Feb 17, 2003

The most recent AAII sentiment survey came out with a notable increase in bearishness, as the bulls dropped to 21% (from 22% last week) while the bears rose to 58% (from 38% last week - thanks to Mark Young at Traders-Talk.com for the early release). The bull ratio (bulls / (bulls + bears)) dropped 10% to 27%, which is the lowest bull ratio since 1992. This is a very noisy survey week-to-week, so I prefer to watch it on a 4-week moving average basis, and that average dropped from 39% last week to 36% this week, the lowest since 1993. While the average is low and suggestive of a rally, the weekly number is hinting that it may not have yet arrived, as this survey curiously appears to lead the market by one-to-three weeks. The table below outlines the lowest point the AAII bull ratio reached near each major low over the past five years:

SURVEY DATE*	BULLS	BEARS	RATIO	4-WK AVG	MKT LOW	# DAYS
08/21/98	23	45	38	44	09/01/98**	8**
10/01/99	34	28	55	59	10/18/99	12
03/24/00	39	50	44	66	04/14/00	16
03/16/01	28	47	37	48	03/22/01	5
09/10/01	31	44	41	47	09/21/01	6
07/12/02	26	54	33	43	07/24/02	9
10/04/02	29	55	35	39	10/10/02	5
Average	30	46	40	49	N/A	9

02/14/03	21	58	27	36	?	-1 ?

We can see from the table that we are more extreme now than prior to any other major low. We have the fewest bulls, the most bears, the lowest bull ratio (of course) and the lowest 4-week average. However, one thing stands out - at every major low, the AAII bull ratio reached its lowest point at least 5 days PRIOR to the ultimate low in the market, and the average waiting period was 9 trading days. If last week was the low, then this would be the first time that the market bottomed before the survey did. If we go by previous lows, then we should expect the ultimate intermediate-term low for this move to occur sometime between today (didn't happen) and March 7th, with the "average" date being February 26th (next Wednesday). While this is simply a general guideline and I by no means would trade off of it, I do think there is some underlying mass psychology at work that is reflected in the survey data. I think what we see most often in this survey is a hint of panic when it is apparent that technical support is broken or some terrible business or economic issue makes it on the evening news. This results in the bull camp becoming scarce as investors become bearish or neutral at worst, and the bull ratio of the survey becomes relatively extreme. Then, as the market is in free-fall stage, some investors wise up to the fact that it may actually be a buying opportunity, that panic is a chance to get long, and they begin to migrate back to the bull camp once again. At this point, we've now seen the "panic" reading from the survey, and I suspect lower prices would be met with higher bull ratios (provided we don't suffer an all-out crash).

One positive for the market continues to be the action in the Rydex mutual fund complex. I highlighted the asset movement in the index funds this weekend, and I want to revisit it again. Since February 12th, $360 million has left the short-side funds. Only $145 million of that money (or what we assume to be that money) made it to the bullish funds and $81 million to the money market. $133 million has left the index funds or money market altogether. Most notably, of the money that left the bearish funds, 77% of it left the leveraged funds and only 23% left the Benchmark, or un-leveraged, funds. Also, of the money that then flowed into the bullish funds, only 55% of it went into the leveraged funds while 45% went to the un-leveraged funds (even though Nova is leveraged 1.5-to-1, I considered it un-leveraged for the purposes of this discussion). To me, this is a revealing look at the psychology behind these market timers. They believe that there is a chance for a stiff rally, thus the heavy outflow from the leveraged bearish funds as opposed to the un-leveraged ones. However, they do not have enough faith in higher prices to place their bets on further upside, thus the tepid move into the bullish funds, and the leveraged funds in particular. Normally, when we see a 5% rally in the indexes, we would see a larger absolute move into the bullish funds, and especially into the leveraged funds as these traders try to take maximum advantage of the nascent rally. You may have noticed that the RSI spread indicator is now in bearish territory - that is more a function of funds moving OUT of the bearish funds than assets moving IN to the bullish ones. So is this a good or a bad omen for the market? Common sense (well, common sense to a contrarian, anyway) would suggest that since there is apparently disbelief that this rally could be "the" low according to these timers, then there is a good amount of apprehension, and thus future buying power. That's bullish. On the other hand, as I've pointed out several times, these timers can often be a good non-contrarian indicator in the very short-term. When the market moves higher, but these timers bet AGAINST the rally, then the rally often stalls or fails outright. Conversely, if the market is declining but the Rydex timers add to the bullish funds, then often the market puts in at least a short-term low. Considering this information, the recent activity here, since the timers are essentially betting against further upside, is bearish. My take is that since the asset moves are not directly betting on further downside, but instead not much upside, I think the action here is moderately bullish.

