http://www.sentimentrader.com/subscriber/subscriber_home.php
THURSDAY, AUGUST 16, 2007
Now Can We PLEASE Get Some Follow-Through?!
08/16/07 5:00 PM EST
There are only so many words that describe what we saw today, like "panic" or "fear" or "capitulation", so they become overused and cliché-ish. I try to avoid using any them because of that, but today is an exception.
One of the best ways to measure uncertainty is to see how much money is flowing out of the riskiest assets and into the safest. And there is no (perceived) safer asset than short-term U.S. Treasury Bills. We can easily see the money flows into them simply by watching their yield (a good proxy we can watch intraday is the CBOE short-term yield index, ticker IRX).
Over the past two days, money flows into those Bills has been incredible. Since yields move in the opposite direction of prices, when money flows into these short-term Bills, it pushes prices higher - and yields fall. Over the past two days, those yields fell somewhere in the vicinity of 20% by mid-day today.
That is an incredible move. Depending on what data you use, it may be nearly unprecedented. Using data provided from the Federal Reserve from the 1950s, I can find only two other instances - immediately after 9/11, and late February 1958. Both instances marked the end of a severe leg down in stocks (or very close to it), and by three months later the S&P 500 was about 8% higher both times with very little drawdown.
There were multiple signs of capitulation today. The best is perhaps the number of stocks on the NYSE that hit new 52-week lows. By today's close, that number swelled to 1,121 according to my quote vendor, which is second in history to the bottom of 8/31/98 when 1,183 stocks hit a new low.
When we express that number as a percentage of all stocks that traded today, it exceeds 33%. That's a truly incredible number that has been matched or exceeded only three times in the past 20 years: 10/19/87, 8/23/90 and 8/31/98. All three marked the end of the panic phase of the decline as stocks rocketed higher in the days following, with the S&P higher by +10.4%, +4.7% and +2.6%, respectively, three trading days later. Prices came back down to test those lows each time within the next month or so, but they still marked important inflection points.
I mentioned yesterday and again this morning that I was expecting any move down towards 1375 - 1380 to mark the end of this phase of the decline, and we overshot that for just a few moments and just a couple of points before the reversal this afternoon. Combined with all the other stuff we looked at during the day today, that makes me more confident that that area is a place of major support.
Given the big reversal bar that is going to look so pretty on traders' charts as they study them tonight, I suspect we're going to see more buying interest as the days wear on, and the impact of options expiring perhaps adds a bit more fuel to the fire.
Have a great night and we'll see you tomorrow!
Scramble for Safety at "Capitulation" Levels
08/16/07 2:15 PM EST
As I mentioned in my prior note, the STEM.MR Model was inching its way towards "maximum" oversold at 90%, and it's about there now. The same model for the Nasdaq 100 has also entered extreme territory.
There are a number of interesting developments today. The yield on short-term Treasury paper that suffered such a big decline yesterday is showing an even larger drop today as traders and investors scramble for what is viewed as the ultimate risk-free asset.
I touched on this before today's open, and the situation has become only more extreme. In the past 50 years, I can find only two other time periods when short-term yields declined this much over a two-day period: right after 9/11, and February 24, 1958. The latter marked the end of that year's correction and the S&P was up nearly 8% over the next three months. That's about the same return seen in the months following the 9/11 tragedy.
I was also intrigued by an alert from my broker earlier this afternoon. They suggested that clients URGENTLY re-assess their risk if they are holding futures or options on the VIX. Due to the increase in volatility, they cannot be sure what is going to happen, but they believe that margin requirements for VIX spreads are going to double or triple at today's close.
Along with the CME's announcement earlier that margin on some futures contracts will be increased today, we might see some volatility as traders are forced to re-adjust their portfolios (or brokers do it for them...). Any such adjustments would likely be worked through by tomorrow morning, however.
I cannot get around the fact that these types of announcements are so often made near turning points, and it lines up with everything else I'm seeing. Like the fact that the number of stocks hitting new 52-week lows on the NYSE spiked to over 900 for only the second time in over 10 years, with July 24, 2002 being the only other day with that high of a number.
