sentimenTrader.com
All rights reserved.
Sunday, November 17, 2002
Once again, an intermediate-term neutral stance has saved us from multiple
whipsaws, as the S&P has had several wide swings lately, and is still
within 1.6% of where we closed last week. The NDX did a bit better,
higher by around 5% this week. The little rally to end to the week
certainly looks bullish, but may be too little, too late.
____________________________________
Indicator: RYDEX RATIOS
Status: BEARISH
Comment: I would consider this complex of indicators as bearish, but not in a major way. As I said last week, the 50-day moving average of the bullish and bearish ratios still include the extremes we saw in October, so the 10-day average, expressed as a percentage of the 50-day average, looks more extreme than it normally would be. Our other shorter-term measures are now neutral. After hitting a high last week, the 3-month bullish ratio stochastic has dropped some 25 points as the market has churned around. And the 3-day RSI, after pinpointing market turning points to the day over the past week, has returned to a firmly neutral condition. What all of this tells us is that although the asset flows from the bearish funds to the bullish funds had real momentum about a week ago, that has subsided to a great degree.
Bottom Line: Until we see more of these traders stampeding to one side of the boat or another and creating another extreme, there's not much to glean from this data.
____________________________________
Indicator: STEM MODEL
Status: BEARISH
Comment: In the "bottom line" area last week, I said:
"Watch for a spike above the bullish band in this model if we return down to the lower end of the trading range in the S&P 500 and Nasdaq 100. That would create a high-odds, multi-day trade on the long side."
We came very close to that exact scenario, but not enough to do any good. The S&P did trade to the lower end of its trading range, but the model only came within 5% of its upper band. This week, while the model has not yet pierced its lower trading band, meaning it is not "officially" negative, the chart below shows each instance over the past nine months or so where the model reached the level it is at now relative to its bands.
Obviously, the connotations are bearish, and I'm kind of splitting hairs here by not putting a negative indication on the model. The current model reading is 25% while the lower band is now sitting at 24%, so we're within a percent of the band. Each of the past times the model has come within a percent of its lower band, it soon pierced it, so I find the current reading to be troublesome. We'll be dropping a series of low readings beginning Monday, so in order to see a further deterioration in the model, the market would have to rally right from the get-go next week. This already has me on alert for the short-term, but if we get enough upside that the model pierces the lower band, then I think a tradable move down will soon follow.
Bottom Line: The STEM model is urging caution from the long side now, and if we rally further on Monday, it will create a high-probability scenario of a tradable move lower soon after.
____________________________________
Indicator: TRIN
Status: NEUTRAL
Comment: For the past couple of weeks, I've been pointing out the similarities in the NYSE and Nasdaq TRINs between now and last November. Last week, I suggested that what we were seeing was close to the small decline we had last November before we rallied once again. That again proved to be a good guide, and the analogy continues to work. Obviously, at one point soon this comparison will fall apart since each moment in the market is unique, but for now these indicators continue to remain neutral and suggest continued movement in a trading range is the most likely course of action.
Bottom Line: These indicators suggest more time in a relatively tight trading range is likely.
____________________________________
Indicator: VIX and VXN
Status: NEUTRAL
Comment: On Thursday, I mentioned that the large drop in the VIX that we saw only occurred 20 other times during this bear market, and 13 of those resulted in a lower market 5 days later. However, of the 7 "failed" signals, all 7 occurred after a major decline, and not after a large rally like we've just seen. The VIX has now dropped 15% over the past two days, which is the largest two-day decline since August 15th. That time, the market continued to rally for another week before topping on the 22nd, and is probably the closest approximation to what we are seeing now. The VIX has now also closed more than 10% below its 10-day average, which last occurred on 10/15 and the topping week of 8/14-8/22 prior to that. While not a particularly effective market timing strategy unto itself, this 10% deviation has been a good heads-up to watch for possible reversals within the next week, especially if we get two or more days of the VIX more than 10% below its average.
Bottom Line: The VIX is showing some signs that often preclude a short-term high. If we get another day or two of the VIX being stretched from its average, we should become much more worried about a possible short-term high being put in.
____________________________________
Indicator: COMPOSITE MODEL
Status: NEUTRAL
Comment: As shouldn't be a surprise considering market action over the past 5 days, this model is essentially unmoved from last week. For the coming week, a deterioration in the sentiment surveys and commitments of traders information will be a further drag on the model, so a rally accompanied by overbought readings in some of our shorter-term indicators could cause a quick drop in the model, although I doubt we would see it hit the lower standard deviation band. At most, we would likely see a reading equivalent or slightly lower than the reading at the August high.
Bottom Line: Again, not much to read into here unless and until we begin to see a confluence of extremes in the component measures.
