sentimenTrader.com

All rights reserved.

Thursday, August 15, 2002  11:15 PM EST

No commentary this evening.

Wednesday, August 14, 2002  11:13 PM EST

My apologies, but tonight's comment will be rather short.

The rally today certainly had nothing to do with sentiment, as the persistent move began when our short-term sentiment measures were on the lower end of neutral at best.  It's quite rare to see such a move when sentiment is as ambiguous as it was, so it caught me (and I suspect quite a few others) by surprise.  Today was certainly out of the ordinary.

I try to avoid discussing technical or fundamental analysis in this forum, but I've mentioned the triangle pattern that has been forming in the broader market averages since the rally on July 24th.  I think this pattern is being watched by a lot of money, and a breakout on a daily basis could see some fresh money being pumped into equities.  These common patterns can often trigger violent moves, so I would be hesitant about taking trades against a rising market until I had more confirmation than usual, especially with the decent volume we saw today.  Looking at a daily chart of the S&P, it's easy to see the upside possibilities.

At this point, our shorter-term sentiment measures are mostly overbought, but not exceedingly so, allowing for a possibility of more upside.  The intraday cumulative TICK indicators made large moves and are quickly approaching their upper ranges.

The S&P 500 put/call bid/ask bias ratio closed at a very bearish 0.11.  This is the lowest since 0.09 on August 9th.  There was a very clear and consistent pattern of selling calls and buying puts by the institutions.  Here is the breakdown of trading volume for the three nearest expirations:

Volume in the October expiration is still very low, so it's difficult to read much into that.  Otherwise, you can see the clear pattern of calls being traded at or below bid while puts were traded at or above ask.  That is bearish.

Bottom line, it's likely we will have follow-through on the upside here, but our shorter-term indicators will soon become very overbought.  That in combination with the exceedingly bearish institutional options activity will have me looking for downward reversals on a shorter-term time frame.  Don't forget (the indicator bias graphs on the home page are a reminder) that the intermediate-term sentiment picture is still rather positive, so any bearish remarks I make are primarily directed at short-term (less than 5 days) moves.

Tuesday, August 13, 2002  11:20 PM EST

So with a 200+ point drop in the Dow, the VIX and put/call ratios actually drop today?  The NEGATIVE stage change in the STEM.MR model was issued with the S&P at 910 when the model was at 11, and the S&P ends 26 points lower at 883 and the model only rises 8 points?  That smacks of a buying-the-dip mentality that is a definite short-term negative. 

The fact that the VIX dropped along with the OEX is quite unusual.  Since 1986, the OEX has declined on 47% of the days.  Of those days, only 12% of the time did the VIX also drop.  When the OEX drops more than 2%, as it did today, the average increase in the VIX is 15.4%.  I looked at those days when the OEX dropped 2% or more on any given day, and compared the times the VIX rose versus when it dropped, to see if it led to a difference in performance going forward.

When the OEX dropped 2% or more and the VIX also dropped, it appears that it lead to short-term underperformance, but that it recovered after 10 days.  We can see that the average next-day return when the VIX rose was 0.1% with the OEX up 59% of the time.  But when the VIX fell, the average next-day return was minus 0.2% with the OEX up only 38% of the time.  It appears as though the VIX drop today was a negative development.

In going back and forth with some of the subscribers during the morning trading today, I suggested that one of the reasons for the large call sale yesterday was the assumption that with the uncertainty of the Fed decision out of they way, implied volatility might drop even with a drop in the broader market.  This would give a call sale an edge over a put purchase.  It looks like a pretty sharp institution figured that out and played it that way.  The 10,000 calls they sold yesterday for around $50/contract are now going for an average of $34.50/contract.  Total one-day profit?  $15,500,000.  I saw no evidence of a covering of those sales, but that will be infinitely more difficult to detect.  The put/call bid/ask bias ratio remains low and bearish at 0.47, even with the drop today.  The bias was particularly evident in the August expiration, although with expiration coming fast, these figures may become quite distorted.

I mentioned the total put/call ratio vs. the OEX put/call ratio yesterday and how it was becoming disturbing, as the small-money retail traders were sanguine while the bigger-money OEX players were becoming bearish.  That trend continued today, as the spread between the 5-day moving averages is the widest since May (I also keep a 10-day moving average, but the 5-day is of course more responsive):

I have drawn arrows showing the last two times such a wide spread has been reached in the past few months, and the results were not encouraging.  As you can see in 2001, the spread got much wider before any ill effects were felt.  If this is indeed a new intermediate-term trend we're in, we could see this spread widen considerably, but it does seem very early to see such a discrepancy.  Effective tonight, I have added the 5-day and 10-day moving averages of this spread to the regular indicators.  You can find them in the indicator table or under "CBOE Ratios" in the Indicator section.

