sentimenTrader.com

All rights reserved.

Sunday, June 2, 2002

I stated last week that I felt we didn't have enough pieces in place from a sentiment perspective to expect a change in trend.  And, with the performance last week, I think it's safe to say we didn't get one.  If you've been reading the daily commentaries, you should know that with any further selling, I've been becoming more and more bullish as many of our sentiment indicators are flashing oversold.  Beginning Friday, I stated that many pieces are beginning to fall into place to expect a short- to intermediate-term bounce.  Friday was NOT a positive day for the market in any way, shape or form, as we exceeded the narrow range day I talked about on Thursday but couldn't hold the gain.  About the only thing to hold on to was the light volume, but that has been the case for quite some time now.  The action on Friday does dampen the short-term bullish case, but not the intermediate-term in the least.  I still feel the odds favor a rally within the next month, which the models and indicators would suggest to be about 5-8%.  I detest giving market projections and prefer to just go with the flow, but I wanted to throw that out there as something to consider.

And now let's look at our indicators...

BULLISH

NYSE MEMBERS REPORT - This set of indicators will now be posted weekly to the site beginning today, under the "Indicators" section.  To kick things off, I also posted a study I did on the current reading of the Specialist Short Ratio, which has been firmly in bullish territory for the past several weeks.  You can view the results and read the conclusions in the Research section.  In essence, this indicator is suggesting that the recent negativity has been excessive, as determined by a high percentage of the short sales being accounting for by the public.  The specialists on the NYSE rarely account for less than 40% of total short sales in any giving reporting period, but they did several weeks ago and continue to hang around that level now.  When we get to such a point of public shorting, the market has a strong tendency to outperform up to six months from the reading.  There is a two-week lag in the data, but the market has not moved much since those numbers were reported.

COMPOSITE MODEL - I mentioned this model in the daily commentary on Thursday and suggested that you peruse the study of the current reading posted to the Research area.  Once again, the study suggests that when we have readings such as the current one, the market has a strong tendency to outperform over the intermediate term.  The model dipped a bit on Friday, thus making Thursday's reading a "unique" occurrence, so we can now place more weight on the second set of data that was referenced in that study.  The data is overwhelmingly bullish, with odds that should make any trader salivate.  For instance, after 30 days the average return was 8.6%, with 90% "winners", and an average gain of 10.4% versus an average loss of 5.4%.  One other thing to note is that we had a reverse divergence in this model.  In early May, the model gave a reading of 63% with the S&P around the 1050 level.  Then on Thursday the model gave a higher reading of 66% while the S&P closed at 1064.  So that tells me that even though our broadest gauge of sentiment conditions was saying market participants were more pessimistic on Thursday, price was able to hold above the recent lows. Now THAT'S bullish.  A key test will come if we do indeed head lower from here and meet or exceed those early May lows.

SENTIMENT SURVEYS - The surveys dropped a bit of their bullishness in the last survey period, except of course for Investor's Intelligence.  The main reason I am suggesting these surveys are now bullish are two measures I decided to post to the "Sentiment Surveys" section of the site.  You will now notice two new charts, one titled "Market Vane Relative Value" and the other "AAII Bullish % Moving Average" (catchy, heh?).  The Market Vane chart shows the most current reading divided by the 52-week moving average.  This is simply to account for any secular changes in the survey itself.  This survey has a tendency to migrate from one trading range to another, where a reading of, say 25% bulls will be quite extreme one year, but not as extreme several years later.  This could be said with most of the surveys, but I've found it particularly true with this one.  So this ratio is simply meant to eliminate any of those potential complications.  The reading we are now getting is .61, which places it in the bottom 4% of all readings over the past 20 YEARS.  The new AAII chart is just the 4 week moving average of the bullish % of that survey (bulls / (bulls + bears)).  That ratio is now giving a reading in the bottom 13% of all readings over the past 15 years.  The current reading of that ratio is lower than it was at the March and September 2001 market bottoms.  Those two in combination show quite a bearish opinion of the market by futures traders and individual investors, two segments of the public known to be most wrong most often.

CBOE RATIOS - In the Research area, I have posted a study I did this weekend on the 10-day moving average of the put/call ratio.  It hit a high of .83 on Thursday before dropping to .81 on Friday.  The results of the study are eye-opening, and exceedingly bullish.  For ease of viewing, I also included a chart of the last 7 years, with situations similar to the current one circled.  Since 1995, buying the OEX when the 10-day p/c moving average hit .83 or higher and reversed led to an average return of 6.0% with a "win" rate of 94% after 30 days.  And lest you think this was a bull-market aberration, the greatest returns occurred AFTER the bubble burst in 2000.  The CBOE put/call open interest ratio is also very bullish, as you can see from the chart posted to the "Indicators" section.  This is one of my all-time favorite indicators, because 1. It works and 2. I know other people follow it, but there aren't many!