____________________________________
Indicator: SENTIMENT SURVEYS
Status: NEUTRAL
Comment: Not much change here, as the Market Vane and AAII surveys shed a bit of bullishness while the Consensus and Investor's Intelligence polls added some. There were no major changes in any of the surveys, and I don't expect there to be any surprises next week either. Thus is life in a trading range. The only thing to be gleaned from this indicator set is the negative divergence I've been talking about for a couple of weeks now (i.e. price has made a lower high from August, but the sentiment surveys have registered a higher high, suggesting optimism is getting ahead of price).
Bottom Line: While the divergence is negative, none of the surveys are yet showing troubling amounts of optimism.
____________________________________
Indicator: AIM MODEL
Status: NEUTRAL
Comment: The essentially unchanged nature of the sentiment surveys has manifested itself into an unchanged reading here (for all practical purposes), as the model dropped a whopping 0.5%.
Bottom Line: Same as last week - the divergence between now and August is troubling, but the model itself is not low enough to trigger outright concern.
____________________________________
Indicator: COMMITMENTS OF TRADERS
Status: NEUTRAL
Comment: For a down week during the latest reporting period, it's relatively unusual to see the action we did. The commercial hedgers in the S&P 500 futures shed some long contracts, but mainly increased their shorting activity, to a net result of an increase of 5,000 contracts to their net short position. Meanwhile, the small speculators in that same contract added a couple of thousand long contracts and bought back a fairly large number of short contracts, for a net increase to their net long position of about 8,200 contracts. These position changes caused the three-month stochastic for the commercials to drop from 45 to 29 and raised the stochastic for the small specs from 32 to 55. The one-year stochastic suffered a substantially lesser change. While negative, these changes are still not extreme enough to suggest that these traders have positioned themselves in a manner that implies imminent downside - there's quite a ways to go before we can say that. I don't have a high degree of confidence in the efficacy of the Nasdaq commitments, but in that index the commercials have become the most net short they have been in nearly a year, while the small specs have gotten relatively long (with stochastic readings above 80). Theoretically, these are both negatives, but since the effectiveness of the Nasdaq commitments is in doubt, I would consider it only minor.
Bottom Line: It's safe to say there's a trend of the commercials becoming more net short and the small specs more net long since the low in early October. Overall, this is normal behavior during an intermediate-term advance and has not yet reached a stage that should be considered unhealthy.
____________________________________
Indicator: SEASONALITY
Status: NEUTRAL
Comment: The seasonality forces in force last week didn't play out much at all like they normally do. Veteran's day, normally very positive for the market, ended down substantially, and expiration day ended positive (though with a weak morning). About the only thing that went according to schedule was the generally positive bias experienced during expiration weeks. The only historical bias this week is the slight negativity that we have tended to see after option expiration. During this bear market, the Monday following expiration has been negative 63% of the time, second only to Wednesday (58%) and Friday (also 58%). This is only a minor factor, and I wouldn't read a whole lot into it.
Bottom Line: This week has a minor negative bias historically, but it comes in the context of one of the strongest months.
____________________________________
Indicator: STEM.MR MODEL
Status: NEUTRAL
Comment: Similar to the STEM model, I'm kind of splitting hairs here by not tagging a negative indication to the STEM.MR. The current reading of 22% is close enough to the 19% lower standard deviation band to consider it "dangerous", although it's not officially negative. This is all just semantics, and the main point is that in the short-term, we're seeing a relatively high amount of optimism. Over the past two months, the only time this didn't lead to weakness (about 20 S&P points) within a day was on 10/11, where we dropped about 10 points before another large rally. Like the STEM model, if we see some sustained upside on Monday, it should provide a good opportunity to tighten stops on any long positions and begin to hunt for short candidates as a high-odds, short-term trade.
Bottom Line: This model is urging (but not yet yelling) caution, and further market gains early in the week should set up a solid negative condition here.
____________________________________
Indicator: PUT/CALL RATIOS
Status: NEUTRAL
Comment: For the most part, these indicators are very close to bearish territory. I normally post the CBOE-only put/call ratios (i.e. options traded on the Chicago Board Options Exchange), but I thought I would post an all-exchange put/call ratio today. This ratio takes into account volume across all U.S. exchanges, so it's quite comprehensive. The chart below is the 10-day average of the ratio.
