Trading Range Sucking In More Believers

Bottom Line: The bullishness of newsletter writers has barely waned during the decline of the past two weeks, but it is still about average for the kind of declines we’ve seen. Still, history suggests that it would be much better if we could see a more pronounced decline in bullish opinion.

From the close on June 25^th to the close on July 9^th, the S&P 500 lost 1.9%, while the Nasdaq Composite lost 3.9%. In spite of those losses, the amount of bullishness in the Investor’s Intelligence sentiment survey decreased only slightly. I went over the sharp decrease in bullishness in the lowrisk.com survey in the last comment, but these different survey populations certainly had different ideas about the decline over the past couple of weeks.

While this stubborn bullishness seems unusual, in fact it is not. The table below outlines the typical moves in the percentage of bulls, bears and the bull ratio (bulls / (bulls + bears)) in the I.I. survey when the S&P lost between 1.8% and 2.1% over any two-week period from 1971 to the present. Also presented is the data for two-week losses in the Nasdaq of between 3.8% and 4.0%.

	Bull % Change	Bear % Change	Bull Ratio % Change
S&P	-3.2	+1.0	-1.9
Nasdaq	-1.3	+1.1	-1.4
Current	-1.9	+1.4	-2.1

From the table, we can see that over the past 30+ years, whenever the S&P had dropped around 2% over a two-week period, on average the bull ratio of the I.I. survey declined by 1.9%. Whenever the Nasdaq had declined around 4%, the bull ratio dropped by 1.4% on average. Over the most recent two-week period, the bull ratio slipped by 2.1%, which is greater than both averages. So the past couple of weeks have NOT been unusual in that respect.

Does it even make a difference if the survey showed an increase in bearishness when the market declined? To see, I again looked at those times the S&P declined about 2% over a two-week period, and when the Nasdaq declined about 4%. I then looked at the times bullishness dropped the most, and those times it actually increased despite the losses in the market, and compared that with how the S&P and Nasdaq performed over the next several weeks.

	2 weeks later	4 weeks later	6 weeks later
S&P
Largest drops in bulls	1.0%	1.4%	1.6%
Largest rises in bulls	-0.5%	0.1%	-0.4%
Nasdaq
Largest drops in bulls	-0.1%	1.6%	3.2%
Largest rises in bulls	0.1%	-0.2%	0.8%

For both indexes, we can see that when the survey respondents became significantly more bearish as opposed to more bullish, the market had a tendency to outperform. Looking out 6 weeks in the future, the S&P was an average of 1.6% higher when bullish opinion dropped markedly, compared to 0.4% lower when bullishness actually rose in spite of lower prices. For the Nasdaq, we can see that when the bulls ran for cover, the index did quite well by tacking on an average of 3.2%, as opposed to only 0.8% when the bulls were confident enough to stand their ground.

While these results shouldn’t come as a surprise to those with a contrarian bent, it is apparent that if we want to see prices rise, it is much better to see other traders turn bearish than it is to see them remain bullish in spite of falling prices. Our current situation is about average in that we have seen the population of the Investor’s Intelligence survey pull in their horns about as much as they usually have given the losses in the market, so there is nothing exceptional about the past couple of weeks.

What is unusual is that bullishness remains so high on an absolute level. With bulls currently making up well over 50% of respondents, and bears less than 20%, the amount of bullish opinion is still more than 2 standard deviations about the long-term norm. By definition, that’s a very high display of confidence and I’ve shown many times before how that most often leads to underperformance.

Bottom Line: The 10-day average of this popular indicator is once again giving an historic extreme reading. Precedent is good for some kind of snapback to occur when things become so extreme.

The TRIN on the NYSE has once again given traders something to talk about. With the 10-day average now sporting a reading of 1.77, we are seeing one of the highest two-week readings in its history. In addition to the NYSE, the 10-day TRIN on the Nasdaq today hit 1.67, its highest reading in over a year.

Back on March 22^nd, I also noted this type of extreme, as that was the last time the measure had reached such heights. Let me parse a couple of paragraphs from that comment:

“Since 1940, after any day that showed a 10-day TRIN reading of 1.75 or greater, the Dow Jones Industrial Average was higher 71% of the time 30, 60 and 90 days later, with average returns of 1.5%, 2.4% and 4.4% respectively. After six months, the Dow was higher a remarkable 86% of the time (145 out of 169 days), with an average return of 6.5%.”

“If we just look at the 55 days since 1950 when the 10-day average reached 1.75 or higher, then the Dow was higher after 90 days 52 times (95% of the time), sporting an average return of 8.7%. The average gain was a hefty 9.4%, while the average loss was 2.3%. Speaking of those average losses, as I stated there were only 3 days out of 55 that showed a loss after 90 days. Two of those days were in early June 1962 when the market was accelerating its downtrend before forming a major multi-year low later that month. The other day that showed a loss was in early July 2002, when (once again) the market was in its acceleration phase before forming a major low later that month.”

There are some caveats with this indicator, as I’ve noted many times before. What I find fascinating is that every time these kinds of extremes show up, so do the doubters who talk at length about why this time it won’t work. While I think it’s important to note the failures as well as successes, to me the bottom line is that historically (even very recently) these types of readings from this particular measure have resulted in higher prices for the broader market a remarkable percentage of the time. In my opinion it is not reason alone to make an investment decision, but it should certainly at least be factored in to your analysis.

We’re now seeing some breadth measurements that are suggesting the broader market is as oversold as it was in March and May. However, there are several notable differences that I think are as equally important.

THEN: Investor opinions were suggesting that some doubt was forming about the sustainability of the bull market.

NOW: Our AIM Model is still in deeply overbought territory, telling us that there is little concern, at least among those who respond to these surveys (with the exception of the lowrisk.com survey, but that is only one week’s worth of data).

THEN: The real-money metrics devised from Rydex asset flows told us that these traders were pulling money out of the bull funds in sizable amounts.

NOW: There is no panic among these traders, at least not yet. Our short-term measures here are somewhat oversold, but take a look at the Bull Stochastic – we are nowhere near an extreme in pessimism here.

THEN: Traders were bidding up the price of options, as implied volatility indicators (i.e. VIX and VXN) spiked higher by 40%+ as the market declined.

NOW: As recently as yesterday (pointed out in the intraday notes), the VIX was more than 10% below its 10-day moving average, which itself is quite low. While there may be structural reasons for the suppressed volatility measures, in general it is helpful to see these spike higher as the market falls.

With traders estimating that future volatility will continue to be low, and not becoming panicked as we fall, it is apparent that the idea of a lengthy trading range is catching on. While it has been my thought since January that this is what we’d see, this type of behavior is giving me some second thoughts as to its length. My approach to this decline has been that I felt it would give a high-odds, low-risk opportunity to once again trade from the long side once we approached the lower end of the range, but trader attitudes towards this fall is suggesting that we may need to see a break of the May lows to actually get some measurable doubt into traders’ psyches. So far, only a small smattering of indicators are showing that kind of doubt – it’s nowhere near what we saw in March or May. Lower prices are not the only thing that could increase the concern, as an extended period of time with little upside movement could serve the same purpose. Either way, it does not appear as though a sustainable, intermediate-term rally is imminent.

Short-term, we still have not reached a confluence of oversold readings in our shortest-term indicators. Any long-side trades should remain very short-term, as there is scant evidence that pessimism has reached a point where the realization that things aren’t so bad can carry us higher over a several-day period.

NOTE: Due to demand, the intraday comments are now available via email! To sign up, please click the link for “Get Emailed Updates” under the Subscriber Support section on the left.

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.