Aug 18, 2003

http://www.sentimentrader.com

Thursday, August 21, 2003 10:01 PM EST

I spent the afternoon at a doctor’s appointment with my wife (she’s fine), so tonight’s comment will be unfortunately brief.

In the model portfolio change this morning, I made reference to the low-volume overbought condition we are currently experiencing, and that it was one of the most egregious examples in history. The computation of this is fairly straightforward – I simply take the NYSE up volume ratio and divide it by the Volume Differential indicator that’s posted to the site. This Volume Differential is the deviation of the 10-day moving average of total NYSE volume from the 200-day average. As volume increases and becomes greater than average, the Volume Differential will rise above 0.

I’m admittedly lacking when it comes to creating catchy monikers, so let’s just call this Volume Adjusted Breadth, or VAB for short. The point of the exercise is to see how much volume is accompanying each market swing. According to the textbooks, if you’re bullish you would like to see increasing volume as the market advances, and decreasing volume as the market declines.

It’s easy enough to calculate on your own:

Calculate the NYSE up volume ratio: UP VOL ÷ (UP VOL + DOWN VOL).
Calculate a 10-day moving average of step 1.
Calculate the Volume Differential: 10-DAY M.A. OF TOTAL NYSE VOLUME ÷ 200-DAY M.A. OF TOTAL NYSE VOLUME.
VAB = step 2 ÷ step 3.

Another use for this is to determine extremes in either direction. For example, if the market becomes oversold on huge volume, that usually marks some type of capitulation move and a low is often soon to follow. Conversely, if you see overbought readings along with very low volume, that suggests a lack of interest (or concern – i.e. complacency) and lower prices are often the result. So, the VAB is constructed to show in one number how breadth is in relation to total volume.

HIGH volume + OVERSOLD breadth = LOW V.A.B. (bullish for the market)

HIGH volume + OVERBOUGHT breadth = MODERATE V.A.B. (neutral)

LOW volume + OVERSOLD breadth = MODERATE V.A.B. (neutral)

LOW volume + OVERBOUGHT breadth = HIGH V.A.B. (bearish)

The chart below shows how this worked at the extremes. Again, a high VAB (the blue line) shows overbought breadth with low volume, while a low VAB shows oversold breadth with high volume.

We can see that our current reading is the highest seen in years. In fact, we would have to go back 14 years to find a higher reading. The current reading is more than 3 standard deviations above the mean since 1965, which basically means it is extremely rare. Even if we take out the very low volume day of August 15^th, we’re still more than 3 S.D. above the mean. Historically, there were 8 other periods in time where we saw VAB readings as high or higher than our current one. While far from a perfect sell signal (the market was higher about 50% of the time over every time period up to 60 days out), the S&P 500 did have trouble making much headway in the face of such a reading (the average return was under 1% over every time period up to 60 days out). The only real failure of the indicator came in 1988 in the wake of the 1987 crash. The VAB reached very high levels in September and November of that year, but the market chugged higher without skipping a beat.

What I outlined on Monday has not changed much. We have a major confluence of readings that have historically lead to declining prices in the broader market. This is evidenced by the extremely low level of the broad-based Composite model and intermediate- and short-term indicator “scores”. In fact, the Composite model and the intermediate-term score are flirting with three-year lows. It’s hard to go against the increasing number of breakouts to new highs in some of the narrower sectors, and if the S&P joins them, as I said, I would not want to continue holding short positions, at least in the short-term. Unless or until that happens, my preference remains with staying with the short side.

- Jason Goepfert

Disclosure: long OEX puts, long BKX puts

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.

Monday, August 18, 2003 11:50 PM EST

I may not be able to get a comment out Tuesday evening, but I want to touch on a few things after Monday’s session.

Throughout this two-month trading range, we’ve seen traders becoming increasing hesitant about taking a firm position in the market. That has been beginning to change in the last few days, and was especially apparent after today’s move. The type of speculation I’m seeing after recent sessions is equivalent to what was seen at the peak in mid-June. While the Dow Jones average and some other important sectors made new highs today, the broader measures such as the NYSE Composite, S&P 500 and Nasdaq 100 did not. Traders are apparently quite confident that these broader measures will soon break out as well, and that type of confidence must be viewed with a wary eye.

While expiration has doubtless been playing with the put/call figures, today’s numbers were once again very low. The five-day average of the equity-only put/call ratio (minus QQQ options) is now sitting at 0.45, which has been seen only three other times since this data has been available – late May 2001, late October 2001 and mid-June 2003. The occurrence in May 2001 pinpointed a top of some import, of course, while the reading in October 2001 lead to only a brief correction while we were in the midst of the recovery from the September low. The reading in mid-June 2003 of course coincided with the top of the move leading to our current trading range. This put/call ratio has been quite effective at highlighting risk from a contrarian point of view, and right now it is flashing yellow. A few more days of this type of activity will change that to bright red, suggesting upside speculation is out of hand.

