11 in a Row, There's More to Go (?)

http://www.sentimentrader.com

11 in a Row, There’s More to Go (?)

Monday, December 29, 2003 9:33pm EST

11 in a Row and Counting…

Bottom Line: Long streaks of higher lows in the S&P 500 have preceded further gains over both the short- and long-term.

For the 11^th straight day, the S&P 500 (cash index) has put in a higher low than the day before. Just to be perfectly clear, this means that the low today was higher than the low the day before, which was higher than the day before that, etc. Such a streak is highly unusual in that there have been only 11 other such occurrences in the past 40 years, the last one being a string of 13 consecutive higher lows in 1998. Interestingly, when the S&P has gotten as far as it has now, it has tended to keep going. The average length of the streaks, once they reached 11 days, was just under 13 days - only three out of the eleven other occurrences stopped once they reached 11 days.

While the initial reaction to this is logically “the index is extended, and due for a pullback”, which was certainly my impression, history does not support it. In fact, the evidence points to quite the contrary. While we only have 11 occurrences to reference, in the past when the S&P has put in such an impressive string, it has portended very positive performance going forward. The table below outlines each past occurrence, showing the date that each streak ended, the number of consecutive higher lows that ultimately were in the streak, and the percentage change in the S&P 500 the given number of days later.

DATE	STREAK	5 DAYS LATER	10 DAYS LATER	20 DAYS LATER	30 DAYS LATER	60 DAYS LATER
01/16/64	15	0.7%	0.2%	1.3%	1.6%	4.2%
07/21/67	13	0.5%	1.9%	0.8%	-0.4%	1.3%
04/18/68	17	-0.5%	1.6%	0.5%	1.6%	2.2%
12/08/70	12	0.2%	0.6%	3.3%	5.3%	10.6%
03/09/72	13	-1.3%	-1.1%	0.6%	0.0%	-0.1%
04/22/83	12	2.5%	3.5%	1.1%	2.7%	2.7%
11/12/85	11	0.3%	1.3%	4.2%	4.8%	8.3%
01/16/87	11	1.4%	2.9%	5.0%	6.3%	7.3%
05/15/90	11	1.2%	1.9%	3.0%	0.2%	-4.0%
09/15/95	12	-0.3%	0.2%	0.2%	-0.6%	6.2%
02/11/98	13	0.8%	2.8%	4.9%	7.9%	8.6%

	Avg Ret.	0.5%	1.4%	2.3%	2.7%	4.3%
	% Pos.	73%	91%	100%	73%	82%
	Max	2.5%	3.5%	5.0%	7.9%	10.6%
	Min	-1.3%	-1.1%	0.2%	-0.6%	-4.0%

From the table, we can see that even though this broad market index was probably extended in the short-term by most measures, it did not lead to weakness in any time frame studied. Five days after these streaks ended, the S&P was still higher 73% of the time. Going out to 20 days, all eleven instances showed a positive market, with the S&P being up an average of 2.3%. Even after 60 days, the market was higher more than 80% of the time, with a maximum gain of over 10% and maximum loss of 4%.

When we have long streaks like this, positive or negative, natural human emotion tugs at us and suggests that it can’t go any further. While there are plenty of reasons for it not to go further, if we just look at how the market has responded in the past when we have seen a setup like this, then the conclusion is counter to what most of us would initially think. Eleven straight higher lows in a broad index like the S&P 500 has portended some great gains over the subsequent weeks, and very few losses.

Selling Pressure

Bottom Line: There was none. Historically, this has lead to some short-term consolidation or declines but doesn’t tell us much beyond that.

There was absolutely no selling pressure today. That can be concluded from many different angles, with one of the most widely-available being the TRIN. I’ve showed many times before the weaknesses of using the TRIN to view widespread buying or selling pressure (namely the ability of one or two low-priced stocks to greatly skew the number), but generally it is still a relatively effective proxy. Also, today’s reading is confirmed by our daily Down Pressure reading for the S&P 500, which was 2%. This is computed by taking an average of the percentage of total points lost by the component stocks in the S&P with the percentage of volume going into the stocks that closed negatively on the day. For example, today I show that the stocks that rose on the day gained a total of 235 points. Those few that declined lost a total of 4 points. So, only 1.7% of the points gained or lost today were lost (i.e. 4 / (4 + 235) = 1.7%). I show that there was 1.1 billion shares traded on the stocks that rose on the day, and 39 million shares traded on the losers. So, about 3.5% of the volume today went into down issues. When we average the two percentages (1.7% of points and 3.5% of volume), we get a Down Pressure reading of around 2%. Past occurrences of such low readings coincide with low TRIN figures from those days.

