sentimenTrader.com

All rights reserved.

Thursday, December 5, 2002  8:45 PM EST

I mentioned the Down Pressure indicator yesterday, and that it would take a reading of around 80% today to push the three-day average to the same level.  We did indeed hit that threshold today, and as you can see from the current chart below, past readings of 80% or higher have generally been a short-term positive for the S&P.

The action is more pronounced in the NDX, as you can see below:

Down Pressure has now reached 86% in the NDX, which as you can see from the green arrows on the chart, has been reached four other times in the past five months.  Each lead to at least a 70 point rise in the Nasdaq 100 over the ensuing week.

Those of you who receive the intraday updates know that I calculate an intraday price oscillator on the S&P and NDX, which has proven to be quite effective, particularly during trading ranges (I discussed the calculation in the November 14th commentary - click here to read).  I also follow the same indicator on a daily basis for both indexes.  As of today, the indicators for both indexes are deeply oversold, particularly so on the NDX which is showing a reading of 18%.  In fact, only three other times in the past FIVE YEARS has the NDX given such an oversold reading.  The charts below show the results of the other three instances:

The occurrence on December 18th lead to another low reading the following day, which in turn lead to a large gap down open.  That sequence then lead to a sharp rally over the next week.  With only three samples over a five-year period, we can't glean any significant conclusions from such a reading, but what we do have suggests that a tradable short-term low is likely near.

The STEM model has gone into strong positive territory as of the close, as it is now poking out of its upper trading band.  The STEM.MR model is not far behind, sitting right on its upper band at the close.  Both are very positive developments if one is looking for some short-term upside.

I showed a longer-term chart of the VIX.MR indicator in the intraday note, and I'm reprising it here.  This six-month chart shows that we have now reached the highest absolute level of any time during this period, which encompasses both the July and October lows.  Remember, this indicator is a one-year rank of deviations of the VIX from its 10-day average.

You can see from the chart that each time we approached or poked out of the upper trading band, the market had at least a short-term rally.  When it exceeded the band by a significant amount, as it is now, the rallies tended to be larger.

All of this could be for naught depending on what the economic numbers are like tomorrow morning.  One thing I want to impress upon you is the importance of sentiment going into important news events.  When we are at an extreme in sentiment, it takes a significant catalyst to continue pushing prices in the direction of the trend, but only a mild surprise to change it.  Let's take Monday for example.  Sentiment was overbought going into the ISM number, so it would have taken a large positive surprise to keep things going to the upside since expectations were already high.  But a number that was in-line, or in this case slightly lower than expectations, was enough to change the trend and send us lower.  You will find that this is a very consistent pattern.

In any event, I am expecting a tradable rally within the next day or two.  Since the econ numbers come out before regular trading hours, if they are negative and create a gap down open, that's something I would look to buy once the selling appears to abate.  If we go the other way, I wouldn't be excited about buying a gap up, but there's some room to go on the upside before shorts become highs odds once again.  Obviously, I believe that in the short-term (less than 5 days), the best risk/reward lays with the long side.  However, this is with the knowledge that the intermediate-term sentiment condition has deteriorated considerably and can now be considered negative.  What that tells me is that it's OK to go long for a few days (if it fits with your methodology), but keep tight trailing stops on any open long positions.

Wednesday, December 4, 2002  9:00 PM EST

With 51.1% bulls and 25% bears in the most recently released Investor's Intelligence sentiment survey, the gap between the two groups is becoming pronounced.  During this bear market, 16 other weeks have shown such a wide gap or wider.  The chart below shows the average performance of the S&P 500 a certain number of weeks after each reading, with the percentage of time the S&P was positive during that time.

We can see that after 6 weeks, for example, the average return in the S&P was -5.94%, with a probability of it being positive of 6%.  With only 16 events the sample size is quite small, although I believe it's still instructive to see what has happened during recent conditions.  If we expand our time frame a bit and go out to the past decade, things get better.

