Starting To Reach Overbought Again

10/17/08 2:05 PM EST

As of: SPX 1045 HELP  ARCHIVE

It's been a pretty steady march higher right from the opening gap, and the indices are hitting their highs of the day.

In the process, our most sensitive indicators have started to cycle into "excessive optimism" territory again.  The Cumulative TICK is at +7,00 as I type this (it may not close this high before the next half-hourly update), which is one of the highest readings in the past few years.  That is the most-stretched since January 31st of this year and January 4th of last year before that - neither necessarily good times to be looking for further gains.  The Price Oscillators are also at the tops of their ranges.

Even during very positive uptrends, stocks usually have trouble maintaining additional rallies when we get this stretched.  While they sometimes continue to chug higher in the very short-term, over the next few days we most often see those gains clawed back.

The one consistent exception from that rule is when a market is deeply oversold on an intermediate-term basis and all of the latent buying pressure overwhelms any short-term overbought readings.  That's what we discussed on Tuesday, as a test of the idea that we were bottoming after Friday's reversal.

Stocks failed that test miserably, as we rolled over immediately from Tuesday morning's gap open.  We still have some room here before the STEM.MR Models get into truly overbought territory, and with the cross-currents from one of the largest option expirations in history, anything can happen heading into Monday morning when the option-related adjustments run their course.

Besides the looming overbought readings, I'm a bit concerned that some credit-related indicators aren't acting as well as they probably should be.  While short-term T-Bill yields have doubled again today (a good sign that traders are becoming more risk-seeking), there hasn't been much improvement in other indicators like the cost of CDS protection for non-bank corporations.  Even more troubling, the HYG (high yield) and LQD (investment grade) closed-end bond funds are actually down today.

Today's recovery from the gap open certainly looks like further confirmation of the idea that we're in the process of forming that intermediate-term low we've discussed since late last week, but again the main focus from here will be Tuesday's highs and if we can carve out a classic double-bottom.  A move over those highs in the face of short-term overbought conditions over the next several sessions would be about all the confirmation we could ever ask for.

Divergences Galore

10/17/08 9:15 AM EST

Good Friday morning...We begin the day with stiff selling pressure in the pre-market futures after another extremely volatile overnight session.  The S&P was up as much as 15 points from yesterday's close, and down as much as 30.  A positive article from Warren Buffett in the NYT seemed to help spur a 30 point rally off the lows, but we've backed off nearly 20 points in the past hour.  These are the kinds of movements that used to take a week to play out.

An aspect to yesterday's trading that is getting quite a bit of attention is the number of divergences among sectors and indicators, compared to the extremes we saw at last Friday's low.

For example, yesterday the Nasdaq 100 made a slightly lower low, but the other major equity indices remained well above their own lows.  On the Nasdaq, there were more than 1,600 stocks hitting new 52-week lows last Friday, but at the worst point yesterday there were "only" 390 that did so.

Anyone who's read these comments for any length of time knows that I'm not a big fan of divergence analysis.  It's too subjective, and we never know when the divergence will end, so it's often useless from a timing perspective until after the fact.  There are some exceptions to that, but they're rare.

Anyway, one of the more-discussed divergences from yesterday's trading is that between stocks and the "fear index", or VIX.  While stocks were (mostly) holding above their worst levels from last week, traders were bidding up the price of options to the point where the VIX hit a new high.

Like most divergences, though, I can't find much that's predictive about this.  Going back over the past 20 years, there were 11 occasions when the VIX hit at least a one-year high, but the S&P 500 didn't even make a one-month low.

Three trading days later, the S&P was positive 7 times by an average of +0.6%.  Two weeks later, it was still up 7 times with an average of +0.5% and a month later 8 times with an average of +0.8%.  None of those are appreciably different than random.

In the past, we've discussed divergences between the indices or between indices and sectors, and those have generally been more consistently predictive than divergences with various indicators like the VIX.

If we look for times when the Nasdaq 100 hit a new yearly low, but the S&P didn't even make a one-month low, then three days later the S&P was up 39% of the time (9 out of 23 instances) and showed an average return of -1.4%.  That was a short-lived negative, though, as two weeks later the S&P was up more than it was down, and by a month later it sported a positive return 65% of the time averaging +1.6%.

The indices didn't just diverge yesterday, they also reversed in a major way.  If we look at the same study above, but look for times when the NDX managed to reverse enough to close in positive territory, then we're only left with 5 instances.  Unfortunately, over the next three sessions the S&P 500 showed a negative return each time, averaging -3.5%.  Once again, it was short-lived - by two weeks later, the S&P was up 4 of the 5 times averaging +1.4%, and by a month later all 5 times with an average of +4.0%.

Over the past couple of weeks, due to the severity of the selling pressure and the historic volatility, we've throttled back on the number of studies we do that compare current market activity with that of recent years.  Really the only valid comparison we have to what we've seen over the past month is 1929.

There is still no evidence that volatility is going to decline as we return to a more "normal" market, but if last Friday was an intermediate-term low then that should begin shortly.  Instead of 5% - 10% intraday swings, we should be heading towards 2% - 3% max.

That kind of environment will help us return more to the kinds of indicators and behavior we've been studying over the past five years.  One example of what we used to be able to look at is behavior surrounding option expirations.  I don't know how much weight to place on these kinds of studies in this market environment, but generally when the market gaps down on an expiration, it hasn't been all that great of a sign.  Out of 9 times the S&P dropped by 1% or more at the open on an expiration, it closed higher than the open only 3 times, though the average return was only -0.1%.

Given everything we've discussed over the past couple of weeks, and especially the past week, my thought has been that last Friday should by all rights have been "it".  We shouldn't see appreciably lower prices from that point in the short- to intermediate-term.  The market's behavior since that point has been pretty much in line with past major lows, though the retracement we saw into yesterday morning was certainly deeper than I was expecting.

Yesterday our shortest-term guides nearly made it to oversold before the rally began, but indices like the S&P were knifing through any technical level that should have been support.  In order to even consider taking long positions, I wanted to see enough buying interest to take us back over 900 in that index, which we did manage to see and got some good follow-through into the close.

The obvious question now is whether this qualifies as a successful test of Friday's panic.  Technically, we have to wait until we overtake the highs from Tuesday in order for such a bottom to be "confirmed", and it's something longer-term traders may want to consider.  We've seen a level of volatility that rivals any in the past century, and so waiting for some confirmation of buying interest and a reduction in volatility isn't a bad idea.  There should still be decent gains in the intermediate-term even after waiting through a short-term rally.

For the short-term, I'm a bit troubled by the stats we discussed above in terms of the behavior after previous divergences between the S&P and NDX, and also option expiration.  Again, because of the outstanding circumstances we're seeing, I just don't know how much weight we can place on studies like that - the market is being driven by emotion, not by "rational" behavior.  Today is one of the largest option expirations we've ever seen, so we could see further wild and unpredictable movements in stocks right into Monday morning.  I'm not doing much at all in the short-term because of it.