sentimenTrader.com
All rights reserved.
Sunday, January 19, 2003
In the December 29th commentary, I said that a rally of 5%-8% would likely
set us up nicely for a high-odds shorting opportunity. We rallied
about 7% off those prices in the S&P, and that did indeed set up a couple
of high-odds shorting opportunities this week. I've been stressing
over the past two weeks that further strength should be viewed as an
opportunity to hedge or sell intermediate-term long positions, or initiate
short positions. Hopefully most of you were able to see the
opportunity to save or make money this week instead of be blindsided by
poor reactions to decent news. When expectations are already so high
going into unknown economic and corporate news, it takes a heck of an
upside catalyst to keep the party humming. Mediocre or, worse yet,
disappointing news has a tendency to set off action like we saw late this
week. If you take one thing away from the study of sentiment, I hope
this is it...fundamental news reports or even widely-watched technical
levels are not necessarily important in the short- to intermediate-term.
What affects price action the most is investor perception going into those
news reports or technical levels, particularly when they are at an extreme.
Find a system for
gauging these perceptions, and use it to your advantage. Be the
shepherd, not the sheep - you'll get sheared less often.
____________________________________
Indicator: COMMITMENTS OF TRADERS
Status: BULLISH
Comment: In a very unusual development, the commercial traders continue to reduce their net short position while the small speculators reduce their net longs, in the midst of generally rising prices. Typically, commercial traders will scale out of their short positions during market declines, as they buy contracts from the small speculators who are busy selling. Over the past five weeks (the reporting weeks for the COT end on Tuesdays), the small specs have reduced their net long position by over 52,000 contracts, in spite of the S&P 500 being up more than 3% during that time. During the bear market, when the S&P was up more than 3% during a five-week span, the median change in small spec net positions was +4,700 contracts. A change of -52,000 contracts is more than 3 standard deviations from the mean value of +2,600 contracts, so what we're seeing now is unusual. Conversely, the median change in commercial positions was -8,500 contracts, while we have seen a change of +21,000 contracts during the last five weeks. Our current figure is nearly two standard deviations from the average. If we look at these changes together, there has been a 73,000 contract swing in a positive direction, which is by far the biggest swing during an up market (i.e. up more than 3% over a five-week span) during the entire life of this data. The closest comparison would be mid-October 2001, when we saw a +50,000 swing during an up market. That, of course, lead to another leg higher into year-end. About the only other comparison would be early May 2001, when there was a +35,000 contract swing. That instance lead to a quick 6% rise in the S&P which failed just as quickly.
Bottom Line: The recent action of commercial traders liquidating their net short position at the same time the small specs are liquidating their net longs, all during an up market, is extremely unusual (especially to this degree). We only have two other recent reasonable comparisons, both of which lead to continued upside which ultimately failed. These position changes suggest that we may see another upside attempt fairly soon.
____________________________________
Indicator: VOLATILITY MEASUREMENTS
Status: NEUTRAL
Comment: After reaching overbought early last week (stretched from 10-day moving average with low RSI), the VIX has gradually moved higher since then, and is now neutral. Curiously, the VXN declined substantially on the day even though the NDX was down more than 4% (the VXN tracks the implied volatility on NDX (Nasdaq 100) options). Due to the nature of how the VIX and VXN are calculated, it is rare for these volatility measures to move in the same direction as their underlying index. Typically, when the NDX declines, the VXN will rise and vice versa. In fact, since 1995, when the NDX closed down on the day, the VXN rose 92% of the time. But when the NDX closed down more than 3%, the VXN rose 99.3% of the time. There have been only 14 occurrences of the VXN falling when the NDX fell more than 3%, the last one being on December 13th. Similar to most of the past instances, the market staged a two-day rally before continuing downward. Of course, expiration may have had an affect on the VXN, but I still believe it urges caution. The other volatility measures I have mentioned recently (average TICK range, average S&P range and NYSE volume patterns) are near the lower end of their ranges.
Bottom Line: I don't think there's a whole lot to be read into the volatility measures currently. They are just beginning to recover from readings that said complacency was rampant, and it will take time and/or lower prices to offset that.