I mentioned earlier that I expected the S&P 500 cash index to bottom around 1375 - 1380 at its worst today. It briefly dropped a few points below there before reversing, but it has now made a fierce comeback this afternoon. If the reversal sticks, we'll surely hear all about the reversal day and the number of extremes we've seen, and I suspect that will help fuel a further recovery in the days ahead.
Panic Readings Beginning to Register
08/16/07 12:15 PM EST
Last Friday, I mentioned that commercial hedgers in the VIX implied volatility futures had established a new all-time record long position. I thought that was very curious, and unfortunately troubling, since it suggested that the smartest traders were betting on an even bigger rise in volatility (which usually goes hand-in-hand with lower stock prices).
Since last Tuesday, which was the cutoff for reporting those positions, the VIX has risen almost 70%, so it looks like they knew something yet again. Today's high of 35.29 on the VIX is the highest since March 2003.
We have a fairly limited history of that index, going back to about 1990. But since that time, any time the high of the VIX hit at least 35 on the same day the S&P 500 was hitting at least a four-month low, the S&P was higher three months later 18 out of 20 times. Many of those days were clustered together in late August 1998, mid-September 2001 (after 9/11) and late July 2002.
Out of those 20 days, there were really 7 distinct time periods covered. The day like today marked the exact low or very close to it four of those times, while the three mentioned above were the ones that led to more short-term pain before the recovery began.
We're seeing quite a few readings now that compare favorably with those time periods according to a number of different measures. They strongly suggest that prices will be higher when looking at the intermediate-term, but the short-term timing is questionable. The safest bet would be to hold off on any long positions until a sign of recovery took place, since in all of those three cases, prices simply kept hitting new intraday lows on almost a daily basis until we bottomed. Waiting for a day where we didn't make a lower intraday low was a fairly good hint that the bottom was in.
Another indicator that just triggered a comparison to those times is our Intraday Cumulative TICK. It recorded a reading of -7000 today, which has been matched only twice: 9/21/01 and 7/24/02. Both notable dates.
With that, the STEM.MR Model is approaching "maximum oversold" as it inches closer to 90%. It has never traded higher than 91% in the five years of history that we have. I like to wait for that model to begin curling higher before considering it a buy signal, and it should do so this afternoon as it would obviously be enormously unusual for it to go beyond 90%.
Adding to the evidence, my quote vendor is showing 880 stocks on the NYSE currently hitting a new 52-week low. That is the 2nd-most in the past decade, next to 7/24/02 when it hit 938.
I mentioned this morning that I was looking to add some long exposure on the gap down open, and another leg should we trade down to 1375 - 1380. That is still my plan, as we are now getting a fair amount of "puke" readings as we approach that area. I will look to add yet more exposure should we close near the day's lows, then I would be holding off until we got more signs that an actual price recovery was taking place (such as seeing a day where we don't make a lower intraday low). Risky stuff, for sure, but I have to go with what the historical evidence suggests is the highest probability thing to do.
Remarkable Times
08/16/07 8:45 AM EST
Good Thursday morning...we begin the day with what is indicated to be another large gap down opening as foreign markets feed off yesterday's U.S. sell-off. That in turn impacts pre-market futures here in a vicious circle that will only end by a solid rally in the U.S.
Yesterday's trading brought with it yet another round of extraordinary readings. Breaking to new lows tends to generate a few extremes, and this time was no different.
With option expiration looming, we're likely to see several extremes in various put/call ratios, and the past couple of days have not disappointed. The Equity-only Put/Call Ratio from the Chicago Board Options Exchange closed above 1.0 for the past two days. The last time we saw more equity puts than calls traded for two straight days was in February 2003, and the only other instances were right after 9/11 and in mid-September 2002. None of them coincided with the exact market low, however, as stocks continued to slide for several days to a few weeks afterwards.
One reason for these abnormally high p/c ratio readings is the fact that we've fallen so hard coming into expiry. Since the options markets began, I can find only four times that the S&P 500 declined at least 8% over the past month, then at least 3% coming into expiration. Those instances were all notable: March 2001, September 2001, July 2002 and September 2002. If you pull up the third Friday of each of those months, you'll see that they were very near major market lows.
