Reversal May Not Be As Good As It Looks

09/05/08 3:30 PM EST

As of: SPX 1251 HELP  ARCHIVE

To cap off a wild week, we've seen some interesting trading in the indices and several sectors today.  The S&P 500 dropped more than 1% from the open, just enough to have it almost exactly kiss its closing low from July, before rebounding strongly.

The reversal we've seen this afternoon looks good on a chart, but we can't always count on such pretty patterns to follow through.  We'll touch on that in a moment, but first I want to address a piece of feedback I've often received about these comments.

We have subscribers from all walks of life, making it impossible to meet everyone's expectations in terms of content.  Some want only long-term stuff, others only short-term, some want more detail, some less, etc.  One of the areas I've felt could really use some shoring up was the presentation of the various studies we go over in the notes.

Going forward, we're going to use a standard format for the studies we discuss.  When mentioning them in the notes, I will discuss them generally, then link to the study's details elsewhere on the site.

That way those who only want an overview can get that without wasting their time slogging through details, and those who want more a more in-depth look can get all the dates of previous occurrences along with exhaustive display of the innards of the study for further review.  They will be cataloged in the Signpost section if current, and then archived in the Research section when they pass their effective time frame.

So let's start it off today by looking at what's happened in the past when the S&P 500 enjoys a major reversal during a bear market.  For these purposes, we'll look at times the S&P lost at least 1% intraday, then reversed to close in positive territory.  If we fall apart into the close, of course, this won't be applicable, but we're trying to be proactive here and look at a best-case scenario (click here for study details).

During the last bear market, there were 12 times we saw reversals such as this.  The following day, the S&P followed through to the upside six times with an average return of +0.4%.  A few days later, we still had six winners but the return climbed to +1.4%, with the average winner about doubling the average loser.  That's OK, but too inconsistent to trade.

The same holds true when looking back to 1950.  Trading these reversals and holding for three days would have netted you 121 points on the S&P, but that turned into a 222-point loss if holding the trades for two weeks.  Other than the questionable potential for some short-term follow-through, I don't find much bullish about this reversal.

Earlier today, the Nasdaq 100 was on route to carve out its fifth-straight 1% daily loss (allowing for some wiggle room on Wednesday's not-quite-1% decline).  This would have been the first time since February 2002 it has accomplished such an infamous feat, and the sixth time in its history.  The others were 10/20/87, 08/22/90, 04/14/00, 09/21/01 and 02/07/02.  It bounced back heartily the next day each time except in 1990, after which it dropped one more day before rallying about 9% over the next two days.

Combined with the stats from this morning, it looked like a decent bet that weakness today and especially Monday morning would have led to a decent - and probably tradeable - bounce, but we got the reversal early.  I didn't see much of anything that would have predicted the reversal, other than perhaps our short-term guides which pushed solidly into oversold territory.

It's pretty much the same picture for the intermediate-term.  This morning we touched on a few notable indicators (OTC transaction volume, Rydex mutual fund flows and corporate insider trading), and they were about as muddy as everything else.  Bottom line, I'm not seeing much at all that would suggest we're near another intermediate-term rally.

Given what we've discussed, and the bad technical shape of the indices, I'm more interested in looking for a setup on the short side.  A rally back towards 1260ish on the S&P 500 that generates some overbought readings would likely trigger such a trade, and it's what's most prominent on my radar at this point.

Exhaustive Selling Perhaps Good For Short-term Bounce At Best

09/05/08 9:20 AM EST

Good Friday morning...We begin the day with a large gap down in the major indices, as a weaker-than-expected payroll report threw the pre-market futures for a loop.  Oil is a tad weaker (though its correlation to stocks has been breaking down of late) and foreign markets were very weak across the board.

With a 3% loss in the major indices, it should be easy to spot any number of potential positive ramifications.  As we've seen multiple times over the years, the typical reaction to an extreme move in the broad stock market is a counter-reaction the other way.  There are some of those evident now, which we'll cover near the bottom of this note.

Going through all of our indicators last night, a few things stuck out.  One was bad, one good, and one indifferent.  But all notable for one reason or another.

Let's start with the good.  The latest data from the Nasdaq regarding Over The Counter transactions showed a big drop in speculative activity last month.  The OTC market is also known as the Pink Sheets, or penny stock, market - they are stocks that don't meet the requirements to trade on the Nasdaq exchange.