Today’s TRIN for the NYSE closed at 0.35, one of the lowest readings seen over the past few years, with the only lower readings this year being 0.22 on March 17^th and 0.24 on January 2^nd. The reason the TRIN was so low is because there was a minimal amount of volume in those stocks that closed negatively. Volume going into down issues today amounted to only 129 million shares, the lowest amount of down volume since January 2^nd. This is qualified somewhat, however, since these types of readings are most commonly found on the day immediately before or after a holiday. For example, the only other days with lower down volume than today over the past few years were the days before Thanksgiving 2000 and 2001, the day after July 4^th, 2002 and the day after New Year’s 2003. The results following those other low-down-volume days were mixed, so it’s difficult to infer too much from those instances.

One other measure that shows buying and selling pressure is the TICK. The TICK shows the net number of stocks that last traded on an uptick minus those that last traded on a downtick. If IBM traded at 93.50, then traded again at 93.53, then it would be considered an “uptick” since that last trade was at a higher price than the previous trade. If you have 1,500 issues on the NYSE that last traded on an uptick and 500 that last traded on a downtick, then the TICK figure would be +1,000. I keep close watch on these figures, as they can be excellent guides as to buying and selling exhaustion, though it has gotten more difficult lately with program trading accounting for 40% of total NYSE volume.

The intraday cumulative TICK indicators that I show on the site sum up TICK readings taken every ½ hour. What is notable is that I have not recorded a negative TICK reading for 65 periods – the equivalent of around 5 full trading days. This does not mean that there haven’t been ANY negative TICKS, because of course there have been. But when I take my readings, every one of the last 65 has been positive. Since I began recording this data over three years ago, there has not been a longer streak of positive TICKS than right now. The closest we’ve come to our current streak was 53 straight half-hourly periods that ended on September 2^nd of this year.

After days with very lopsided breadth (either up or down), it usually pays to look in the other direction over the next couple of days as traders digest their gains or losses. Today certainly qualifies as lopsided, as breadth was the most positive we’ve seen in three months. Since June, any day that the A/D figure was +1,500 or higher (meaning that at least 1,500 more stocks rose than declined on the day), lead to mixed performance over the next few days. The S&P barely showed a positive return 1, 3 and 5 days later, with 5 out of 9 instances being positive. This reinforces the study I showed several weeks ago that suggested that it is better to be a seller after very good days, and a buyer after very bad days, at least in the short-term.

Conclusion

On December 17^th, I showed a five-day average of the equity put/call ratio with QQQ options removed, and noted that it had fallen below 0.45. Every time the ratio had become that low in the past six months or so, the market had seen some type of weakness. This time around, however, it made no difference as the S&P has powered higher by about 3%. As of today, the ratio is back below 0.45, and in fact is lower than where it was on the 17^th. In addition, the 10-day average is now about 10% below where it was then, and the 21-day average is about 8% below where it was – meaning that not only have we seen extreme emphasis on calls versus puts in the short-term, the longer-term measures are now also extreme. At 0.48, the 21-day average is now the second-lowest it has been in years, with the lowest occurring near mid-June of this year.

It seems silly to talk about the risk of complacency when the market has had nary a significant correction for nine months, and the biggest (opportunity) risk has been not being aggressively long. I always cringe when I hear someone say “it may not seem like it matters now, but it will at some point” – so I’m not going to say that. I will say that it would be highly unusual to see more significant upside over the next couple of days, based on the market’s very consistent pattern of digesting large moves. However, year-end can throw a wrench into historical patterns, and the positive seasonality we have coming up makes a lot of traders hesitant about betting too heavily against the market.

My longer-term thoughts have not changed, as I continue to believe it’s just as likely we will be lower than recent prices a month from now as higher. When I see that kind of toss-up, I prefer to shorten my time frame and keep longer-term positions very light.

- Jason Goepfert

Disclosure: long QQQ puts, long QQQ

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.