Just waiting a couple of days, then buying and holding for a week resulted in very positive performance; holding for a month resulted in positive returns all four times, and by an average of a resounding +11.3%. The average maximum drawdown during that month was fairly stiff at -3.9%, though it pales in comparison to the average maximum gain of +12.1%. All of the drawdown occurred in the short-term.
We're also seeing some panic from Rydex traders. The Rydex Beta Chase Index has spiked down to 0.3, meaning that investors in those funds are more than three times as likely to invest in a "safe" fund than a "risky" one. That indicator has been more extreme over the years, but it's at least worth a mention that the last two times were July 18, 2006 and March 6, 2007. Those instances of extreme risk-aversion resulted in a market low soon thereafter.
Also among the Rydex funds, only 2 out of 33 sector funds are showing assets greater than their 50-day average. That's evidence of wholesale selling pressure, and it's gotten to a degree last seen way back in mid-March 2003. It got lower during the teeth of the 2000-2002 bear market, but again it was fairly accurate at hinting that we were close to being washed out enough to rally.
Something else notable about yesterday's session comes from the Treasury market. According to Bloomberg, the yield on three-month T-Bills fell the greatest amount since October 1989. That kind of rush to safety is obviously very rare, so I checked historical precedents going back more than 50 years using the Federal Reserve's data on secondary market rates for 3-month Bills.
Looking for any day where the yield on those instruments fell as much as they did yesterday, while setting a two-year low in the process, I came up with 18 days. Going out one month later, the S&P 500 was positive 15 of those times, for an average of +2.7%.
I mentioned yesterday that a move down to 1375 - 1380 is being mentioned by a lot of traders using different methodologies as a likely support area. Because it's being talked about so frequently, I suppose that means we're either going to overshoot it to the downside or undershoot it by bouncing before we get there. I tend to think we're more likely to bounce before we get there, but I hate that game of trying to outguess the guessers based on nothing but anecdotal evidence. I'm going to go with the idea that that general area, if we get there, is a place where I'll likely want to look to more aggressively establish short- and intermediate-term longs. Let's cross that bridge if/when we get to it.
There are a bunch of things coming to a head today, mainly dealing with options expiration. I mentioned yesterday that according to several sources, market makers are short a record or near-record number of put options, which could put additional pressure on the market today if traders continue to buy puts, but by tomorrow those index options will be expiring and those market makers' hedges will have to be unwound - meaning an added source of buying pressure in the index futures.
I have also heard reports that the CME will be raising margin requirements on some index futures today, which could generate a round of margin calls should the limit be raised a significant amount. I have not been able to verify that with the CME, however, so at this point it's just a rumor.
This is the first time since July 24, 2002 that the S&P is indicated to gap down at least 1% on the open after yesterday's close hit at least a four-month low. I can find 9 other such instances in the history of the S&P 500 tracking stock, SPY. Five of those occurred in the waterfall decline during June and July 2002, but still if we had bought that opening gap and held for three trading days, we would have had 8 out of 9 winning trades with an average return of a whopping +4.3%.
Whenever I send these notes out before the open, the stats can obviously change if pre-market futures recover. But there's a fine line between being too early and not being early enough (the stats don't do as much good after the fact...), so I like to err on the side of being too early.
I'll be looking to add some long inventory with this gap down open, though I will by no means be going leveraged or crazy-long. If we would happen to trade down to the potential support area I mentioned earlier, then I will look to add another leg of exposure. This is a very fluid market, things can change on a dime, but that's my plan of attack as it stands right now.
All the best,
Jason Goepfert
President and CEO
Sundial Capital Research, Inc.
Forwarding or otherwise distributing this copyrighted material is a breach of your subscriber agreement. Violators are subject to termination of their subscription with any received subscription fees forfeited. Any references to historical performance are based on data we deem to be reliable, but are based upon feeds from third parties. We do not recommend subscribers take positions based on data presented here alone, but rather incorporate it into a comprehensive investment outlook.
© 2007 Sundial Capital Research, Inc. All Rights Reserved.