Although the results over the past decade show significant improvement, even these figures are considerably more bearish than random periods taken during the same decade.  In fact, the figures in the chart immediately above show an average return about 2% below random and a probability of being positive about 20% below random.  So even when we look at a period which encompasses one of the greatest bull markets in history, when the bull/bear spread in the II survey becomes as distorted as it is now, the S&P has a much less than even chance of being positive up until 3 months out.

Certainly, when coupled with many of our other intermediate-term indicators (such as the breadth ratios, the Rydex ratios, put/call moving averages and other sentiment surveys), the outlook several weeks out must be considered negative, or at the very least definitely not positive.  About the only saving grace I can find is that we'll be running into some very positive seasonality towards the end of the month and we always have the possibility that our current extremes can, well, become even more extreme.

In the shorter-term (less than 5 days), there are a few things for the bulls to look to.  Although the longer-term indicators I create with the Rydex ratios continue to be negative, the short-term RSI spread is now at -60% (meaning asset momentum has shifted significantly to the bearish funds over the past three days), a threshold which has proven to be consistently positive for the market over the ensuing few days. 

The proprietary Down Pressure indicator (discussed in the November 11th commentary - click here to read) is currently giving a reading of 70%, which is also on the threshold of being oversold and bullish.  A reading above 80% could be considered extreme, and it would take a reading of 80% or more tomorrow in order to get the three-day average to that level.  For sake of comparison, today's reading was 78% and yesterday's was 81%.  On November 11th, the indicator reached 88%, and that was quickly followed by a 60 point rally in the S&P 500.

Wall Street analysts have been relatively quiet with their downgrades lately, as downgrades as a percent of total ratings changes is down to 57% on a three-day moving average basis.  I've discussed the possible future efficacy of this indicator several times, as the continued governmental pressure on these analysts to "do good" by their clientele my prompt knowledgeable market timing by this group to an extent never before seen.  The chart below shows that the past few times the downgrade percentage has reached such a low level, the market typically performed relatively well over the next week or so.  I wouldn't place a whole lot of weight on this development just yet, but like I've said before, it's worth monitoring.

So there are some positives from a sentiment perspective, but virtually all of them are short-term in nature.  This would lend credence to the possibility of another rally attempt followed by a severe, multi-week decline, which I see being discussed heavily as the most likely course of action.  The ultimate contrarian would say that since it's being discussed so much, it's less likely to happen, so you almost have to be contrarian about being contrarian.  I prefer to not get so cute and try to outguess the guessers, so I'll go along with what's quickly becoming a crowd and suggest that it's likely we'll get a rally attempt soon, followed by a more severe correction than we've seen since the October low.  If we get a heavy down day tomorrow, I think that would set up a relatively high-odds trade from the long side lasting a couple of days, but let's discuss that tomorrow if it happens.

Tuesday, December 3, 2002  9:20 PM EST

Not much new to go over tonight from a sentiment perspective.

About the only thing of note is as I mentioned in the model change alert and intraday note - our shortest-term sentiment measures are becoming quite stretched.  The STEM.MR model has now busted through its upper standard deviation band for the first time since the low in early October.  Obviously, each time the model exceeds its band does not correspond to such a violent market move, but a great majority of the time the S&P stages a rally of at least 10-20 points within a day of such an occurrence.  It's not just one component that is having a large impact on the model, as each of them are either in oversold territory or near to it.  When we have such a confluence, the signals tend to be more reliable than when one particular indicator is having too great an influence.  This would have me looking to try the long side - for a short-term trade only - if price begins to firm or even if we have a large gap down tomorrow morning.  Again, I would be looking to capture 10-20 S&P points from such an attempt.

Longer-term, really nothing has changed with today's action.  Most of our intermediate-term indicators are in the slow process of working off their overbought conditions, and it will likely take weeks before they will cycle back around to a more constructive position.  Until then, I continue to urge caution on the long side for those looking for multi-week positions.

