Morning Market Comment - October 28, 2009

Good morning...Well, after nine months of being seizure-free, our daughter had another epileptic episode yesterday. She was indisposed at the time (OK, she was on the potty), and all we heard was a bang as she slumped over and hit the floor. Besides a bruise on her chin and a couple of parents that aged a few months in two minutes' time, she came out of it just fine. I swear I'll never get used to this, and I'll never get accept the fact that we won't be able to protect her 24/7. Sigh.

We begin the day with a modest continuation of the weakness we've seen for the past few days. The futures are recovering a bit after the Durable Goods report, but are still lower than yesterday's closing levels.

Looking at the S&P 500 exchange-traded fund (SPY), this would be only the fourth time since March that we've seen three straight down days then a gap down open. The others were May 26th, September 2nd and October 2nd. All of them rallied almost immediately (losing no more than -0.25% from the open) and continued to do so over the next week at least. If we can't accomplish a similar feat here, then it would be another "something's changing..." moment.

In late September, when the markets had suffered a Key Reversal Day, we looked at that phenomenon from a number of different ways, all of them dispelling the fear of a major decline that the textbooks have instilled in technicians.

Among the studies we looked at was a technique we've used many times over the years - how a market responds to overbought or oversold conditions in the short-term often gives very good clues about the probability of a longer-term reversal.

Yesterday we touched on the McClellan Oscillator, and the fact that it was already in oversold territory despite the S&P being within 1% of a new yearly high that day. That's never happened before. Generally, however, oversold breadth conditions when the market is trading near a high has been an intermediate-term negative.

With bad breadth in yesterday's session again (the Up Issues Ratio was only 34%), the Oscillator has dropped all the way down to -84, which is the most extreme reading we've seen since early March.

Since 1940, the S&P 500 has had a tendency to bounce in the short-term after such extremes, sporting a positive return three days later 62% of the time and one month later 58% of the time.

Let's see if we can improve upon the accuracy of that longer-term number by watching how the market responded in the short-term, in terms of whether the Oscillator kept falling or started to rebound immediately.

First, let's look at how the S&P performed going forward when the Oscillator hit -84, then kept falling the next day.

S&P 500 Performance If Oscillator <-84 And Then Kept Falling
	1 Week Later	2 Weeks Later	1 Month Later	3 Months Later
Average	-0.5%	-0.8%	-1.3%	+1.5%
% Positive	52%	48%	42%	60%

S&P 500 Performance At Any Time
	1 Week Later	2 Weeks Later	1 Month Later	3 Months Later
Average	+0.1%	+0.3%	+0.5%	+1.5%
% Positive	55%	56%	58%	61%

We can see that when the Oscillator reached such an extremely oversold level and then kept falling, the market not only didn't do all that well on an absolute basis, it significantly under-performed any random time, especially when looking out one month.

So let's flip it around and look at those times when the Oscillator curled up the next day:

S&P 500 Performance If Oscillator <-84 And Then Rebounded
	1 Week Later	2 Weeks Later	1 Month Later	3 Months Later
Average	+0.6%	+1.0%	+1.4%	+3.8%
% Positive	61%	69%	65%	73%

Quite a big difference here, particularly when looking out a few weeks. Two weeks later the S&P was positive nearly 70% of the time with a meaningfully positive average return. A month later, its consistency dipped a bit, but the average return climbed, showing almost the mirror image of the return when the Oscillator kept falling.

This of course brings up the "what if" with today's session. If breadth is as bad today as it was yesterday (about 1,300 more declining issues than advancing issues), then the Oscillator will keep falling, dropping to about -92. That's the bad part.

But the good part is that we actually won't have to see positive breadth today in order to turn the Oscillator - we just need to see less-bad. The advance/decline figure for today could close around -600, and it would still result in an Oscillator reading that's higher than yesterday's...and trigger the more-positive go-ahead performance from the tables above.

Sentiment-wise we haven't seen much of a change. The shortest-term guides are still snuggled in oversold territory with a few more triggering yesterday, but the longer-term ones haven't had a change to catch up to the decline of the past few days. The latest survey from Investor's Intelligence showed some bulls moving to the "correction" camp, but some bears did too, leaving the Bull Ratio precisely where it was last week.

By the way, Investor's Intelligence would really like to see some traffic from our members in order to allow us to continue to post their charts. We get nothing from them other than their permission to display their data, so if you want to keep receiving those updated charts on a weekly basis, please click here and check out their site.

Signal Strength Breakdown	Notes
Sentiment:	Smart/Dumb Confidence Spread is neutral
Trend:	Positive given a rising 200-day average, uptrend from March is still intact
Support / Resistance:	1100 has been stiff resistance, but multiple layers of support remain below
Other Tendencies:	Pullbacks after highs have been positive, but we're seeing more and more "toppy" kind of behavior

Beginning in early March, we discussed a large number of reasons to expect an imminent rally of one to three months' duration, or possibly even a new bull market. During April and May, we went over several studies that suggested that "this time is different" in terms of bear market rallies, as we reiterated in early May.

Since July, the market has consistently rallied smartly from the shortest-term oversold readings, as a healthy market does, and it has rolled over almost all hints of bearish conditions. That kind of momentum tends to persist for long periods of time.

We've seen some periodic bouts of excessive optimism along the way, and the market has pulled back in the very short-term after them. But each time, the major indexes have held technical support, and rallied from even intraday oversold readings, so we've seen little change in the uptrend's character.

Over the past week or so, we've gone over a few modestly convincing studies that show some cracks in the uptrend's potential. We're seeing some very volatile swings in breadth, a deterioration in the number of stocks rising along with the market, a number of big reversals after tests of recent highs, and a hesitation to rally from short-term oversold readings. All of these are warning signs, but until we actually see more of a breakdown (a loss of 1050 on the S&P would be the first step in that regard), we're giving the bull market the benefit of the doubt.

Signal Strength Breakdown	Notes
Sentiment:	Our shorter-term indicators are oversold
Trend:	Short-term moving averages are pointing higher, but there are now a series of lower highs and lower lows
Support / Resistance:	The S&P failed to hold 1070, but due to the big rally there are a series of support levels below, such as 1050
Other Tendencies:	The market has rallied well since March after multiple down days and oversold conditions

As we touched on above, the market has had a consistent tendency, especially since early this year, to rally after multiple down days and then a gap down open.

Even if we don't gap down this morning, the tendency after three down days and three days with very bad breadth figures has been to see the market respond in a reflex rally of one to three sessions. That would coincide with the end-of-October / beginning-of-November period which has been extremely strong over the years.

I'm not at all impressed by the market's recent limp reaction to the short-term oversold conditions which we had yesterday, and the easy violation of what had previously been support around 1070. Combined with a few other "toppy" kind of behaviors that triggered over the past week or so, that's a caution signal.

But the bottom line is that the market is still in a clear intermediate-term uptrend, we're short-term oversold, the price patterns are bullish, and we're about to enter a period of strong seasonality. That latter argument is always secondary, but it's nice to have the wind at your back instead of in your face (believe me, I biked for 20 miles yesterday with 35 mph winds and it...was...not...fun).

So I'll be looking for a bounce here, and that will probably be in effect unless we convincingly lose 1050ish.