Inflection Point

Bottom Line: The type of breadth readings we’ve seen the past two days have usually marked lows – or the beginning of waterfall declines. The action over the short-term should tell us which.

As you have no doubt heard by now, the selling yesterday was some of the most indiscriminate we’ve seen in a very long time – in fact, we would have to go back to October 1997 to see a larger number of stocks that declined on the day. It’s rather fruitless to look at the absolute number of declining issues over time, because the total number of stocks traded has of course increased. If 2,000 stocks decline today when there are 3,500 total stocks traded on the NYSE, that’s a lot different than, say, 1975 when there were half as many. So instead of looking at the absolute number of decliners, I think it’s much more instructive to look at them as a percentage of total stocks traded. When viewed in that manner, yesterday came in with 72% of all stocks being down on the day. That is still quite remarkable – over the past 40 years, or 9,847 trading days, yesterday’s percentage of declining stocks ranked #34.

These types of days bring out two distinct opinions depending on whether the speaker happens to be bullish or bearish – either it is evidence of a washout move that will allow the market to rally, or it is a sign of serious selling pressure that signals more selling to come. Both are nothing more than opinion, so if we look at the facts of what these days have typically preceded, we can get a better feel for who may be most likely correct.

Selling Pressure S&P 500 Performance After 72% of Stocks Decline on the Day 1965 - 2004
	3 Days Later	10 Days Later	30 Days Later	60 Days Later	90 Days Later	120 Days Later
After 72% of all stocks declined on the day…
Avg Return	1.4%	2.1%	1.9%	5.5%	7.6%	8.6%
% Positive	64%	79%	70%	82%	73%	76%
After an average day…
Avg Return	0.1%	0.3%	0.9%	1.8%	2.8%	3.8%
% Positive	54%	56%	60%	62%	63%	65%

From the table, we can see that the S&P typically outperformed an average day after seeing the type of broad selling pressure that we saw yesterday. After 10 days, the index was an average of 1.8% higher than an average day, and was positive 23% more of the time. Even after 60 days, the S&P showed a return 3.7% greater than average, and was positive 20% more of the time. After 60 days, the index was positive 82% of the time, as 27 out of the 33 instances were positive. There were two notable exceptions to this rule. The first one occurred in July 1974, as the S&P was 22% lower after 60 days. In October 1987, the index was 13% lower. The four other losses averaged 1.7%, so the damage was minimal other than those two. This compares to an average gain of 8.2% from the 27 winners.

The action today complicates matters somewhat (or simplifies them, depending on your perspective). The number of decliners on the NYSE today amounted to 55% of all issues traded – another fairly rare occurrence. Even rarer is to see two days in a row of at least 55% of all issues being down on the day, which we have seen over the past two days. Over the past 40 years, this has only occurred 20 times. While my natural instinct was to believe that if we saw one such day, like noted in the table above, and it lead to a very positive market, then two such days in a row probably lead to an even more positive market. It turns out that that is not the case. When we saw two days in a row with a large percentage of declining stocks, it underperformed an average day. 60 days later, the S&P showed an average return of 0.2%, with 60% of the occurrences being positive.

In looking at the data, however, something notable popped out - how the market performed the first day after these two lopsided selling days had a good deal of accuracy in forecasting what was in store later on. If we look at those times where day following the two consecutive selling days was positive or negative, it appeared to have a great impact on whether the market would be higher 60 days later as well.

If the day after these two days was positive, then the S&P was higher 60 days later 9 out of 12 times, for an average return of 3.3%. If the day after was negative, however, then the S&P was higher 60 days later only 2 out of 7 times, with an average return of -4.9%. This “day after indicator” was the best predictor of future returns - better than looking at the first 3 days, 5 days or 10 days.

