incredible 75% of all securities on the NYSE hitting new 52-week lows, the DJIA suffering 7 straight days with at least a 1% decline, the worst week of price losses ever, etc.).

Heading into mid-week, we looked at several factors suggesting that Friday should be the exhaustion day, with the market forming a short- and intermediate-term low at that point.  As the selling severity increased, so did my expectations of the subsequent rally.  Instead of the usual bear-market reaction (+5% - 15% over one to three months), my expectations changed to +10% - 15% in the short-term and +20% - 30% over the intermediate-term.

Looking at the instances above, we may have to adjust those expectations upward, at least for the short-term, with a possibility of even a +25% gain from Friday's lows.  That would take the Dow to very roughly 9,800 and the S&P to 1025 over the next one to two weeks.  I think that's as valid an outlook as any, and they are my approximate targets against which to expect a probable pause in the post-crash rebound.

We've discussed before that large (2% or larger) gap openings in the S&P have a tendency to be poor predictors of future strength, but obviously it's a bit difficult to read a lot into that given the circumstances.  As per usual when we see this large of a gap up opening, I'll be watching the first hour of trading very carefully - if we can make higher intraday highs after the first hour, my expectation will be for additional strength into the close, with the probability of a downside reversal becoming less and less likely as the day wears on.  If we fail, it will most likely be within the first hour.