Sep 29, 2003

http://www.sentimentrader.com

Thursday, October 2, 2003 9:40 PM EST

My friend Tony Dwyer, Managing Director at FTN Midwest Research and frequent contributor to Minyanville.com, asked me to take a look at breadth on days similar to yesterday. The reason is because NYSE breadth was extremely strong, as I show that we had 2699 advancing issues to 569 decliners, a ratio of 4.7 to 1. To see another day with advancers outpacing decliners by more than 4.5 to 1, we would have to go all the way back to April 1994. Historically, these days are extremely rare. Since 1965, I show only 53 other occurrences out of 9,696 days, a ratio of 0.5%.

Tony was most curious to see how the market performed after these occurrences, especially in light of the underlying trend at the time. It turns out that the market performed exceptionally well afterward.

Out of the 53 days that showed advancing issues outpacing declining issues by 4.5 to 1 or greater, the S&P 500 was higher 30 days later 70% of the time, with an average return of 2.5%. 60 days later, it was higher 77% of the time with an average return of 5.2%. 90 days later, which I will concentrate on the most, the S&P was again higher 77% of the time (41 out of the 53 times), and the average return was 8.3%.

These numbers are a little misleading, since some of the occurrences were bunched together. For example, in the Fall of 1974, there were five instances within two months of each other. So, next I looked at only those readings which were at least 90 days from a previous reading. That way, the 90-day return of one instance would not overlap into another. After weeding out those bunched-up readings, I came up with 29 distinct occurrences. Here, the S&P 500 was higher 90 days later 20 out of the 29 times, with an average overall return of 6.4%. The average gain was 12.1% while the average loss was 6.2%, for a greater than 2-to-1 ratio. The maximum positive return was 29% and the maximum loss was 17%. These returns are assuming that one held throughout the entire period, and do not consider the drawdown one would have suffered (or potential maximum gain one could have enjoyed) within that 90-day window. Perhaps most notably, there were 12 distinct occurrences where the 90-day return exceeded a 10% gain; there were only two that exceeded a 10% loss.

To drill down further, we next looked at whether the S&P was trading above or below its 200-day average at the time of the lopsided breadth reading. This turned out to create a big difference in performance, as the table below shows. The green shading highlights which trend indicator (i.e. above the 200-day average, or below) performed better relative to the other one.

How the S&P 500 Performed 90 Days After a 4.5-to-1 Positive Breadth Reading 1965 - 2003
	When > 200 DMA (11 occurrences)	When < 200 DMA (18 occurrences)
% Positive	91%	56%
Average Return	8%	5%
Average Gain	10%	15%
Average Loss	4%	6%
Maximum Gain	19%	29%
Maximum Loss	4%	17%

What stands out immediately is that when we saw a breadth reading of 4.5-to-1 or better AND the S&P 500 was trading above its 200-day moving average, the S&P was higher 90 days later 10 out of the 11 times. The average gain was a large 10%, as opposed to an average (and only) loss of 4%.

By only taking these signals if they occurred when the S&P was above its 200-day average, you would have missed out on the biggest gains seen during this time (as the market snapped back from depressed levels). However, you also would have avoided several large losses as the market kept trending lower after these one-time breadth thrusts. Still, whether the S&P was trading above or below its 200-day average, buying the market after such lopsidedly positive breadth would have been a winning strategy after 90 days on balance.

As always, there are caveats. 9 of the 29 occurrences transpired while the U.S. was in one of the greatest bull markets in history, from 1982 through the 1990’s (although only 2 of the top 5 largest returns in the study occurred during this period). Also, the NYSE switch to decimalization and the increasing number of non-operating companies may have an impact on these figures.

As I mentioned in the intraday comment this morning, the 3-day cumulative TICK indicator on the NYSE that I’ve discussed several times before is now over +20,000 for only the second time in three years (that other instance was on 1/6/03). I’ve included a chart below of the current reading, with red arrows highlighting the other times the indicator reached at least +18,000. We can see that even when the S&P 500 was in an strong uptrend throughout April and May, when the TICK became nearly as extended as it is now, the market took at least a small breather in the short-term. High TICK readings can be a sign of accumulation, and is not necessarily a bad thing coming off a low. But when the longer-term measures such as this become as extended as they are now, at the very least the market tends to rest in the short-term.