One reason is because this type of selling pressure on consecutive days tended to coincide with either the very low of a move, or the middle of a severe decline. For example, the last two instances were 7/23/02 and 3/30/94 – both of which nailed significant market lows. However, we also saw instances on 10/15/87 and 4/24/74 – both of which occurred during the middle of extreme selloffs.

If the market was not able to immediately recover from this severe selling pressure (such as in 1987 and 1974), then it was a good tip-off that there was quite a bit more selling in store over the intermediate-term. If these selling days marked the low (such as in 2002 and 1994), then it boded very well for the next few months. I never like to give too much importance to a single day, but it does seem as though tomorrow, or the next couple of days at least, may be an important tell as to whether we are in the middle of a decline or near the end.

I want to note one other thing about today’s breadth. While the number of declining issues far outweighed the number of advancers, down volume was not nearly as skewed – so even though there was widespread weakness today, the TRIN closed at 0.65. That is a reading more closely association with overbought conditions than oversold, and is one of the reasons I have picked on the indicator from time to time. What today’s TRIN tells us essentially is that the volume flowing into all those down issues was not as dramatic as the negative a/d figure would have us believe. I checked every other time where declining issues were at least 55% of the total number of stocks traded, and the TRIN closed at 0.75 or below. Interestingly, the S&P 500 was higher 60 days later 11 out of 12 times, and with an average gain of 6.9%. After 90 days the average return climbed to 8.7%, again with 11 out of 12 being positive. After 120 days, the return climbed again to 10.1%, though one of the instances slipped into negative territory.

Bottom Line: A cumulative TICK measure is registering the most oversold reading since the low in October 2002, and underscores the importance of the next few days.

Another one of the remarkable notes about yesterday is the action in the TICK indicators. Recall that the TICK is simply the number of stocks that last traded on an uptick minus those that last traded on a downtick. It is calculated for the NYSE and the Nasdaq, updates continually throughout the day and typically ranges from -1200 to +1200. Yesterday the NYSE TICK barely traded above the zero line all day, but touched -1000 at least a dozen times. Today wasn’t a whole lot different for much of the day, and this activity has pushed our indicators that track that information to true extremes.

On the site, I update a cumulative TICK for both the NYSE and Nasdaq, which is really a summation of the last day’s activity. I also keep a similar cumulative TICK, though it takes a longer view and looks at the past three days. This indicator is now the most extended to the downside it has been since the low in October 2002, and the low in July 2002 prior to that.

I am always troubled when such overwhelming selling indications come after rallies and close to new highs, and I think it underpins the importance of the coming week. If we cannot rally off of these readings, then it would seem increasingly likely that this is an initial wave of selling that will be followed by still more.

I often read about certain times being “critical” for the stock or bond markets, and I think most of it is hyperbole. But from some of the readings we’ve seen over the past couple of days, it really does seem as though how the market acts over the next few days could hold the key to where it ends up several weeks or even months from now. I have been saying that the readings we saw last month, and how the broad market reacted off of those readings, were consistent with an intermediate-term low – meaning we shouldn’t violate those lows anytime soon. I still believe that, but a few things over the past few days, such as the comments above, and the performance of REITs that I wrote about on Monday, have me back on my heels a bit. I’ve been looking at the 1120ish area on the S&P as a likely place for that index to reverse higher if it was going to, and we did indeed stop near there this afternoon. A violation of that general level will have me tempering my longer-term bullishness to some degree, but if we decline below the March lows, that would be the ultimate (and obvious) signal that something is very amiss with the uptrend.

In the short-term summary on the Daily Overview of the page yesterday, I noted that a gap down open this morning would likely lead to something of a bounce, given how far many of our short-term measures had become. We did get a bounce, and it was enough to relieve most of the indicators of their oversold conditions. We’re back to neutral again this evening, and once again I don’t have a high degree of confidence with a trade in either direction. We should see more of a rally in the coming days, and that is what I will be looking for, but I am leery of a failure here. If we cannot hold higher prices, we may well be in for a tough few weeks.

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.