May 19, 2003

http://www.sentimentrader.com

Wednesday, May 21, 2003 9:35 PM EST

Next week will bring Memorial Day, and as we all know, especially with the Easter action fresh in our memories, holiday seasonality can be a strong and consistent influence. Memorial Day is not a particularly reliable holiday as far as biases go, as the table below demonstrates. I have used data only since 1971, when it was made an official federal holiday, and declared as the last Monday in May. The S&P 500 is used as a proxy for "the market".

	2 Days Prior	Day Prior	Day After	2 Days After	Random
Avg. Return	0.02%	0.15%	0.21%	0.02%	0.03%
% Positive	47%	59%	38%	59%	52%
Maximum	2.95%	1.59%	3.45%	1.88%
Minimum	(1.79%)	(1.21%)	(1.48%)	(1.67%)

We can see that Memorial Day follows a fairly common pattern among holidays - the market is often up the day preceding the holiday, and down the day after. 59% of the days preceding the holiday were up during this time period, as opposed to only 38% of the days immediately after. This hasn't been an especially reliable pattern lately, as four out of the last five years have seen the trading day before Memorial Day end negatively. If the day before Memorial Day ends in the red, then there is a 54% chance that the day after the holiday will also be down. However, if the day before the holiday closes positively, then there is only a 32% chance that the day following the holiday will also close positively. These aren't particularly strong patterns, and the sample size is relatively small, so I certainly wouldn't place trades based on this information. However, it may make sense for short-term traders to at least be a bit more willing to trade strength on Friday and weakness on Tuesday, all else being equal.

Chartcraft's Investor's Intelligence survey was released today with another jump in bullishness. Bulls increased to 56% (from 54.4%) and bears tumbled to 20.9% (from 23.9%). This pushed the bull ratio (bulls / (bulls + bears)) up to 73%, the highest reading since 1992 and in the top 13% of readings in the past 34 years. This is the least amount of bears since 1992 as well. The chart below shows a long-term view of the ratio of bulls to bears. The current ratio is 2.68 (obtained by taking 56% bulls and dividing it by 20.9% bears). The red horizontal line in the bottom portion of the chart highlights our current level.

We can easily observe that we are currently seeing the most extreme reading in 15 years, other than two weeks in 1992. However, historically, this is nowhere near the most extreme - in the 1970's, there were 8 weeks with a ratio over 10 (the graph has been cut off in order to magnify current readings), and in the 1980's there were many weeks where the bullish percentage tripled the bearish percentage.

The Rydex mutual fund timers, who never really embraced this rally to begin with (perhaps one of the reasons it has lasted so long), have turned their backs on it wholeheartedly. Only a couple of days off the high, we're seeing readings similar to what we saw at the March low. The flow out of the bullish funds and into the bearish ones has been enough to bring our shortest-term indicator, the RSI Spread, down to -85, which is the lowest reading since April 11th. The total shift of assets over the past five days has been a little over $500 million (out of the bull funds and into the bear funds), which is about 15% of total assets in the index funds. This is the largest amount of dollars shifted during a five-day period since mid-March. Such extreme readings tell us that there has been a sudden and severe shift of funds to the bearish side, which from a contrarian point of view should be considered positive for the market, as these traders are rarely so right.

The total put/call ratio has stayed at 1.0 for the past three days, which has been getting quite a bit of press. However, I think that headline figure is misleading and not nearly as bullish as it appears. When we take out some of the component measures, the readings become more troublesome. The OEX put/call ratio, for example, has been very high the past three days while the open interest ratio has stayed relatively steady. This suggests that OEX traders, who are certainly better market timers than equity option traders (in general), remain bearish on the market for the time being. The equity put/call ratio has also been elevated the past few days, but again QQQ options are skewing this ratio. QQQ put options have accounted for more than 20% of all equity put options this week, and accounted for more than 30% for two of those three days. When QQQ options are backed out of the equity p/c ratio, it drops dramatically. The 10-day average of the non-QQQ equity p/c ratio remains a very low 0.56.

