May 26, 2002

sentimenTrader.com

Sunday, May 26, 2002

From the commentary last week, there was 1 item in the bullish column, 4 in the bearish column and 8 in the neutral column. From that, a logical conclusion would be that sentiment was then neutral to bearish for the market, which turned out to be pretty accurate. If you've seen the addition to the subscriber's home page, titled "Daily Overview", there's a new section called "Indicator Bias". This will be updated daily and hopefully it will serve as a quick reference for those of you looking for a take on the current sentiment situation on three major time frames. Although by nature subjective, I will be objective as possible when updating these color bars.

We just completed the lowest-volume week since late December, and with that lack of volume was a lack of change in much of the sentiment condition. We are basically in the same spot this week as we were last week.

Let's take a look at those indicators one by one...

BULLISH

SPECIALIST SHORT RATIO - This long-term indicator continues to be bullish. At 40%, it is on its third week in what I consider to be positive territory. Within two weeks, I will have an extensive page up in the Indicator section dealing with this indicator and possibly some of its derivations. It will help you put some current readings into an historical context.

STEM.MR - Our shortest-term model is now on the cusp of being bullish. The action on Friday made the model creep up all day, finally finishing at its highs for the day near the close. With any more of a market decline early next week, we will most likely get a BUY signal here, but I'm not sure how confident I would be recommending a long bias in the current environment. I will issue whatever the model gives, but I do have leeway in the ratings. And I can't really imagine giving a three-star rated BUY while we're in this quagmire.

BEARISH

COT - Although the commercials added about 10,000 contracts long, they also added about 5,000 contracts short, for a reduction in their net short position of around 5,000 contracts. The small specs didn't make much of a change, except to become a bit more long. They are now the net longest they have been since the decades-long high this past March. The situation now isn't as dire as it was then, as the commercials have somewhat reduced their short exposure, but it is still a long-term negative. The Futures Balance Matrix shows a 24% long-side commitment by the commercials as opposed to a 76% commitment by the small specs. Although that is a positive change from last week, my studies show that the S&P 500 underperforms in every time period up to six months following such a reading.

TRIN - Even with Friday's somewhat high reading, the 10-day average is still on the low end of its range. I said last week that even with a market drop, we could still see this indicator go into the bearish column since we were dropping so many high readings. On Friday, we dropped 2.43 and added 1.58, and the moving average dropped to .96. Although that is not extreme and there is certainly room for it to drop further, it is suggesting we currently have overbought conditions. The last couple of times we saw this area was early March and late December/early January which preceded substantial market weakness. Many traders feel that the TRIN is ineffective, especially on the overbought side, but what I've found is that it goes through periods of high and low trading ranges. The current trading range indicates overbought conditions when the 10-day moving average is around .90, which is close to where we are now. Although I don't believe in changing one's parameters to constantly fit the market, I do believe in remaining flexible enough to understand that sometimes the underlying dynamics of the markets change, and that causes changes in the indicators based on those markets. That doesn't necessitate abandoning effective filters, but it does mean we need to keep our eyes open to the possibility of these indicators entering new trading ranges.

NEWS ITEMS - It was an extremely slow news week, particularly for economic news, but that will change soon. A couple of things to note is that even though Home Depot came out with numbers that were generally higher than expected, the stock ended down 12% on the week after gapping up a dollar on Monday. And the bigger story in my opinion - Merrill Lynch - is even more bearish. After gapping up 5% Tuesday on news of their analyst conflict-of-interest settlement with the NY AG, the stock actually finished down about 1% on the week. This stock has gone nowhere but down since January, with the excuse that this potential litigation was hanging over it. Now that the uncertainty is over, traders are finding new excuses to not buy - such as it opens Merrill up to new investor suits. This helped push the XBD (Securities Broker/Dealer Index) down 4% on the week. It will not be a positive for the market if this particular industry sector cannot rally. On the economic front, the Leading Indicators numbers came out lower than expected on Monday, and the market responded accordingly (by going down). On Thursday, Durable Goods came in quite a bit higher than expected and the market couldn't do anything with it. There was a nice late-day rally, but it was given back on Friday. The GDP number on Friday was just a touch lower than expected. There's nothing for the bulls to hang their hats on here.

