http://www.sentimentrader.com

Wednesday, June 18, 2003  8:45 PM EST

For the past two days, the S&P 500 has closed within 0.2% of the previous close.  It’s rare for three consecutive closes to all be within such close proximity to each other, as such instances have happened on less than 5% of the trading days since 1950.  Historically, this hasn’t meant much going forward.  But over the past five years, after the third day of such close closes, the S&P has often kicked into gear one way or the other.  By the fourth day, the average close has been over 1.3% away (more often than not to the downside, but I think this is more a function of the sample period than an inherent bias).  This tells us that after three days of indecisive behavior, look for the market to try to trend again.  For day-trading purposes, I would be more inclined to ride whatever trend asserts itself, instead of trying to play bounces within a trading range. 

Not only have we not seen much change on a closing basis, the daily range has also been quite small for the past two days.  When expressed as a function of price, the two-day range (cumulative) is the smallest seen since late December 2002.  Based on historical precedent, we should expect to see a range closer to 14 points or more in the S&P tomorrow, as volatility begins to pick up after a two-day lull.  These short-term nuances aren’t of much use to longer-term traders, but for very short-term traders it suggests that once we break this range we’ve been in (and we likely will tomorrow), the market should carry for a bit. 

I’ve mentioned the high level of the OEX put/call open interest ratio several times recently.  Again, this shows the amount of OEX put option contracts open in relation to call contracts – the higher the ratio, the greater the concentration in puts.  Currently, we’re seeing the highest ratio since 1999, meaning there are a ton of put contracts open as opposed to calls.  This is the fourth-highest ratio two days before an expiration over the past decade.  If we compare the highest ratios two days before an expiration to the lowest ratios, a bit of a bias stands out.  For the highest ratios, the S&P 100 was positive going into expiration day 70% of the time.  This contrasts to 55% of the time for the lowest ratios.  However, the Monday following expiration was positive only 25% of the time for the highest ratios, while it was positive more than 50% of the time for the lowest ratios.  Expanding the look out to the entire week following expiration, the OEX showed a bit more of a negative bias after seeing high put/call open interest ratios than it did after low ratios.   

Shifting gears to volume patterns in equity options, the 10-day put/call ratio (minus QQQ options) has dropped to a new low, going back to May 2001 (the furthest back this data is available without QQQ options).  At a current reading of 0.46, the ratio has eclipsed the previous low readings seen in May and November 2001.  This has been a consistently accurate contrary indicator during the past two years once it reaches an extreme.  High readings, which show excessive put activity in relation to calls (usually seen in panic-type situations) have coincided well with market lows, while very low readings (high call activity compared to puts) have pinpointed short- to intermediate-term highs with few exceptions.  This time is a bit different, however, in that the ratio dropped to a very low 0.49 a couple of weeks ago, and we’ve rallied a little under 3% since then.  This is similar to November 2001, when the 10-day average dropped to 0.50 early in the month and the market continued to rally about another 5%.  Then in late November, the put/call ratio came down once more which corresponded quite closely with that market peak, which the following chart shows. 

On Monday, I presented the results of a study showing that extremely low one-day readings (which we saw on Monday) typically lead to more weakness than strength over the next 1 to 10 days.  I see no reason yet why it wouldn’t still be applicable, so that remains a bit of a concern. 

There is certainly the possibility that these put/call ratios could be entering a new, lower trading range, similar to what we saw from 1997 – 2000.  During that time, the equity-only put/call ratio spent most of its time between 0.35 and 0.55, rarely venturing beyond those bounds.  From 2001 on, the ratio spent most of its time between 0.55 and 0.80, a much higher trading range than during the latter stages of the bull market.  If we are now transitioning to a longer-term bull phase in the market, then I would fully expect the put/call ratios to trend lower for some time to come.  That should be taken into account when discussing how low the ratios are now compared to what we’ve seen over the past three years.  While what we’re seeing now is extremely troublesome in the context of recent memory, historically the 10-day equity put/call ratio is still above its all-time mean. 

In the Rydex series of mutual funds, traders have become extremely gun-shy about betting against this rally, and in fact have become quite comfortable with the long side.  The two funds most leveraged to the long side, for the S&P 500 and the Nasdaq 100, have been hitting recent asset highs in the past couple of days, while bear funds like Arktos are hitting new asset lows.  This phenomenon has arguably become more extreme than we’ve seen at any other point over the past couple of years.  The chart below shows the differential between where the average bull fund is compared to its 52-week range as opposed to where the average bear fund is.  Currently, the average long-side fund (in terms of assets) is at 89% of its one-year range, while the average short-side fund is only at 41%.  This difference of 48% is the widest gap seen at any other time during the past two years (the furthest back this type of analysis can go due to data limitations). 

While this may not have an immediate impact (note the “early” signals given in June 2002, and in May of this year), it suggests that the Rydex timers, notoriously wrong at market direction when we hit extremes, have become aggressively long in their asset choices. 