The bottom line is that the breadth measurements are beginning to deteriorate while at the same time short-term sentiment is benign to even optimistic.  That does not bode well for the short-term, and I will continue to favor the short side for now.

Monday, August 12, 2002  9:24 PM EST

One of the most lopsidedly bearish put/call bid/ask bias ratios I've ever recorded (on Friday) manifested itself into a large gap down open this morning, and the markets spent the rest of the day making a decent attempt at recovery.  If today's action in the S&P options pits is any indication, the recovery attempt will be short-lived.  I mentioned in the intraday commentary that the bias ratio was staying low even with the sell off in the indices.  The bias ratio ended at an extremely low 0.13, and shows a clear bearish bias of selling calls AND buying puts.  In the August expiration, 27% of puts went off at or above ask (as opposed to 13% at bid) while 32% of calls went off at or below bid (as opposed to 17% at ask).  In the September expiration, 10% of puts went off at ask as opposed to 11% at bid (no particular bias there).  However, in the most strikingly lopsided example to date, 56% of the September calls went off at or below bid as opposed to only 5% at or above ask.  After I posted the late intraday comment, a massive series of trades went through in those Sept calls.  Between 3:07 and 3:21pm EST, an institution sold 10,000 contracts well below bid in the September calls.  The trade was spread evenly with 2,500 contracts each being sold in the 880, 885, 890 and 900 strikes.  Each trade went off between 2-5% BELOW the bid price.  These contracts were trading around an average price of $50 apiece at the time, which makes this a $50 million trade.  Of course, this could have been different institutions, but it's highly unlikely.  Also, there's no telling what strategy was being implemented (it could, in fact, be a bullish trade), but it does not appear as though there were any offsetting options transactions.  It's possible that this trade could be a hedge on a large long stock position, but I feel that's also unlikely.  The late-day weakness could possibly be explained by the counter-party to this trade, as someone had to short stock to offset the risk of buying those 10,000 calls.  This is conjecture on my part (but at least it's educated conjecture).  I've spent a good deal of time trying to work out various scenarios where selling these calls would make sense, and they're almost all bearish.  Given the size and timing of the bet, it's a seriously negative sign for the short-term.

Adding to the negative tone from the options pits, the total put/call ratio remains neutral while the OEX put/call ratio continues to climb, today reaching the highest reading since early July.  This hints that retail traders are relatively sanguine while the bigger money is getting increasingly bearish.

The advance/decline line 10-day moving average dropped quite a bit today, as I mentioned it would since we dropped a reading of +2014 while adding a reading of -327.  We have two more positive readings to drop before we begin dropping a string of large negative readings.  This tells us that the oscillator will most likely drop over the next couple of days before making another attempt at overbought.

With the large drop in the A/D oscillator and the improvement in some of the weekly sentiment surveys and COT data, the Composite model jumped up to 64%, bringing it back to POSITIVE status.

Today was an inside day (lower high and higher low) on both the S&P and NDX.  The S&P had the narrowest range since June 28th while the NDX had the narrowest since March 27th (and December 24th before that), making it the third-smallest range in the past two years.  The NDX has now made two relatively low range days in a row.  I have been conditioned to believe (and anecdotal "evidence" confirmed) that narrow-range days preceded large-range days.  In other words, low volatility begat high volatility.  However, in looking over the past two years in the NDX, I checked the narrowest range days (defined as the bottom 1% of all daily ranges) compared to the rest of the days.  I defined the range as the true range (as opposed to simple range) expressed as a percent of closing price.

So, this table tells us the opposite of conventional wisdom - that low volatility actually begat more low volatility, at least over the short-term.  The major difference this time is implied volatility.  The average VIX reading during the bottom 1% of ranges was 22.7 (with a standard deviation of only 2.5), as opposed to 40 currently.  I'm not quite sure how the current high implied volatility will play out, but it makes me a just a bit less inclined to try to play breakout strategies from this narrow range we're in.

Our "sneak peak" lowrisk.com sentiment survey didn't show much of a move during the most recent reporting period, suggesting that the major surveys released later this week will also be relatively unchanged.

The results from the Fed meeting tomorrow will undoubtedly rule afternoon trading tomorrow, and if history is any guide the morning will be relatively quiet in anticipation.  If the Fed does cut rates, here's what's happened over the past year when the day ends up or down:

As you can see, even during a roaring bear market, when the Fed cuts, it's a positive short-term catalyst for the market a majority of the time.

- Jason Goepfert