Trend Change

There's something happening here

What it is ain't exactly clear

"For What It's Worth", Stephen Stills (1966)

The something that's happening in the markets, in contrast to the vision of Buffalo Springfield's song, is starting to become clear.

We appear to be in a trend change.

The bull and bear signal numbers out of my analytical universe of about 2,300 symbols tell the story:

On May 3 I found 132 signals from symbols traded on the NYSE, AMEX and NASDAQ -- the major American exchanges. That's a signal rate of 5.7%.

By last Friday, May 24, the number of breakouts had dropped to 39, for a signal rate of 1.7%.

Put in the scariest of terms, that's a 70% decline in the number of bull and bear signals in a period of three weeks.

The decline was underway from May 3 onward, with total signals declining from 132 to 37 by May 21, rising to 64 on May 22, and then declining again on May 23 and 24 down to 39.

My screening method picked up a change well before it showed up on the S&P 500 chart. The index closed on May 3 at 1614.42, set higher lows consistently as it rose to close at 1666.29 on May 20, hit an all-time high of 1687.18 two days later on May 22, and only then declined into the next day. On May 24 the index reversed for a rise intra-day that failed to set a higher high or a lower low.

To be exactly clear: The S&P 500 chart doesn't yet show any change in the daily-chart trend. To do that it would need to set a lower low by closing below the base of the current up leg. That' s 1536.03, the low set on April 18, a 9% decline before the all-time high.

In order to verify the downtrend, that lower low would need to be followed by a high below 1687.18 followed by a reversal to yet another lower low.

The weekly chart, which tracks the longer-term trend, shows the present up leg beginning on Oct. 3, 2011 from 1074.77, which would be a 36.3% decline.

But these things happen. Big declines aren't all that rare. Three weeks in late July-early August 2011 saw the S&P 55 decline by 18.1%. 

Mid-October to mid-April 2012 saw a five-week decline of 8.2%. That was the fall that immediately preceded the present leg up.

The recession crash from autumn 2007 to spring 2009 carried the S&P 500 down 57.7%.

These things happen.

As a trader, I'm gratified to see my methods picking up on major market changes early. The decline in the total number of signals tells me that something is happening, and the ratio between the bull and bear signals gives sense of the direction and magnitude of the change. 

I went from a bull/bear ratio of 6.8:1 on May 3 down to 1:5.5 on May 24. It was a steady decline on the bull side and rise on the bear side throughout the three weeks covered.

I think it will pay, going forward, to give attention to those numbers. Potential trades were still making it through my screening process, and so there was a chance that I would still be served bad trades that would cost me if the trend change in fact is happening.

A rule that says I only take trades when the bull/bear ratio is at a certain level in favor of the direction of that trade would make my system more robust: Maybe 3:1 for a bull trade or 1:3 for a bear trade. Those ratios would have halted all trades on May 16, resumed to the bull side on May 17, would have halted all trades again on May 20, and moved me to the bear side on May 22.

We'll see what happens. I'm not going to write this into a rule based on so little data, but I shall begin to include the ratio as part of my analysis and reporting.

One "gotcha" in this discussion is the fact that I'm analyzing a subset of all stocks traded on the major exchanges: Stocks tracked by the rating aggregator Zacks that have market capitalization of $1 billion or greater (mid-cap, or by some definitions, mid-cap and the top of small-cap). I have no limits on price or volume.

This means that the stocks in my universe have a following among some analysts -- even if it's the CEO's first cousin once removed. Because all or most of the small-cap stocks are cut out, the stocks I analyze will tend to be active and to give heavier weight to the most liquid issues -- the household names that dominate market coverage, the sorts of stocks, in fact, that show up in the S&P 500;.

So test the extent of the trend change I think I'm seeing, I applied the analysis to all 9,353 symbols in my major-exchange database. On Friday, May 24, there were 127 signals, 39 to the bull side and 88 to the bear side.

The signals account for 1.4% of the total universe of symbols analyzed, quite close the 1.7% figure obtained by my more limited list.

The narrower list of stocks -- my analytical universe -- was more bearish than the market as a whole: 84.6% of signals were to the bear side on the short list, compared to 69.3% on the whole market.

The bull/bear ratio for the whole market was 1:2.3; for my analytical universe, it was 1:5.5.

Those results are close enough to persuade me that my analytical universe is a fairly good avatar for the whole market. What's happening here, whatever it may be, is a market-wide phenomenon.

References

My trading rules can be read here. And the classic Turtle Trading rules on which my rules are based can be read here.

Disclaimer

Tim Bovee, Private Trader tracks the analysis and trades of a private trader for his own accounts. Nothing in this blog constitutes a recommendation to buy or sell stocks, options or any other financial instrument. The only purpose of this blog is to provide education and entertainment.

No trader is ever 100 percent successful in his or her trades. Trading in the stock and option markets is risky and uncertain. Each trader must make trading decision decisions for his or her own account, and take responsibility for the consequences.