Thursday, April 24, 2003 8:50 PM EST
I've received a lot of requests to update the breadth combination chart I presented a few days ago. This indicator is simply a combination of the 21-day average of the advance/decline line and the 21-day average of the up volume ratio. The way I've presented it below is adjusted to make a neutral reading equate to 0%. The red area shows readings that are greater than zero, and when they reach 8%+, the market is becoming overbought. The green area shows readings below zero, and when they reach -8% or below, the market is becoming oversold.
Yesterday's reading of +15% was the most overbought in nearly five years. During the bear market, the next highest readings were +14% on 5/22/01 and 11/7/02. As I stated earlier, this is indicative of unsustainable pressure in the context of a persistent downtrend.
I've also been asked to present again the 21-day NYSE cumulative TICK. The last time I showed it, we were hovering around +60,000. As of today's close, we reached +81,000. Again, this is a new bear-market record.
On Tuesday, I mentioned the high level of put open interest in OEX options, and that this was a bearish sign. These traders continue to open put contracts in a greater proportion than calls, pushing the ratio between the two to 1.43, one of the highest readings seen during the bear market. Since OEX options have an American-style expiration (meaning they can be exercised anytime), institutions tend to shy away from selling short these contracts. We can make the assumption that much of this increase in put open interest is coming from traders buying puts to protect or speculate on declining prices. From the record of these traders calling market turns, this is a mildly bearish development.
There were some interesting developments in the major sentiment surveys this week. You've likely already heard about the large drop in bullishness displayed in Chartcraft's Investor's Intelligence survey. The bull ratio (bulls / (bulls + bears)) dropped from 62.5% to 55.1% while the S&P rallied over 2% during the reporting period. This is HIGHLY unusual, as this type of large drop in bulls during a 2% up week has happened only three other times in the past decade. The first two were in 1997 and lead to continuations of the strong uptrend already in place. The other took place during the week of the October 2002 low. Obviously, this appears market-positive, although with such a small sample size it's impossible to derive any solid conclusions.
Disputing the readings from the II survey somewhat are the latest indications from the American Association of Individual Investors (AAII) poll. The percentage of respondents who are bullish rose from 46% to 63%, while those bearish dropped from 32% to 19%. The chart below shows all instances of the percentage bullish above 60% over the past decade. The vertical lines represent those weeks that showed greater than 60% bulls (if the weeks are consecutive, only one line is shown).
Incredibly, during the upside run from 1995 through 1999, there were only TWO weeks which showed a bullish percentage greater than 60%. Even during that great run, such optimism lead to market weakness soon after. In contrast, during just one year beginning in November 1999, there were sixteen such weeks. This was the heyday of investment clubs and internet chat rooms, so it should be no surprise that it coincided with the height of bull-market mania. Also no surprise, there were five weeks with a bullish percentage over 60% in 2001 alone, as those false rallies sucked in investors hand over fist. Apparently, the recent market activity has been different enough that a majority of these respondents feel more positive about the future now than they did during the rallies last year. From the looks of the chart above, unfortunately I cannot share their optimism for the next few weeks at least.
All of these inputs have dragged the broad-based Composite model into strong negative territory for only the third time since 2000. The first was August 2000 and the second was March 2002. Honestly, I've been surprised that the model has dropped so far so fast, particularly with the positives from the Commitments of Traders data and Specialist Short Ratio. But the historic levels of overbought breadth (advance/decline, up volume and cumulative TICKS) combined with the sustained low relative VIX have weighed heavily on the model and are the largest reason it's now giving such a negative indication.
I've been allowing for a move to the upper end of what should prove to be stiff resistance around 930/940 on the S&P. Somewhere between here and there, I believe that from a risk/reward standpoint, it makes the most sense for intermediate-term traders and investors to be getting out of or aggressively hedging their long positions, and possibly even looking for short candidates. In order to justify being aggressively long in here, one has to believe that we are in the transitory stage from bear market to bull. While that could certainly be the case, and I see the possibilities, I prefer to make the "bull market" prove itself before I'm willing to jump on board. In the interim, selling such high hope has proved to be the best trade out there.
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.
Tuesday, April 22, 2003 10:00 PM EST
Once again today, we saw a large trader in QQQ (Nasdaq 100 tracking stock) put options. Since QQQ options are considered "equity" options instead of "index" options as far as the CBOE is concerned (even though they track an index instead of an individual security), they tend to skew the equity put/call ratio that I post to the site. This has happened frequently over the past few months. Interestingly, even though QQQ volume has declined as a percentage of NYSE volume over the past year, QQQ option volume has been increasing as a percentage of total option volume. Today, for example, QQQ put volume accounted for a whopping 49% of total equity put volume. While I believe that QQQ options are a haven for unsophisticated options traders due to their low premium, they also attract institutional investors because of the high liquidity. When a large fund steps into these options, it can greatly skew the ratio as we've seen several times lately. Because of these frequent distortions, and because of the large impact QQQ options have on the equity put/call ratio in general, I think it pays to back those options out and see what happens. For instance, the equity put/call ratio ended up closing at 0.77 today, which is historically high and would lead one to conclude that options traders were betting against this rally. That would be a sign that the rally is probably going to continue. However, if we back out the QQQ options, the equity put/call ratio drops all the way down to 0.43 which is historically low, and suggests that if it weren't for one large trader in QQQ options, in general traders were quite active in call options as opposed to puts - a bearish sign.