That is already the case among Rydex traders, at least in the short-term. These traders have gradually been warming up to the rally we’ve seen over the past week, but today they embraced it whole-heartedly. Our Beta Chase Index, which measures the relative amount of money flow among the highest-beta (risky) and lowest-beta (safe, or counter-market) funds, today hit one of the highest readings in its history, at 7.56. Very roughly, this means that there has been 7 times the amount of funds going into the risky funds as opposed to the safe ones. This type of reading has been matched only a few times in the past three years. Every occurrence of such speculation preceded at least a short-term high except one, which occurred when we were coming off the major low in October 2002. Our longer-term measures in the Rydex complex are still neutral, in fact possibly even moderately oversold, but this type of speculation usually results in at least a short-term rest.

Last week, I went over the results of the sentiment survey done by lowrisk.com. At the time, I showed how the respondents were extremely bearish, giving off one of the lowest bull ratios in its six-year history. That changed in a big way this week, as the most recent survey (which includes survey responses through this past Sunday) is showing one of the larger BULLISH readings in its history. The percentage of respondents who thought the Dow would rise by at least 2% over the next 30 days went from 17% last week to 55% this week. Those who thought the Dow would decline 2% or more went from 57% last week down to 28% this week. Traders were pretty confident of their opinion, as only 16% of the respondents were neutral, which is quite low historically. Compared to the other instances of extremely low bull ratios that I pointed out last week, this one is different. None of those other occurrences showed such a large shift to a bullish posture so soon, and the average bull ratio the next week was 32%, quite different from the 66% we’re seeing now. In fact, even up to five weeks later, not one of the other occurrences showed a bull ratio this high. This is a disturbing change so quickly after the bearishness reported last week, and is another sign that traders are being very quick to adopt a bullish posture here. Even with the weaknesses in the survey methodology that I pointed out, this now has to be removed as one of the few bullish signs from a sentiment perspective.

Market breadth is now becoming overbought. Not extremely so just yet, but getting there quickly. Our Down Pressure gauges, which measure the number of points gained or lost in the components of the S&P 500 and Nasdaq 100, as well as the flow of volume into those issues, are now overbought once again. This is confirmed by a number of other readings, including our cumulative TICK indicators. In the past, I’ve included a chart of one of the longer-term intraday cumulative TICK indicators I keep. The intraday indicators as posted to the site are 13-period sums of the NYSE TICK recorded each ½ hour throughout the trading day. I also keep a 39-period sum (three trading days), which is shown below.

We can see that the TICKS have now moved back to the upper part of their trading range, last seen in early July as the S&P was challenging the mid-June highs. While sustained high TICK readings do not suggest that the market necessarily HAS to pull back (note the high reading in late May that just lead to more upside in the market), it does so with a high degree of consistency, at least in the short-term. Out of the last 104 half-hourly readings I’ve taken, I show only 6 with a negative TICK reading. This is extremely unusual.

This bout of speculation across a variety of measures has pushed the Composite model to a very low reading. The chart of this model as posted to the site is a five-day moving average of the daily readings, and it is now approaching its lower trading band (as are all the other models as well). However, today’s daily reading of 23% has been seen only three other times since 2000 – August 2000, May 2001 and March 2002, all of which of course preceded highs of some import.

The breakout in the Dow, Semiconductor Index and other sectors should be given some respect. So should the fact that the market rallied well today (albeit with pathetic volume) in spite of the short-term negatives I outlined late last week. As we saw this Spring, when the market does what it “shouldn’t” do, that deserves some measure of respect. Until today, we haven’t really seen traders get too enthusiastic about a move in one direction or another. While this could be the initial “blast off” phase of a new rally, the fact that we’re already seeing excessive short-term speculation in the face of large resistance in the broader averages seems like overkill. I said this weekend that a move (and hold) above 1005 in the S&P 500 would have me backing off the short side for the time being, and for the moment I am sticking with that, as I could see the S&P then quickly moving to larger resistance around 1015. A break of 1015 would force me to defer to the long side, at least in the short-term, as I don’t have any inclination to short a new high. However, unless and until those levels are breached, and with the high level of speculation and outright complacency we’re now seeing, I want to be more aggressive on the short side and will be using a break of 990 on the downside as evidence that a larger correction could be in store.

- Jason Goepfert

Disclosure: no positions