High volume in the OTC market is a sure sign of speculative activity and it has tipped us off a few times over the years to dangerous conditions.  Following is the current chart:

The red dotted line shows the current activity in transaction volume.  Looking back, we can see that the current volume is the lowest since the spring of 2003, after dropping nearly 30% over the past two months.  Since the explosion in volume that marked the blow-off top in the Nasdaq in 2000, the only months showing lower volume than last month were June 2002 - April 2003.

That drop-off in OTC volume apparently didn't translate to the Rydex family of mutual funds, since we saw a very unusual occurrence yesterday.  With the S&P 500 down so much, we typically see those traders flee the long-side index funds, and rush into the inverse funds that profit on a further decline in the market.

Yesterday, however, assets in the leveraged long fund shot higher, as traders not only bet on a rebound, but did so in a leveraged way.  I'm making some big assumptions there, as it may just have been a single large fund that took the trade, and it could have been to hedge a bunch of short positions.

Generally, though, assets in the funds work in a contrary manner.  Traders usually follow the trend of the market, and extremes in the assets are effective contrary indicators.  That's why yesterday's activity sticks out like a sore thumb.

Since the inception of the fund in 2000, there has never been a time assets in the fund rose so much on a day the S&P fell more than 2%.  The only time assets in the fund rose at all with a 3% loss in the S&P was on February 27, 2007.  We did see a quick rebound the next day, then fell back to a lower low.

There were 10 days where assets in the fund rose on a 2% or greater loss in the S&P.  I couldn't find much that was consistent about it going forward - the market rose the next day 6 times averaging +0.3%, with the losers about equal to the winners in terms of size.  It was pretty much the same story across all time frames.

I would generally consider this behavior to be bearish, but I can't justify that by looking at history.  Again, it could just be some rogue fund at work and not a bunch of individual investors that suddenly decided to be contrarians in a major way.

One indicator that did strike me as more bearish than not was the latest report from InsiderScore.com.  Their ratio of corporate insider buying and selling activity showed that these "smart money" executives are not buying the latest chop in stocks, and have pulled back to one of their least-aggressive stances in the past year.

The latest data doesn't include the carnage from yesterday, and the week was shortened by an exchange holiday, so perhaps we'll see a quick turnaround next week.  It's not a good sign, though, that we're not seeing more interest from insiders in the broader market and some of the more heavily-hit sectors.

Given the nature of yesterday's selling, there should be some indications that stocks should rebound, and indeed that's the case.

*  There have been 5 times the S&P dropped 1% the day before a payroll report, then gapped down -0.75% or more the morning of the report (futures are recovering just a bit as I type, but the stats remain mostly the same).  The S&P closed higher than the open 4 times by an average of +1.3%.  Most impressively, the average maximum drawdown (i.e. loss) during the day was only -0.4% while the average maximum gain was +1.9%.  None of the instances lost more than -0.8% during the day.  Holding for three days resulted in all five winning trades with an average of +1.2%.

*  Disregarding yesterday's activity, any time the S&P has gapped down 0.75% or more the morning of a payroll report, it closed higher than the open 8 out of 14 times with an average of +0.4%, and three days later 10 of 14 by an average of +0.4%.

*  Ignoring the payroll report, the S&P has gapped down 0.75% or more after four straight down days 14 times, and closed higher than the open 8 of those times by an average of 1.5%.  By three days later, it was up 11 of 14 by an average of +3.1%.  If we stipulate that yesterday was also a 2% or greater decline, then the S&P closed the day higher 4 out of 4 times by an average of a whopping +5.5%.  Several of those four dates were notable:  10/28/97, 06/15/01, 09/21/01 and 07/24/02.

I could point out several more, but it just becomes redundant.  The bottom line is that activity like we're seeing this morning, after selling pressure like the past several days, has been exhaustive in the past - meaning that we almost always see a short-term snapback, usually immediately.

I'm going to assume the same for today, with the usual caveat - if we trade below the low of the first hour of trading later this morning or afternoon, then the chances of a later-day upside reversal diminish greatly.

There isn't much I'm seeing here from a sentiment-, breadth- or price-based approach that suggest we're anywhere near another intermediate-term low.  We're not seeing "fear" in many of our gauges at all, and there's few hints of even a heightened sense of concern.  Given the very dubious technical condition of the major indices, the only long-side trade I would consider here would be just that - a short-term trade only, based on the tendency for markets to snap back from extreme selling pressure like we've seen, capped off by a large gap down open, due at least in part to the reaction to an economic release.