Our short-term measures overall are neutral. However, with the heavy pessimism from the Rydex timers and relatively positive seasonality, we should see some type of tradable upside attempt yet this week. Such a rally, should it occur, will run smack into some strong negatives that are showing in our longer-term measures. Market breadth remains overbought, and speculation has been running rampant according to several measures. Unless we see more of a correction first, additional upside attempts from here will most likely fail in the intermediate-term. As I've been stating for some time now, unless and until the 955 level on the S&P is taken out, we've only completed one step of a four-step process to suggest we're in a new bull market and overbought readings should be given diminished weight. Until those four steps are complete, the odds are still better selling such an egregious sentiment condition than buying it.

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, May 19, 2003 8:17 PM EST

The selloff today was fairly extreme as far as short-term sentiment goes. Our shortest-term model, STEM.MR, went from 18% on Friday morning all the way up to a very oversold reading of 52% as of today's close. This is the most extreme reading since March 24th, which lead to a quick 17 rise in the S&P before that short-term downtrend resumed.

Our Down Pressure breadth readings, which measure both the volume and points gained or lost each day for each of the components of the S&P 500 and Nasdaq 100 indexes, will almost certainly enter oversold territory tomorrow if we have another down day, even a moderate one. If one concludes that the shorter-term (30-minute) trend is up, which I do, then an oversold Down Pressure reading would give a good indication that we will likely see something of a bounce sooner rather than later.

The put/call readings looked fairly extreme today but again they were skewed by trading in QQQ options. The QQQ put/call ratio came in at 3.0, and QQQ put options accounted for over 30% of all equity put options. The traditional equity put/call ratio is showing a very oversold reading of 0.90, but when we back out the QQQ options, it comes all the way down to 0.68 - barely on the high end of neutral. The 10-day average of the non-QQQ equity p/c ratio is still a very low 0.55, which continues to suggest that equity option traders are too optimistic, and that normally spells trouble for the broader market.

In the OEX (S&P 100) options complex, the open interest ratio remains steeply skewed, even after last week's expiration. It is fairly unusual to see the open interest ratio so stable after an expiration, as put and call open interest often jumps sporadically when near-term contracts expire. That the OEX ratio remained so stable this time around shows that these traders did not open all of those put contracts in the May series - they were longer-dated options. The current open interest ratio is the most skewed the day after an expiration in almost six years, which again tells us that OEX traders are pessimistic on the outlook for the 100 largest companies in the S&P 500. Since these traders tend to be right more than they are wrong, this is a bearish indication for the market.

In data released over the weekend, the Commitments of Traders data was essentially unchanged. The futures traders in both the full contract and the e-mini were apparently content to sit on their existing positions. The story here has not changed for about two months now - the full contract has recently given us the most bullish indications since the data began in the 1980's, and has backed off of that only slightly in the past couple of weeks. In the e-mini, the situation appears to be the mirror-image, as commercial traders remain heavily short while small speculators are very net long. This should be bearish, but the contract has not given consistent enough readings in order to read too much into it.

For the period ending May 2nd (the most recent available), specialists on the NYSE picked up their shorting activity enough to push the Specialist Short Ratio to 38%. This is right on target, and I would expect to see this week's reading somewhere closer to 40%. This is the area that has coincided with recent market peaks, though it is still relatively low historically. I showed a very long-tem chart last week showing that this data lends some credence to the "new bull market" theories.

The most current sentiment survey results continue to exacerbate the overbought extremes they reached last week. The recent changes were enough to push the AIM model under 40% and into extreme negative territory for the first time since January 2002, and January 2000 before that. The model is now more than 30% below its 12-month average, a reading matched only twice since 1991 (March 1995 and June 1997 were the other two instances, however overbought readings in a longer-term uptrend are not particularly effective). Since that reading in 1997, there have been 5 other readings of the model being stretched 20% or more below its 12-month average, each of which turned out to be excellent intermediate-term sell signals.

If you look at the "model glance" on the subscriber's home page of the site, you can see that the models are making almost a perfect diagonal line, from extremely positive in the short-term, to extremely negative in the longer-term. This coincides with our indicators, which are suggesting the same thing - much more of a decline should be met with some buying pressure, but that rally will more likely fail than not.

- Jason Goepfert

Disclosure: no positions