We’re now very mildly overbought on a short-term basis, and the overwhelming sense of pessimism that I noted on Tuesday appears to have dissipated to a large degree (a 3% rally tends to shake that out). As I said then, from what I was seeing, the most likely scenario from the information I look at would be a bounce over the next few days, which we got, then a failure and a further decline. I was put off on getting too anticipatory of a decline by the pessimism that I noted, but now that has been worked off to some degree. In the very short-term, like the next day or two, I don’t see a real edge and will wait to see how the jobs report tomorrow is accepted.

Longer-term, now that we’ve got the three failures I had been looking for, I expect that there will be more downside to come and won’t be interested in intermediate-term long positions from a risk/reward perspective until more of the speculation that we have seen has been worn down. My intended aggressiveness on the short side, however, has waned somewhat. Our Composite model, and the “score” for our intermediate-term indicators (see chart here) are in or very near positive (for the market) territory. In fact, the indicator score is that its highest level since the March low. Part of this is due to the Commitments of Traders information, which as I’ve been saying is questionable, but even without that data the score would be quite high - as least as high as it was in early August.

- Jason Goepfert

Disclosure: no positions

This disclosure is not intended as trading advice in any form. It is meant as a note to subscribers that the author may have a position directly affected by the market outlook reflected in the commentary. Although the author takes great pains to remain objective in any commentaries, it is only fair that readers should know that the author may have taken positions in accordance with his market outlook. Positions can and do change at any time, without notice to the reader.

Tuesday, September 30, 2003 10:40 PM EST

With a combination of conflicting sentiment inputs, a bed-ridden wife (doctor’s prescription for late-stage pregnancy), and two sick children, I will be cutting short tonight’s comment, so my apologies up front.

This weekend, I stated that if the market couldn’t rally several percent immediately, then that will be the third failure I was looking for to determine if it was likely that the larger trend had changed. Though we did rally quite well on Monday, we were not able to hold it, and to me that qualifies as a failure. Since we’ve now failed at a previous high, broken the simple downtrend line from August 6^th, and failed to create a sustained rally from deeply oversold short-term conditions, the odds are high that the intermediate-term momentum will now rest with the short side.

The problem with this thought is that it now seems as though everyone else has it too. I can’t remember a time where there has been as much of a consensus as now from those sources that I read and talk to – that the daily trend has changed from up to down, and rallies are now meant to be shorted. Almost to a person, those that I come in contact with are looking (and hoping for) a rally so that they can initiate short positions. I am not a fan of relying heavily on anecdotal evidence, but I was truly struck today by how universal the strategies are among both trend-followers and those who normally trade in a contrarian manner. Even though we’re only 7 days and roughly 4% from a new recovery high (in the S&P 500), the pessimism I’m seeing is palpable. This is also being reflected by some of our real-money sentiment indicators (e.g. the equity p/c ratio with QQQ options removed, the Rydex RSI Spread and Beta Chase Index, and the VIX/VXN – check the site for reference).

For a couple of weeks, I’ve been laying out what I needed to see to determine that the short side was the higher-odds bet. Now that we have it, I’m a bit hesitant to become too aggressive because the boat seems awfully one-sided. Like I said, I can’t remember a time where so many traders had exactly the same idea. IF we were to rely on this anecdotal evidence (which I wouldn’t necessarily recommend), then we could make the assumption that what everybody thinks is going to happen will most likely not. So, the most likely scenarios going forward (in the order most likely to happen) would be a stiff rally from somewhere in here, a trading range where neither side makes much headway, or an immediate and powerful decline with no bounces to allow short-sellers to leg into positions. The market action so far is certainly reminiscent of the last scenario.

Bottom line, I think that if we can regain some upside momentum very quickly and overtake some short-term resistance points, then there is enough of a belief that we’ve seen the highs of the year that we could rally enough to at least challenge them, if not overtake them. If not, and this weakness persists, then the consensus may have it right and we will continue to fail. My work suggests that we should see something of a bounce over the next few days which ultimately fails and we head lower, but again I am cautioned by the wide acceptance of that same opinion.

Again, I apologize for the brief and unconventional comment this evening. As I recover from today’s scramble on the home front and have more of a chance to analyze the recent activity, I may post more in the intraday comments.

- Jason Goepfert

Disclosure: no positions