TOPM MODEL - In this week's issue of Barron's (www.barrons.com), Kopin Tan states in the Striking Price column that "The clamor for put protection created a skew, with traders noticing richer put prices in many stocks but without a corresponding rise in call premiums -- a sign that investors were more willing to pay for downside insurance than for upside bets."  I happen to disagree with tone of the article, but that quote is true, and it is exactly the situation I track daily with this model.  I included a chart of the model with last week's commentary, but it has gotten significantly more bullish this week, for exactly the reason stated in Barron's above.  The current model reading is .48 and the 5-day moving average is .53, both of which are very bullish.  As traders scramble to buy puts and thus bid prices up, I happen to think that's bullish and not bearish, as this group of traders is historically very wrong (as the CBOE ratios discussion above illustrates).

BREADTH RATIOS - We have somewhat of a mixed market here, as the Nasdaq is considerably more oversold on a breadth basis than is the S&P, but both could be considered at least moderately bullish.  The advance/decline ratio will begin dropping some large negative readings next week, so we're close to being as oversold as we're going to get unless we really get pounded down next week.  The NH/NL ratio also became quite oversold before rebounding a bit on Friday.  The cumulative TICK readings on the NYSE are neutral, but bullish on the Nasdaq.  The daily reading is now negative, and as you can see from the chart in the "Indicators" section of the site, that has corresponded to at least short-term bounces over the past year.  The intraday reading is also oversold, although less so than it was a couple of days before late last week's little rally began.

BEARISH

COT - No question here, this is bearish.  The commercials added 8,000 long positions but increased their shorts by 13,000, for a net short addition of 5,000 contracts.  In addition, the small specs decreased their shorts by 1,000 and increased their longs by 7,000 for a net long addition of 8,000 contracts.  The small specs have now built up their largest net long position in the past decade, which is NOT a good sign.  However, I think the best way to look at the data is through the stochastics, and although they are not maximum bearish (except for the small specs), they are bearish nonetheless.  The Futures Balance Matrix (found in the "Indicators" section under "COT") gives a good look at this data, and you can quickly see that we are not in a position where strong intermediate- to long-term rallies have begun.

NEUTRAL

NEWS ITEMS - About the only news reaction of any consequence this week was the economic numbers on Friday and news that the India-Pakistan confrontation may turn nuclear.  The reaction to the economic news was flat-out bearish, no doubt about it.  For the market not to gain significantly on the good news that was in those numbers detracts considerably from the bullish case.  However, even with the news of a possible nuclear war brewing, the market did not forcibly sell off, with many indexes finishing up on the day.  During the weekend, Pakistan's president played down the possibility of using nuclear weapons, which may help to alleviate some of those fears next week.

VIX and VXN- The VIX spent the end of last week near 5% above its 10-day moving average, which is on the bullish end of neutral.  I mentioned on Thursday that it reached 12% above the average before reversing, which is a bullish indication.  Nothing happened on Friday to change that.  The VXN didn't do much at all last week.  Not much to say here, other than that these volatility measures are neutral to ever-so-slightly bullish.

TRIN - The TRIN 10-day moving average is starting to sneak back up towards oversold.  We will be dropping two overbought readings early next week, so we could see an oversold reading here if we sell off much more, particularly on the Nasdaq.  At this time, however, we are firmly in neutral territory.

AIM - Although I consider the sentiment surveys to be generally bullish, it was not enough to push this model into BUY territory.  We continue to hang just under our signal line, with a reading essentially unchanged from last week.  That stubborn II survey will just not budge, but if it does it will have a profound impact on this model.  Still, if you look at the May 12 weekly commentary, you will see that a reading around this area bodes well for the market, particularly in the intermediate- to long-term.

STEM.MR - This model did a fairly good job of alerting us to the extreme points of short-term sentiment this week.  The rally off of the lows on Thursday and into Friday brought the component measures back to neutral territory, and almost reached short-term overbought near the highs on Friday.  At this point, the model and all components are neutral.

STEM - Without Friday's rally, we could have been near a BUY signal here, but the upside movement brought the model back closer to neutral.  We will be dropping some high readings beginning next week, so it will be difficult to generate a BUY now unless we really sell off significantly.  The model is on the upper end of neutral, so it's still a positive for the market, but we can't consider it all-out bullish.

As of this weekend we have the greatest concentration of bullish indications since early May and before that late February/early March.  If we go down much further, I believe we will also see the VIX and TRIN become bullish, along with the STEM.MR and possibly the STEM models.  I will follow price action wherever it takes us, but if we see those additional indicators flip over to bullish, I will curtail my short activities greatly.  I could see a return down to challenge the recent lows, and maybe even a slight break, but I believe we will be at least modestly higher than where we are now within 30 days.

 - Jason Goepfert