I like to view the equity p/c ratio on a 10-day moving average basis. The chart below shows the traditional ratio over the past year, compared to the ratio if QQQ option volume is backed out.
We can see that the ratio without QQQ volume (the red line in the chart above) tends to be consistently lower than the traditional ratio. In fact, over the past year it has tended to be about 5% lower. Since QQQ option volume has been increasing as a function of total option volume, and the QQQ p/c ratio has tended to run higher than the non-QQQ p/c ratio, the traditional way of looking at the equity p/c ratio has been skewed higher and is becoming more so.
If we look at the equity p/c ratio sans QQQ options, then the 10-day average is approaching its lower trading band, and is close to becoming bearish (but isn't quite there yet). If we see another day like today, where non-QQQ call volume is significantly heavier than put volume, then the 10-day average should drop to levels seen at prior market peaks during the past year.
Also of interest in the options arena, the OEX (S&P 100) open interest configuration is becoming extreme, particularly after the most recent expiration. Open interest is simply the number of option contracts currently open. The ratio of OEX puts currently open versus calls has been steadily ticking higher, and finally entered extreme territory after expiration, suggesting OEX traders let their calls expire while keeping or rolling into longer-dated puts. From a non-contrarian perspective, this is bearish. The chart below shows the ratio of open interest in OEX puts versus OEX calls, on a de-trended basis. The de-trending is necessary since this ratio has shown a secular decline over the past 8 years or so, making it difficult to compare current extremes to those several years ago. To de-trend the ratio, I have taken the deviation of the 10-day moving average from the 6-month average. This shows the oscillations of the ratio around its recent mean. When the indicator is high (say above +20%), then PUT open interest is historically high. When the indicator is low (under -20%), then CALL open interest is historically high. As with most sentiment indicators, this becomes most effective when the larger trend is taken into consideration. So prior to 2000, when the trend was up, only oversold readings (the green line) should be noted. From 2000 forward, overbought readings (the red line) become more effective.
Currently, we are overbought, meaning OEX put open interest is quite high, and as you can see from the chart, that has lead to weakness with a high degree of consistency, particularly during this downtrend.
Market breadth remains overbought of course, and could become historically so should we continue to rally in the short-term. Amazingly, if we have even moderate up days for the next two days, we would likely record one of the most overbought 10-day advance/decline readings in history. We would almost certainly see one of the top 5 overbought readings in the past decade. To go along with that, all of the cumulative TICK indicators that I record are in overbought territory. Several of the longer-term ones (10-day and 21-day in particular) are historically overbought - reaching 5-year record readings this afternoon. I've discussed overbought breadth many times over the past two weeks, so I'm not going to continue going on about it, but when it reaches these types of historic extremes, it should be on the front burners.
I know many of you are sick of hearing about the VIX, but I have to mention the truly historic reading of the VIX Transform. At 15.0%, this indicator is now at maximum overbought, for all intents and purposes. This level has been reached only five other times in the history of the VIX (October 1987, Summer 1988, April 1991, July 1999, and April 2002). No matter the market environment, whether it be bull or bear, the 30-day returns after peaks above such levels have been muted. This indicator should not be used for buy or sell decisions, rather it is best at identifying those times when one should concentrate on intermediate-term long or short setups. Currently, it is suggesting without question that the greatest risk/reward situations for positions held several weeks (at least) lie with the short side.
Going over a cross-section of our indicators, I was surprised tonight at how many are at or within a day or two of becoming overbought in an historic fashion. This is not what I would expect when we haven't even broken the downtrend line from the Spring of 2000. So, obviously, we are overbought - the question is whether we are entering a new phase where overbought readings do not matter. With the major indexes flirting with long-term downtrend lines, the possibility is there, but the proof must remain on the bulls' shoulders. Every reading similar to our current one during this downtrend has resulted in weakness in the intermediate-term, so until that changes, I believe the trend that is currently in place should be respected.
Today's breakout was very pretty from a technical standpoint, and I suspect we may carry a bit higher because of that very reason (as new trend-followers are brought on). I've been suggesting that if we break 905, then the 930/940 area looks like the next likely stopping point. If we reach that level, then I think it would pay to become quite aggressive on the short side, but even before that level is (potentially) reached, I think longer-term traders should be concentrating on short candidates. The truly historic readings in many of our measurements are suggesting strongly that this breakout should not be chased for anything other than a very short-term trade.
- Jason Goepfert
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.
© 2003 Sundial Capital Research, Inc. All Rights Reserved.