____________________________________
Indicator: TRIN
Status: NEUTRAL
Comment: The 10-day NYSE TRIN reached a low of 0.82 on Tuesday, while 10-day Nasdaq TRIN hit 0.86. Both of these readings were below their respective one-year 1.5 standard deviation bands, and they once again gave nearly perfect sell signals. When these indicators have approached their lower extremes during the bear market, an ensuing sell-off has been a very high-odds affair (exactly opposite to approaches to the upper end of their trading ranges during the bull market). The three-day decline to end last week brought with it heightened TRIN readings, so our 10-day averages have now pulled out of overbought territory.
Bottom Line: The exceedingly low TRIN readings we had been seeing over the past two weeks are just now beginning to be replaced with higher ones, so it will take at least another week before these indicators are likely to cycle back to oversold. Once again, it will take either time and/or lower prices to work off that exhaustive thrust of buying power.
____________________________________
Indicator: STEM.MR MODEL
Status: NEUTRAL
Comment: Our shortest-term model has spent the better part of the last two weeks working off the extremely overbought condition it reached with the two explosive upside days off of the December 31st low. The main property of this model is that it measures deviations from a mean for four sentiment indicators. Since the mean of the indicators became low (overbought) with the early January rally, the past couple of days of high readings have pushed the deviations to relatively high levels, thus giving the model a reading close to oversold. In fact, the model almost reached its upper (oversold) standard deviation band during Friday's selloff but didn't quite make it there. If we see a continuation of Friday's selling pressure early next week, this model will quite likely move into strong positive territory, suggesting we look for a quick rebound.
Bottom Line: Continued selling pressure early next week will most likely prompt a positive indication out of this model, suggesting the selling pressure is becoming exhaustive and indicative of an upside retracement soon.
____________________________________
Indicator: SEASONALITY
Status: NEUTRAL
Comment: I don't see anything particularly unusual about the coming week's seasonality.
Bottom Line: N/A
____________________________________
Indicator: SENTIMENT SURVEYS
Status: NEUTRAL
Comment: The story remains the same here, as it has for the past two months. The amount of bullishness exhibited by the respondents to the major sentiment surveys has not materially changed since mid-November. I've expounded on why I think this is long-term bearish several times in the past few weeks, so I'm not going to rehash that. The only survey that doesn't quite fit is AAII, where the 4-week moving average of the bullish ratio has dipped to 47%. Readings under 50% can be considered excessively bearish (bullish to us), although it reached as low as 38% during the July and October lows. While some dismiss the AAII survey as "too noisy", I find value in it on a moving average basis. However, the fact that this low reading is not being confirmed by the other surveys dampens its impact as far as I'm concerned.
Bottom Line: We continue to see an elevated amount of bullishness from the individual investors and newsletter writers that constitute the majority of these survey responses. In the context of a clear long-term downtrend, I believe this is troublesome.
____________________________________
Indicator: AIM MODEL
Status: NEUTRAL
Comment: There wasn't a great deal of change in the latest round of sentiment surveys, and that is reflected in this model, which only changed by .5%. The current reading of 52.1% is firmly neutral.
Bottom Line: N/A
____________________________________
Indicator: RYDEX RATIOS
Status: NEUTRAL
Comment: Over the past two weeks, we saw our shortest-term indicators here hit extremes that suggested there was too much optimism on the part of these market timers. The stochastic, RSI spread and Beta Chase Index all flashed extremes during the last push higher. It will take much more time for most of the indicators to work off their extremes, but the RSI spread at least has already cycled back down to oversold, with a current reading of -59. The last time this indicator reached this level was December 31st, and we all know what happened then. While I certainly don't expect the same kind of fireworks this time around, I do think that another day or two of market declines will show that these timers have overreacted on the short side once again, potentially setting us up for a run higher.
Bottom Line: These traders exhibited too much froth at the recent highs, and that will take more than a few days of declines to wear off. However, they did react aggressively on the short side in the past few days, suggesting that another push down may "trap" them into losing positions as is the usual pattern.