Monday, December 2, 2002  9:39 PM EST

The new month finds us with sentiment beginning to enter a territory where caution is warranted.

The sentiment surveys have gradually increased to the point where the AAII bullish ratio, II bullish ratio and Consensus bullish percentage are all challenging their upper standard deviation bands.  These bands are useful for putting current readings into the context of recent history, and have proven themselves as a reliable guide for showing excessive pessimism or optimism.  Shown as a composite via the AIM model, the surveys are reflecting a generally-held opinion of higher prices, or at least not severely lower ones.  The AIM model is now only 4% above its lower standard deviation band, which is the lowest since January of this year.  In fact, since the bear market began, every instance of the model reaching this level has resulted in a market high immediately after.  During the bull-market run prior to this bear, such a high level of optimism didn't deter the market for more than a week or two, so one needs to make the distinction between how this indicator performs in bull or bear markets, and what our current situation is.  A simple trendline drawn from the peak in 2000 on a broad market index such as the NYSE Composite Index clearly shows the trend is still down, so I believe the model can be compared favorably to September 2000, May 2001 and January 2002 - all corresponding to intermediate-term market highs.

The most recent Commitments of Traders report (released today due to the Thanksgiving holiday) showed commercial traders in the S&P 500 futures increasing their net short position by 7,600 contracts while the small speculators increased their net long position by 8,300 contracts.  This has served to push the three-month stochastic reading for the commercials to the lowest level since the peak in March, at the same time the stochastic for the small specs has risen to the largest reading since March as well.  In essence, the extended trading range since late October has given the commercials time to increasingly hedge their portfolios while the small traders try to anticipate a breakout to the upside.  While neither position should be considered extremely bearish, the positive indication these positions had given since early October is gone.

Our intermediate-term breadth measurements are close to becoming overbought in unison.  The 10-day advance/decline line, 10-day up volume ratio, 10-day TRIN (NYSE and Nasdaq) and daily cumulative TICKS are all either in overbought territory or knocking on the door.  It is extremely rare for the broader market to make significant headway when these indicators are in confluence, especially so in the context of a bear market.  The 10-day new high / new low ratio still has plenty of room to go before reaching overbought, and it is just now approaching the area reached shortly after the August high.

If we look at a couple of indicators that reflect actual monetary transactions, the put/call and Rydex ratios, the intermediate-term situation looks even more dire.  The 10-day averages of the equity and total put/call ratios, commonly believe to reflect option commitments of poorly capitalized (and informed) smaller traders, have poked below their lower standard deviation bands.  As these bands contain 80% of readings, a move outside of the bands should be considered noteworthy.  The relatively heavy flow of volume into call contracts as opposed to puts, which now can be considered extreme and persistent, is somewhat rare and almost always precedes a market decline shortly after.  The Rydex ratios are showing a similarly persistent trend of money flow into bullish-oriented funds as opposed to bearish.  The three-month stochastic of the bullish ratio (which shows the momentum of assets flowing into bullish funds compared to bearish funds, weighted by the leverage each fund employs) is currently at an all-time high.  Since the leveraged Dynamic funds were only created in 2000, the history of this indicator is short, but has been effective in the amount of time we've been able to monitor it.  The fact that we're seeing such a flow into the bullish funds (and out of the bearish) is a sign of optimism that is rarely rewarded.

On a shorter time frame, the STEM.MR model is quite high (that's a positive), and the price oscillators and intraday cumulative TICKS are neutral to slightly oversold. 

The heavy decline off the gap up opening this morning has served to push some of our shorter-term measures into oversold territory, suggesting we've rested enough for another push higher.  However, considering the condition of a majority of our intermediate-term indicators and models, any move higher should be muted relative to the risk of lower prices.  Since the risk/reward for positions held at least several weeks is now titled to the short side, further rallies should be used to exit or aggressively hedge long positions, and to scout for selected shorts. 

- Jason Goepfert