Feb 10, 2003

http://www.sentimentrader.com

Wednesday, February 12, 2003 8:00 PM EST

For those who don't receive the intraday notes, I want to reprise a couple of charts I sent out this morning. The Investor’s Intelligence survey came out today with a drop in bullishness. The percentage bullish dropped to 40.2% (from 47.2%) while percentage bearish rose to 31.5% (from 29.2%). That’s a decent start for a bottoming process, but take a look at the following charts. They are the cumulative weeks the bull ratio (bulls / (bulls + bears)) spent above the 60% level. With the most recent reading, we just ended a 15-week string, one of the longest in the past decade. The green bars represent weeks ABOVE the 60% level and the red bars are weeks BELOW 60%. As the weeks pass by above or below that threshold, the bars gain height.

1991-2003:

1996 – 2003:

We can plainly see here that large cumulative weeks above 60% (the green bars) have very often coincided with market peaks. The cumulative reading just ended, at 15 weeks, is the third-largest in the past 12 years.

On the other hand, look what it took to kick off the latter stages of the bull market – a whopping 140 weeks below 60%. Most of the better intermediate-term lows were accompanied by at least 10 consecutive weeks below 60%, although during this bear market we haven’t had the time to accumulate so many weeks before a panic low was put in. Regardless, though, these were PANIC bottoms – something we have not yet seen so far this year. This indicator would suggest that we look for one of two things - either a prolonged bottoming process that gradually builds up the bearishness (with at least 10 or so consecutive weeks with the II bull ratio under 60%) and leads to a longer-term low, or a sharp multi-day selloff that generates the panic-type readings we've become accustomed to during this bear. I think the latter is most likely.

The lesser-known lowrisk.com sentiment survey has shown a notable amount of bearishness over the past few weeks. This is a sentiment survey taken online at www.lowrisk.com that has been in existence for about 5 years now and which asks respondents whether they think the Dow Jones Industrial Average will be up or down by more than 2% (or flat) over the next 30 days. The chart below is a four-week average of the bull ratio (bulls / (bulls + bears)).

We can see from this chart going back to 1998 that the current level of bearishness rivals those found at most other intermediate-term lows in the DJIA. There was one "signal" which preceded a five-month trading range in 1999, and one which reached our current level right around 30% before 9/11. Otherwise, each instance of the four-week average dipping to 30% resulted in a good move up, usually within a week.

During the past week, we have begun to see more and more NYSE issues hitting 52-week lows. While that can be taken two ways - either as a sign of weakness or a sign that we are approaching a low - at this point, I have to conclude that it is most likely the former. The chart below shows the number of NYSE new lows as a percent of total issues traded.

We can see that the past few major bottoms have coincided with new lows accounting for an average of around 20% of total issues. Our current figure is a relatively tiny 5%. In order to match readings at recent lows, we would need to see NYSE new lows reach somewhere around 600 - 700 or more. As a caveat, this figure can become considerably higher. In the 70's we regularly saw readings around 30%, 40%, even close to 60%. So while I would consider a new low reading around 600 to likely be indicative of an imminent intermediate-term low, we should have the possibility of much higher readings in the back of our minds.

Although we have evidence that a good low may be forthcoming, there are still several missing pieces that should be in place before I would consider the long side to be high-odds for more than a day-trade. I mentioned the NYSE new lows above, so those should expand dramatically. We should also see a dramatic increase in volume over what we have been seeing, along with wider trading ranges in the indices and the TICK. While the VIX should also expand, the current level of the VIX Fear Premium suggest that a dramatic rise may not be necessary since there is such high uncertainty (compared to historic volatility) already built into that indicator. Until we begin to see more of these panic-type readings, I continue to believe that there is greater risk to the downside than upside. However, this is with the knowledge that we have seen such heavy short-selling activity recently that should some positive surprise occur, there is enough fuel to push us up dramatically in a very short period of time. These are dangerous times on both sides - caution and patience should continue to serve us well.

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.

Monday, February 10, 2003 8:00 PM EST

As I mentioned in the intraday note, we have seen a pick-up in public shorting of NYSE issues during January. For the week ended January 24th (the latest data available), the Specialist Short Ratio stood at 31%. This means that the specialists accounted for 31% of all shorting activity during that particular week. The average ratio since the beginning of 2000 is 43%, so you can see that we're a ways below even the bear-market average. Perhaps most significantly, the four-week average of the ratio has dipped to 35%. This is the lowest average since the extremely heavy and persistent public shorting we saw from 1995 - 1997, and also puts the current average more than 1.5 standard deviations from the two-year and five-year means - in a word, it is extreme and definitely a positive indication for the longer-term health of the market. Over the history of this data, it is quite rare for the market to not stage some type of multi-week rally (at least) when we see such heavy public shorting pressure.