As you can see from the chart, when the average peaks above .70, it has coincided well with intermediate-term lows during the bear market. Conversely, troughs at or around .50 have preceded market weakness a significant amount of the time. The current reading is .54, which is obviously approaching an area that would urge caution. This fits well with the CBOE total and equity-only put/call ratios, which are both close to their lower standard deviation bands on a moving average basis. On a daily basis, these two ratios pierced their lower bands on Friday, although as I pointed out on Thursday, when the trend is up the put/call ratios tend to be quite low during expiration week. The one kink in this complex of indicators, the one thing that doesn't "fit", is the OEX put/call ratio. The traders who frequent these index options tend to have a much better handle on market direction, and they are quite often right at the extremes, while the equities options traders are almost always wrong. So, the fact that equities options traders have been betting on upside suggests that we should take the other side and bet on downside. But since the OEX traders are also apparently quite optimistic, it throws a bit of water on that line of thinking. The equities traders have a better record of being wrong than the OEX traders have of being right, so overall I would rate this complex as neutral, but very, very close to being outright bearish. In the S&P 500 options, our put/call bid/ask bias ratio has been neutral all week, and was on the low side on Friday, but this indicator is influenced greatly by expiration and I wouldn't read too much into it.
Bottom Line: The poorly-capitalized equities options traders have shown a consistent bias for calls over the past couple of weeks, and it is beginning to reach a critical stage. If we see another couple of days of heavy call volume (in relation to puts), it would suggest that these options buyers will soon be wrong and the market will likely decline over the ensuing weeks.
____________________________________
Indicator: BREADTH RATIOS
Status: NEUTRAL
Comment: I mentioned last week that the 10-day average of the advance/decline line had six straight large positive readings to drop, meaning it would be difficult for us to become overbought here. Well, we have three more readings to go (meaning until Wednesday), before we start dropping some negative days, so an early-week rally would likely not coincide with an overbought situation. This either means that we would have a loss of momentum on any rally, or that there is fuel left for further gains, so I'm not going to try to read anything into the numbers themselves. Rather, I will wait for a clear overbought reading before passing judgment. The same will go for the 10-day up volume indicator, which continues to churn around our overbought level, and likely will for another few days unless we decline early in the week. Our cumulative TICK indicators, both daily and intraday, are neutral but could threaten overbought at any moment. The best guide we have had over the past few weeks of this trading range have been the price oscillators. Again, this is a method of quantifying individual price bars (especially easy to think of candlesticks for those of you familiar with that charting method). I would refer you back to Thursday's commentary for an explanation of how each price bar is scored. The oscillators I post are based on 30-minute bars, but one can also calculate it just as easily for daily bars, which is what I have done below:
This is a three-day average, and when it reaches 70%, it has typically coincided with overbought market conditions and a pause soon to follow. Our current reading is 73% (meaning the average price bar over the past three days received a score of 73%), which puts us in overbought territory. As is true with most oscillator-type indicators, it works best in trading ranges, or when it gives contra indications (i.e. overbought in a downtrend or oversold in an uptrend). If we are overbought here, but break out of the trading range, it will obviously give a false signal.
Bottom Line: Most of our indicators in this bunch are neutral to slightly overbought, which isn't enough of a confluence to have me overly concerned with market breadth at the moment.
____________________________________
Indicator: NYSE MEMBERS REPORT
Status: NEUTRAL
Comment: Pretty much an exact repeat of last week here, as the public still has not let up significantly in their shorting activities. While our latest reading of 38% is not abnormally low (within the context of the past five years), it is quite low considering the rally we've had off of the October low. Normally we would see a more definitive lessening of public shorting than we've seen this time around. Like I said last week, this is one indication of the public (probably mostly hedge funds) not believing this rally will last and continuing to bet on falling prices. This reading is not extreme, but could be considered relatively positive.
Bottom Line: Same as last week - although the current level of shorting activity is neutral, the fact that it has not lessened to a great degree with a relatively large rally off the lows is positive.
I left off last week saying that if we traded through the 900 area on the S&P, we would likely visit the lower end of the trading range. We did so, and obviously rallied off that level. The NDX also traded back through its breakout level but showed some relative strength and didn't quite make it to the other end of its trading range. We're now back to the recent highs (in the NDX, but have a bit more to go in the S&P), so this week should prove telling. My experience is that breaks of obvious highs, lows and trendlines tend to create a self-fulfilling cycle for at least the next few days, as trend followers pile on. Therefore, if we do break the recent highs I would not be aggressive in going against that trend probably for two or three days. If we begin to fail anywhere in here, it may create a double-top formation that will embolden the bears and create some severe selling. As of this weekend, sentiment is neutral to slightly overbought, but if we get one more day of significant upside, it's very likely the STEM and STEM.MR models will go negative. At that time, if we have not broken the recent highs on the S&P, my opinion is that a failure is likely (at least short-term) and one should look for reversals down. If we HAVE broken the highs, then I would back off for a couple of days and let the trend-followers exert their influence before fading the move (unless we break the high and reverse back down intraday or immediately the next day - that would be quite bearish). Although a majority of our intermediate-term indicators are still neutral, no more than one week of generally rising prices should push most of them solidly into overbought. Whatever price level we are at then, the chances of significantly more upside would be limited, and we would more likely suffer a correction more severe than what we've seen recently.
- Jason Goepfert