____________________________________
Indicator: COMPOSITE MODEL
Status: NEUTRAL
Comment: This model reached a low of 35% on Wednesday, which was the lowest relative and absolute level it had reached since March. The past few days have served to add a few percent to the model, as it now sits at 40% (which is still low and relatively bearish). In order for us to see a push under the lower trading band, we would need to see a drastic drop in the VIX and put/call ratios, along with our breadth measurements becoming more overbought than they were. These are unlikely events at the moment, so I suspect the model will continue to rise over the coming week.
Bottom Line: The recent low level of the model suggests that we were seeing a confluence of optimism that, while not "officially" extreme, was certainly bothersome in the context of long-term declining prices. There is a LOT of room for this model to rise, which tells me that the intermediate- to long-term sentiment condition is precarious.
____________________________________
Indicator: STEM MODEL
Status: NEUTRAL
Comment: This model did a good job of alerting us to overheated optimism early last week, as it suggested that we were about to go through a topping process with a low probability of significantly higher prices. The activity over the past week has relieved most of the pressure on the model, which is sitting in the middle of its trading range as of Friday's close.
Bottom Line: The model did its job of alerting us to exercise caution on the long side over the past week, and it is now back to neutral (which doesn't tell us much of anything).
____________________________________
Indicator: PUT/CALL RATIOS
Status: NEUTRAL
Comment: I've highlighted the OEX open interest ratio several times over the past week or so, and said that its current behavior was very troublesome. During the bear market, the S&P had never risen significantly when we saw readings such as we saw this week. In fact, even before the bear market, the S&P had trouble making headway when such extremes were reached. The situation has cooled a bit over the past couple of days, and could possibly change dramatically when the post-expiration figures are released next week. Once again, I believe it will be instructive to see how the OEX traders accumulate their positions after expiration. The more widely-followed traditional put/call ratios didn't move much this week, which is somewhat unusual for an expiration week. We typically see one or two very high put/call readings during expiration week, especially during a down market like we saw in the latter half of the week, as traders take some shots at hitting the lottery. I say "hitting the lottery" because there are situations that tempt almost every option trader at some point in their career. For example, the OEX January 460 put was trading for around $0.50 per contract on Thursday, and hit a high of over $4.00 on Friday. A return of 700% in about 8 hours is pretty good work if you can get it, and is a major reason why we often see a spike in volume during expiration week. In the S&P 500 options, I pointed out the high level of the SPX put/call bid/ask bias ratio on Wednesday, and how it had recently done a very good job at calling market direction the next day. It suggested that we should be see upside on Thursday or Friday at the latest, but of course, it promptly crapped out by giving us a dud this week. The current reading is somewhat high (bullish) but again, could be expiration-related.
Bottom Line: With a combination of a high OEX open interest ratio, low equity put/call ratio and high OEX put/call ratio, this complex of indicators is negative, though not extreme. Overall, it tells us that generally poorly capitalized traders continue to be optimistic while their more savvy index-trading cousins are not, which is a recipe for further downside.
____________________________________
Indicator: BREADTH RATIOS
Status: NEUTRAL
Comment: The extreme nature of our overbought breadth indicators, which was evident early last week, has now begun to dissipate. The 10-day average of advancing issues is has reversed from an extreme reading of .58 on Tuesday to .51 as of Friday. Likewise, the 10-day up volume ratio has dropped from a "maximum" overbought reading of .62 on Tuesday to .51 on Friday. This has relieved the immediate pressure on the indexes, but it will take more of a decline than we've seen to fully work off the excesses. The coming week will see two very large positive readings and two moderate negative readings drop off, so it's likely both ratios will decline further this week. For the first time in eight months, the NYSE new high / new low ratio reached overbought territory, with a 10-day average over .80. This is the most overbought reading since March/April. The current situation in this indicator closely matches that time frame, where the S&P 500 could not exceed the prior high set in January, yet the ratio of new highs to new lows did. While many view the expansion of new highs as a positive, I have found that this type of divergence typically leads to a continuation of the downtrend. If we have more stocks making new yearly highs, yet the indices cannot even exceed the previous swing high, then there is something wrong. In any event, this absolute level of the 10-day average has spelled trouble for the market ever since the mania started to wane in 1998. Since the beginning 1999, there have been seven peaks of the nh/nl ratio above the .80 level. The S&P was lower after 10 days every time, for an average loss of 3.2% from the date the nh/nl ratio peaked. Since our current ratio hasn't yet peaked (it should on Tuesday if we have a daily figure under .81), and if we assume that Tuesday's 10-day average will be less than Friday's, then that average decline projects down to approximately 875 on the S&P over the next two weeks. I don't put much faith in such projections, but it at least lets us know that significant and persistent upside from here is unlikely. Elsewhere in our breadth measurements, all of our cumulative TICK indicators are neutral. The Down Pressure indicators have reversed from their overbought readings on January 6th to almost oversold in the case of the S&P and firmly oversold in the NDX. Over the past 6 months, the NDX Down Pressure indicator has reached this same or higher level of oversold 7 times. All 7 instances lead to a higher market in the short-term (although one was a miniscule one-day affair), so this will be something to watch early next week. If we have another down day on Tuesday which pushes the Down Pressure reading even higher, then I would look for a relief bounce by Wednesday. The type of downward pressure we've seen in the NDX over the past three days is usually unsustainable, even in the midst of a down market.