BREADTH RATIOS - The cumulative TICK indicators are quite troublesome, as should be evident from the study posted to the Research area. It states that 100% of the time over the past four years the market has been lower 30 days from a reading we had this week. The average decline is still several percent more than what we've seen thus far. I said last week that the advance/decline ratio moving average was on the overbought end of its range and could become extremely overbought with any kind of a rally. Since we didn't get much of a rally, we will now be dropping some large positive readings this coming week while retaining some large negative readings. This could push the moving average into bullish oversold territory if the market continues to drop.

NEUTRAL

SENTIMENT SURVEYS - We have a somewhat mixed picture here, as the Market Vane % bulls dropped while the Consensus % bulls increased. The AAII survey became significantly more bearish (bullish to us) while the II survey became more bullish (bearish to us). I don't think there's much we can read into this short-term. The II survey bullish ratio (bullish/(bullish + bearish)) has remained above 60% for 27 straight weeks now (with the exception of a one-week dip to 59.5% in early March). That takes us back to November 9, 2001 with the S&P at 1120 and the NDX at 1514. So with a 4% decline in the S&P and a 17% decline in the NDX, their bullish enthusiasm has not budged. The most recent occurrence of this string of bullishness was 26 weeks above 60% from November 1999 - May 2000, during which time the market chopped around. The end of that string coincided nicely with the start of this bear market.

AIM - Still neutral here, although the model rose on the sharp decline in bullishness from the AAII survey. Since the weighting isn't as great for this survey, it wasn't enough to push the model into BUY territory. With the II survey where it is, it would actually take a reading from AAII of almost 0% bulls to get this model into the BUY zone.

CBOE RATIOS - No change from last week - fairly neutral, although with a slight bullish bias. We continue to see moderate readings on all of these measures, with no extremes so far. Since no particularly aggressive call buying has been evident, and we actually saw a fairly heavy spout of put buying, the moving averages continue to hang around the upper (bullish) end of neutral. The open interest figures and the OI - P/C spread remain neutral to somewhat bullish.

TOPM - We've seen a steady drop in this proprietary index all week, driving the 5-day moving average down. This tells us that puts are beginning to be bid rather aggressively, which from a contrarian's standpoint is bullish (once it reaches an extreme). I will get a page up for this model or add it to one of the existing indicator pages, but in the meantime, here is a chart of where we stand now (click on the thumbnail to see a larger chart):

As the index and the moving average drop, it suggests puts are catching a bid. As they rise, it suggests momentum is moving back to calls. Of course, the more options traders as a group want to own puts, the more likely it is they will not need them. We have not reached a bullish extreme yet, but we'll be dropping a couple of high figures next week, so should put buying remain aggressive, we could dip into bullish territory here.

VIX and VXN - This is from last week's commentary:

"This tells us that if the VIX does NOT continue lower, but instead reverses up from here, probability would suggest that the S&P may be 40 - 50 points lower several days from now."

The close on Friday was around 1106, and we dropped to 1075 by Wednesday, which was a 30 point drop as the VIX rose from those levels. That action did coincide quite nicely with the study - you can see those results on the Research page. As of now, we are fairly neutral on both the VIX and VXN, although I would have to say they are somewhat bearish. The fact that they remained below their 10-day averages during a down market doesn't get me real bullish. As you may know by now, I don't place a great deal of weight on the absolute levels (i.e. the VIX is around 20, which is too low) but instead prefer to look at relative readings (i.e. 5% below the 10-day moving average). Even though the absolute levels are low, the relative readings are moderate to only very slightly bearish.

STEM - After becoming bearish last Friday, this week's decline has brought the model back to neutral. It would take several days of persistent pessimism to get anything close to a BUY indication here. The slow line will be dropping some low figures on Tuesday, but the slow line - the signal line - will not be dropping those figures until later in the week. So a large rise in the signal line would be difficult to imagine unless a market drop was either very severe or took several days.

COMPOSITE MODEL - After coming close to bullish at the early May bottom, this model is now back down to where it was in early April. With the new weekly figures we have this week, the most likely direction for this model next week will be up (bullish) if we get continued selling in the markets. It would not be inconceivable to see a BUY indication here, but it would take a rather severe market beating to accomplish that.

This week is fairly balanced between bullish and bearish indications. When we have situations such as this, I always follow the trend. Sentiment is useful in the extremes, but not so much during the middle (depending on your time frame of course). That's why I always make it clear that I look for extremes, since that's what it takes to make a new trend. In the meantime, stay with the flow and don't look for things that aren't there. At this time, we don't have enough of a confluence to say with any confidence that there may be a trend change.

- Jason Goepfert