Need to Relieve the "Pressure"

Bottom Line: A couple of the most positive days of the year are coming up, and it pays to be aware of them.

I’ve pointed out before that two of the most positive days of the year historically have been those immediately surrounding Thanksgiving. This phenomenon has been in effect for several decades, and it shows few signs of letting up. Since 1950, the day before Thanksgiving has been positive for the S&P 500 more than 77% of the time, with an average return of 0.4%, a maximum gain of 2.9% and a maximum loss of 1.9%. The day after the holiday has been just as positive, with 79% of the days closing higher than the day before. If one had invested $10,000 in the S&P 500 on the close two days before Thanksgiving and sold on the close the day after the holiday, that $10,000 would have turned into just over $15,000, with an amazing 87% of the trades being profitable to some degree. That’s a pretty good return, considering one would have money at risk only 2 days each year.

A common perception is that since seemingly everyone knows about these holiday biases, then traders likely buy earlier and earlier, trying to get a jump on the other suckers that come in later. I’m not sure why that perception exists, because it’s just not true. Market performance around this holiday has been very consistent for the past 50+ years, as the following charts show. It’s easier to look at this type of data than think about it, so the following charts show how the S&P 500 performed each year for the given days around Thanksgiving. On the charts, “Thanksgiving -3” would mean the trading day three days prior to the holiday, “Thanksgiving -2” would be two trading days prior, etc.

If we look at times when the day before the holiday goes against its historical pattern and closes negatively, then it doesn’t have much of an impact on the day after. In fact, the four times when the day before closed down by more than 0.5%, the day after bounced back three times for an average gain of 1.2%. The other time, it lost 1.5%.

We can see from the chart above that the day before Thanksgiving had a streak of nearly 20 years from the 1960’s to the 1980’s where it was positive without fail. Since that time, there have been 7 down days, some relatively large. However, the days before that haven’t become more positive during that time, which would have signaled traders were buying in anticipation of the positive seasonality. While the day before and the day after the holiday are typically positive, the day three days before and two and three days after are historically negative, though not massively so.

There is little evidence that the bias around this holiday is changing from its historical pattern, so although nothing is perfect, this is about as close as you’re going to get, at least with a one-input method. Buying before Thanksgiving and selling one day after has been a winning trade for many years, and there’s no reason to believe it won’t continue.

Bottom Line: The components of the S&P 500 and Nasdaq 100 are extremely oversold in the short-term.

Today’s Down Pressure reading for the Nasdaq 100 reached an extremely high 90%. This means that over the past three days, on average 90% of the total points that have been gained or lost by the component stocks that make up the index have been lost, and 90% of the volume has been going into those issues down on the day. This is a very rare example of persistent and heavy selling pressure. On a short-term basis, at least, this information can be interpreted in a contrary manner. Rarely is this type of selling pressure sustainable for much longer.

Since July of 2002, there have been only four other days that have exhibited a Down Pressure reading equal to or surpassing today’s. The table below shows those days, along with the percentage gain or loss in the NDX the given number of days later.

DATE	DOWN PRESSURE	1 DAY LATER	3 DAYS LATER	5 DAYS LATER	10 DAYS LATER
08/05/02	90%	5.2%	10.4%	9.5%	19.7%
11/11/02	91%	3.0%	8.8%	7.5%	15.9%
07/10/02	91%	4.1%	6.2%	7.3%	-0.8%
09/26/03	93%	2.0%	2.0%	5.0%	7.3%

The same indicator on the S&P 500 is at 84% tonight. Since the components of the S&P tend to be less volatile than those of the Nasdaq 100, and simply because there are more of them, the Down Pressure gauge for the S&P doesn’t quite reach the types of extremes that the indicator does for the NDX. Still, today’s reading of 84% is the fifth-highest since last July, and each of the four other readings higher than today’s had a positive return three days afterwards, with an average gain of 3.6%.

With only a handful of historical readings to rely on, one certainly can’t base a trading decision off of this information. Also, while we saw a similar extreme Down Pressure reading in September, this type of selling pressure is not usually seen in solid uptrends. This data tends to trend, as we see consistently high Down Pressure readings in strong downtrends, and consistently low readings in uptrends, which makes perfect sense. So, while the evidence looks promising for an oversold bounce over the coming days, the fact that we’ve seen this type of lopsided selling over the past few days – especially in the face of relatively positive news – is disconcerting to me.

In the intraday comments yesterday, I noted the extremely oversold nature of our shortest-term intraday indicators, and said that if the uptrend was still intact, then the downside should be limited over the short-term but if not, then that would be another clue that the larger trend is changing. We’ve also broken some short-term technical trendlines and are hovering very close to the widely-watched 50-day moving average in the S&P 500. We are perilously close to achieving all three of the failures I wanted to see before getting too anticipatory of a larger decline. The broader market must rally almost immediately, or the structure we’ve seen for 8 months will be broken. It sure feels to me as though this decline is different, mostly because the selling pressure has been very intense compared to what we’ve seen before, and the rallies have lacked “oomph”.

I’m also seeing some subtle shifts that for the most part had been absent. For example, yesterday, even though the NDX declined by 1%, assets in the Rydex NDX fund leveraged 2-to-1 to the long side rose by $27 million, and assets in the leveraged short fund dropped by $9 million. This type of behavior is not typical of what the Rydex traders had been exhibiting previously, where they were piling out of longs and into shorts at the slightest whiff of a decline. Also, we have not seen any major put volume increases over the past couple of days. In fact, put volume has been about equal to where it has been over the past few months, but call volume has actually been running about 20% above average. Both of these data points suggest small traders are now quite comfortable “buying the dip” that has worked so well so many times this year. That is not a positive sign.

While I am more interested now in possibly shorting a short-term rally should we get one for a longer-term trade, I am concerned about the positive seasonality we are in that I’ve discussed at length. It’s not set in stone that we HAVE to rally during this time frame, but it is a historical pattern that I am wary to fight. As a solution, if I do ultimately take some short positions, they will likely be lighter than normal. Unless this market turns around right quick and puts in some very good days, the upside prospects in the intermediate-term appear limited at best.

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.