Bottom Line: The breadth of the market became extremely overbought by Tuesday, and we are only beginning to work off that condition. This can be resolved by a long sideways market, or a quick decline, but almost certainly not by a price rise. This suggests that any rallies that occur over the coming week will likely fail. From the looks of the Down Pressure indicators, weakness early in the week should lead to something of a bounce, but again, that bounce should not lead to much.
____________________________________
Indicator: NYSE MEMBERS REPORT
Status: NEUTRAL
Comment: The S&P rose nearly 4% during the most recent reporting week here, and so it's no surprise that the specialist short ratio rose a couple of percent to 39%. This is still well below the mid-40% readings we saw at previous peaks over the past six months. One pattern that has existed over the past year is a consistent increase in public shorting. They have been increasing their shorting more and more near each intermediate-term low while letting up less and less on the rallies. I believe that this is a bullish development longer-term.
Bottom Line: Not much change here, despite a healthy rally during the reporting week. While the level of the current reading is neutral, the action of the public shorts versus the specialist shorts appears to be a long-term bullish development.
Over the past two weeks, the majority of our indicators were either in overbought territory or flirting with it. We had mostly negative indications, some neutral and no bullish, so the recent weakness should not come as a surprise. As of this weekend, things have become less intense as far as the absolute and relative levels of the indicators go. However, most of the indicators are only beginning to work off their extreme conditions, and that takes either time in a trading range or declining prices. Nothing I am seeing currently from a sentiment perspective would suggest that we are about to embark on a long journey upward, so I continue to stress that caution should be exercised from the long side in the intermediate-term. We will have short-term upside attempts, and I expect we will see one early next week, but I still believe that higher prices should be viewed as an opportunity to sell or hedge long positions, and look for shorting candidates. The best selling (or short entry) point is likely behind us, but if the current situation is anything like those of the past few years, there should be significantly more downside left to come. One thing that troubles me about this outlook is the action in the futures markets, which I believe to be bullish. I have never been one to rely solely on one indicator, but the commercial and small spec position changes during the market environment we've experienced would put me on guard with short positions. I would still focus on intermediate-term shorts on rallies, but perhaps reduce position sizes or keep tighter stop losses.
On a side note, I want to quickly mention a couple of things about the model portfolios. You can now find the archived rationales on the site under "Portfolios". Simply click on the trade number, and all of the trade adjustments will be there. Also, in answer to some subscriber questions, these portfolios are not meant to capture every swing in the market, by any means. They are meant solely as educational vehicles to show how one trader transforms sentiment extremes into actual buy and sell decisions. Not every major move will be accompanied by a sentiment extreme, so you should not expect the portfolios to capture those. They are meant only to show how I would transfer the odds of a trend change into a trading decision.
- Jason Goepfert
Disclosure: long QQQ puts
This disclosure is not intended as trading advice in any form. It is meant as a note to subscribers that I have a position directly affected by my market outlook. Although I take great pains to remain objective in my commentaries, I believe it is only fair that readers should know that I have taken positions in accordance with my market outlook.