Confirming the public shorting activity in NYSE issues is the most recent action in the S&P 500 e-mini futures contract. Anyone who has read the weekly commentaries or looked at the site know that I only discuss the full S&P 500 contract, and not the e-mini. I've tracked the e-mini for many years, and frankly have found it useless. I've received countless emails from subscribers asking why I don't track the Commitments of Traders data for the e-mini on the site, since it would seem to be the perfect contrary indicator - small speculator positions in a low-margin futures contract. The fact is, it just does not stand up in historical testing. I have not found a way of looking at the data that yields consistent forecasting value, as opposed to the large contract, which has been effective over many years. I've also been asked why I don't adjust the large contract (say, for commercial traders) based on their activity in the e-mini. Since the reporting requirements are different for the large and small contract, we are not comparing apples-to-apples when viewing the various classes of traders. The reporting limit for the large contract is 1,000 contracts but only 300 contracts for the e-mini. That said, over the past couple of months the e-mini contract has been a more effective indicator of future market direction than has the larger contract (I've seen various explanations for this, but as always I'm not especially concerned about the theories behind the indicators). The chart below shows the recent activity in the e-mini:

We can see that throughout November and early December, the commercials were holding steady at around 30,000 contracts net short while the small specs were considerably more optimistic by holding between 20,000 - 60,000 contracts net long. The market soon accommodated the commercials and went against the small speculators by staging a decline into late December. Before the market found a low, however, there was a drastic change in net positions, as the small specs liquidated longs and went hugely net short, to the tune of 110,000 contracts. It should be no surprise the the commercials were net long by this time...by 110,000 contracts. The January rally allowed the commercials to once again pare down their positions while the small specs jumped on the bandwagon and began to aggressively accumulate longs. This was just in time, of course, for the sharp decline we've experienced since. Currently, we've almost come full-circle. Commercial traders have become net long to a degree comparable to mid-December, and the small specs have been aggressively liquidating their long positions. I want to stress once again that this data has not performed especially well over time, but the recent track record is good and I think the action the past couple of weeks is notable and bullish.

The put/call ratio was elevated again today, and once again the talk is that the cause is the same trader who caused the disturbance in the ratio last week. Today there were approximately 72,000 contracts of the QQQ January 2005 45 puts traded, the same contract that was in play last week. This contract accounted for nearly a quarter of all equity option volume, so obviously it distorts the final figure. However, as I said last week, I don't believe it is prudent to try to adjust the figure for what appears to be a single trade. We have no idea what the underlying strategy is, and we also have no idea what other contracts may have been used, or are used every day for trades like this, so it is impossible to create a put/call ratio based only on trades we know to be from wrong-way retail traders with a particular market outlook. In any event, the 10-day moving averages of the total and equity put/call ratios have reached extreme territory and on the surface appear to be a bullish contrary indicator.

Our short-term indicators have worked off their oversold condition with today's "rally" and are in fact now closer to overbought than oversold. The intraday cumulative TICK on the Nasdaq has surprisingly reached a level last seen at the January high, and along with a price oscillator near bearish territory suggest that if we are not beginning a new longer-term trend now, then the very short-term upside prospects for the NDX should be rather limited.

We remain in the same overall situation we were in last week - not oversold enough to anticipate an imminent, 10%+ reversal to the upside, but not overbought enough to create a high-odds shorting opportunity. I continue to believe that with the heavy short activity we've seen (most easily seen via the Rydex ratios and Specialist Short Ratio), the slightest positive geopolitical development could trigger a relentless, multi-day move higher. I would not want to fight that from the short side. However, if that type of occurrence happens anytime soon, it will not have been from a solid enough longer-term sentiment base that would suggest it will last for several weeks or longer. It would most likely be just another speed bump on the bear market highway.

Bottom line: The only edge I see sentiment-wise is the potential fuel for a short-covering rally. If we get positive news that seems to spark a move (like this morning), I think it behooves short-term traders to attempt to ride such a wave - it could carry quite far. Longer-term, I don't see a particular edge either way just yet, and prefer a continued decline to set us up in a better position for an intermediate-term long trade.

- Jason Goepfert

Disclosure: no positions