Wednesday, July 02, 2008

When Trending Trends and When It Ends

If I had to describe my trading approach in a nutshell, it would be to identify what markets have done historically under a set of conditions and then observe whether or not this scenario is playing itself out in real time. We've had an oversold market that, based on historical precedent, should have led to a sharp reversal. Tracking such indicators as the Cumulative NYSE TICK, however, it's been clear that buyers have been absent and that the market has not been following its historical script. Some might call that a failure of historical indicators; I call it useful, practical information.

Above we see how the indicators have been weakening (kudos to Decision Point for the charts). New lows among NYSE common stocks have expanded beyond their March levels (top chart; click for greater detail); the advance-decline line for NYSE common stocks (bottom chart) has also been making new bear market lows.

Notice in the bottom pane of the second chart that the volume-weighted advance-decline line for NYSE common stocks has been much weaker than its traditional counterpart. This suggests that stocks are falling on higher volume than they're rising: just another sign of institutional bearishness.

When indicators and sectors are in gear, we have a trending market. Distinguishing trending markets from ones that are likely to reverse is one of the great challenges of trading at any time frame. Knowing what markets "should" do based on precedent--and observing what they *are* doing--is helpful in making the distinction.


Indicator Review

Sector Strength


CharlesTrader said...

Your comment about the market not following the historical script reminds me of a suggestion once told to me by a more experienced trader.

When I was trading commodities, such as the grains, I religiously kept track of statistical, seasonal trends. The natural psychological tendency is to look for commodities that are following their expected seasonal trends. The more experienced trader, however, told me that what I really should be looking for are commodities that are doing the opposite of what is expected as dictated by the statistical analysis.


Joe said...

Hi Brett!

I am Joe from Hungary.

Amidst the fear in the market and a strong series of selloff Finally my predictor gave me
a much stronger thean Avg Bullish signal for Today, Wednesday.

All of my predcited indexes


Above average Bullish for the day,
relative to previous Close

It is really a question how this will play out, and how potential gains in the morning session will hold in the late afternoon, and for how long.
My predictor is supporting my Day -Trading and overnight trading activities and it is for one Day only.


SSK said...

Thank you for that insightful comment. Steve

Zlatko said...

This is purely anecdotal, but trends down are different from trends up. I mean, haven't you ever looked at a chart upside down and thought "This isn't what a chart of a stock price looks like". (I have).

What is remarkable about downtrends is that they often end in range expansion. Not so anecdotal are some back-of-the-envelope calculations that show that all daily ranges of 5%* or more in the S&P500 have been ends of intermediate-term trends (although not vice versa).

We haven't seen much range expansion lately.

*I calculate daily range in percentage terms like this: (Hi - Lo)/(Hi + Lo / 2) - basically, daily range divided by median price

Brett Steenbarger, Ph.D. said...

Hi Charles,

I think you're right: instead of simply mechanically trading historical patterns, it makes sense to differentiate when markets are following vs. not following those patterns and trade accordingly. It's a very different implementation of market research from either a purely discretionary or a purely mechanical approach.