The type of aggression we’re seeing from those most wrong, most consistently on market direction (i.e. equity option traders; Rydex mutual fund timers; sentiment survey respondents; and to a lesser extent small speculators in S&P 500 futures, especially when the e-mini is considered) has to be considered a negative for further equity advances.  True, we have seen these traders become extreme before, even recently, without it impacting the market, but we have not seen it to this degree.  There is a good deal of real-money “froth” being thrown at the market, and this speculative hot money will likely quickly reverse should the market show signs of vulnerability.  Those signs have not shown up yet, so I still am not enamored with being either long or short at the moment.  Generally, I am sticking with my earlier stated preference of looking long above 1008 on the S&P and short below it.

Disclosure: long OEX puts

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, June 16, 2003  11:30 PM EST

Today’s closing equity put/call ratio (minus QQQ options) was an extremely low 0.38, which shows us that there were more than 2.5 calls traded for every 1 put.  Going back to 2001, the furthest back this data is available without QQQ options, the S&P 500 put in the following performance when we saw readings of 0.40 or below.  There were 14 occurrences. 

This may be a little confusing, so let’s go over one of the categories in detail.  Five trading days after such a low equity p/c ratio, the S&P was 0.8% lower than it was on the day the reading was generated, on average, and it was lower 64% of the time.  The average rise within those five days was 1.3%, with the maximum rise being 3.9%.  This can be thought of as the drawdown had one sold short the S&P 500 on the close the day the low p/c reading was seen.  The average drop within those next five days was 2.4%, with a maximum decline of 4.9%.  This can be thought of as your maximum gain had you sold short the S&P on the day of the low p/c ratio. 

We can see from the table that on balance, selling short the S&P 500 after we see such low p/c ratios would likely have been a winning strategy overall, as the average (and maximum) drawdown is less than the average (and maximum) “win” across every time frame up to 10 days.  I would NEVER suggest anyone trade based simply off these figures, but it gives us good background for has happened the other times we have seen such low put/call readings.  The main issue with this, of course, is that it only goes back to 2001 and does not incorporate a similar market environment to what we are experiencing now. 

For really the first time since February, the Commitments of Traders information showed a moderate negative change as of last Tuesday’s close (the most recent information available).  Commercial traders reduced their net long position by a little over 13,500 contracts in the full S&P 500 futures contract, while small speculators increased their net longs by nearly 11,500 contracts.  This is the largest negative contract adjustment (i.e. negative for the market, since commercials became less long while small specs became more long) since early July 2002, with December 2002 coming close.  This is the first time in two months that small speculators are more net long than the commercials.  However, even with these negative changes, our Futures Balance Matrix, which shows relative long-side commitments of the commercials vs. small specs, is still relatively positive.  These positions would have to become significantly more drastic before they could be considered negative in an historical sense. 

In the e-mini, which I mention more for informational purposes than anything else, small speculators set another record net long position, while commercials increased their net shorts (though not quite to a record).  I think this has to be on the radar, though I’m as yet unclear as to how it should best be interpreted at this point. 

Last week, I showed the results of a study comparing the recent low level of Investor’s Intelligence bears to past occurrences.  Those results were relatively bullish for the market over the ensuing few weeks before a more serious correction typically set in.  However, when we incorporate the other sentiment surveys in terms of the AIM model, we’re now getting the lowest reading since January 2000, which preceded a rather sharp 10% correction over the next few weeks.  A low model reading corresponds to a high level of optimism being displayed across the broad cross-section of investor groups polled by the four major sentiment surveys. This most recent reading is actually one of the lowest seen over the past decade.  Since the weekly trend of the market is now neutral, this extreme optimism loses some of its import, but it should still serve as something of a caution flag.  If an event comes along to throw a wrench into the market at this point, we must be aware that there are very high expectations currently built in.  If (when) doubt begins to creep back in, there could be a tidal wave of traders rushing to protect their gains. 

I’ve often heard the argument that the Investor’s Intelligence survey shouldn’t work, since it reflects what the population (in this case, newsletter writers) are SAYING they are doing, but doesn’t show where they are actually putting their money.  I’ve never understood that argument, since it simply doesn’t hold up when one does a simple test.  The following charts plot the Investor’s Intelligence bull ratio (bulls / (bulls + bears)) versus the dollar value of the small speculator net positions in the S&P 500 futures contracts (combined large and e-mini contracts for the purpose of this example).  The first chart shows the raw values, while the second chart shows how each series compares to its 52-week range (meaning when the lines are near the top of the chart, then that series is at a 52-week high). 

We can see that the two series track quite closely, and supports the argument that those who respond to the sentiment surveys (at least the II survey) walk the talk.  This is why these surveys tend to work consistently - when they show a high level of bullish opinion, then it is likely that they have taken positions in accordance with their market outlook, and it reduces the "fuel" available to spur the market higher.

I said last week that there were enough signs that a move over last Friday’s high of 1008 could carry significantly further, and I would back off the short side until we either fall back below 1008 or sentiment becomes truly euphoric.  I’m sticking with that for now, and until such time, I don’t see a reason to fight this momentum again just yet.

- Jason Goepfert

Disclosure: long OEX puts

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.