Most of our other indicators are essentially unmoved from Tuesday, or sitting in neutral, so there's not much to go over.

From an intermediate-term perspective (weekly), I do not believe we have seen the ultimate low for this move and lower prices are likely in the coming week(s). If looking for a bear market low, or even an intermediate-term low in general, it would be extremely unusual to see the type of volume and volatility we have recently seen (see Monday's commentary). Many other measures are already in place, so it's not completely unreasonable that a low was put in last week. Not likely, but not unreasonable. In the short-term, I am torn. I believe that we had seen a tremendous amount of short bets placed during the recent decline, and that provides a considerable amount of fuel for a market rally. Not all of it was used up in the past five days. The mini-correction of the past two days has been textbook, and there is no technical reason why another push higher should not occur, possibly exceeding Tuesday's highs. On the other hand, we became quite overbought by Tuesday, and overbought in the context of a severe downtrend just begs to be sold, and sold hard. This would argue that we have seen a short-term high and lower prices are imminent. Due to these conflicting ideas, each of which is perfectly reasonable to me, I will be trading both sides of the market beginning tomorrow, and trading intraday only. This is my particular strategy, and I don't mean to suggest it's what you should do (or even that it's correct), but I believe at this moment, it's the highest-odds, least-risky approach.

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.

While the East Coast storm certainly had an impact on trading volume today, the fact remains that we have not seen the type of volume pattern that virtually always precedes a good intermediate-term market low. It also doesn't help that each day of this mini-rally has come on ever-decreasing volume. I've posted and discussed several times the chart of the deviation of the 10-day NYSE volume from the 200-day. I've shown that - especially during this bear market - high volume has accompanied every major low. Usually, the 10-day average volume was at least 20% above the 200-day average. In comparison, today's 10-day average is 7% BELOW the 200-day average. In an admittedly cursory review of the past 20 years, I could not find one major intermediate-term market low that was accompanied by such pathetic relative volume.

Somewhat related to this is the low daily range in the S&P 500 (cash index, not futures). By daily range, I mean the total number of points from the daily low to the daily high. So if the S&P went from 820 to 832 today, then the range would be 12 points. I do not use true range on the cash S&P since it rarely gaps (if you don't know what that means, it's not really important for this discussion). I've previously pointed out that I look at the 10-day average range expressed as a percentage of current price. If the 10-day average range is 17 points and the S&P is at 850, then the reading would be 2% (=17/850). One of the hallmarks of bear market lows is that they are marked by volatility - an expansion of volume and range. Every major low since 2000 has shown a 10-day S&P range of at least 3% of price. Our current reading is 1.85% - not the makings of a major low.

As an indication of how much buying pressure was used up in the past two days, the Down Pressure indicator on the S&P has dropped to a low reading of 26% - anything under 30% can be considered overbought. The indicator that's posted to the site is actually a three-day moving average of the daily readings, which were 8% on Friday and 7% today. This indicator is constructed by averaging the percentage of total points gained or lost that were down points and the percentage of volume that was down volume for each stock in the S&P 500 and Nasdaq 100. Today's reading of 7% was derived by the fact that only 6% of the point moves today were "down" points, and 8% of the volume in S&P 500 component stocks flowed into stocks that closed lower on the day. Amazingly, if we get anything under 30% tomorrow, this indicator would set a new 8 month overbought record. This is a very broad generalization, but a rally of around 10 S&P points would likely generate a reading around 30% - nothing dramatic is needed. As if we need further proof that we have not seen the type of volatility that usually marks low points, the total number of points gained and lost today in the S&P 500 was 292. Near the July low, the we saw readings over 800 (meaning the average stock in the S&P moved 1.6 points in one day), and the October low saw readings over 600. The highest we've seen in the past two weeks is 330.

With today's rally continuation, the STEM.MR model dipped to the lowest reading it had given since early January. When short-term sentiment becomes so frothy that the model drops below its lower trading band, it is very unusual for the market to be able to muster much more strength, especially considering that we are in a downtrend on virtually every timeframe except intraday. This low model reading is, of course, a reflection of its components, all of which are well overbought except for the VIX, which is close.

As I stated in the intraday note, the highest-odds chance for success with the least amount of risk is identifying overbought conditions in the context of a downtrend (or oversold in an uptrend). We've reached that situation now on a very short time frame, so my focus is on short-side setups entering tomorrow. Once price begins to weaken, I think there's a good chance the sharks will begin to circle and it shouldn't take much to push us lower. Last week, I said that there was a tremendous amount of short fuel to propel us higher in quick, sharp bursts. Now that we've used up some of that fuel, about the only catalyst for higher prices is being depleted. I would have to re-think this strategy if we were able to rally and hold above the 870 level on the S&P.