In making investment decisions, you are inevitably drawn to what has served you well in the past.  No matter the particular strategies that you employ, you fall back on your experience to judge how to proceed.  That works very well when past is prologue and trends are long, but not well at all in disruptive environments.

This posting was originally conceived in response to the current challenges faced by those whose investment process is wholly or partly related to technical analysis, so let’s start there.

The universe of market players who use technical analysis to some degree is broader than most people believe.  It is not uncommon to be talking to a portfolio manager who stresses how he makes decisions on the fundamentals, when over his shoulder you can see a chart up on his screen.  “Just for perspective,” he might say if asked about it, but aren’t such conceptual frameworks what technical analysis is all about?

Much of it involves seeing “levels” of one kind or another in one way or another, and making decisions in response.  Once electronic delivery of charts became ubiquitous, the technicals crept into the consciousness of fundamentalists too, so that a statement by someone that a stock is cheap may be some combination of a valuation assessment and a picture floating around in her head.

For avowed technicians, there are levels everywhere.  Sometimes they are clearly and cleanly represented in graphic form, but often these days they are to be found, if you can find them, in the midst of a chart with so many lines that it looks like a picture of multi-colored spaghetti noodles.The Crosshairs Trader | This posting does a nice job of looking at simplicity versus complexity in charting. The goal in this posting is not to describe or compare the formulas and techniques, or to debate their efficacy, but to consider whether they can be applied as they have been in the past.

One thing that interests me (as someone who evaluates processes) is the precision of the levels observed or predicted.  I often see notes that say something like, “35.42 is key,” with recommended actions to take if that level is violated.  How precise or fuzzyBarron’s | Michael Kahn referred to “fuzzy lines” in this piece. should your decision rules be?  False precision is unlikely to help in any way, no matter your approach.

But what of the levels in general?  Do the old rules still apply?  Can you rely on them?

It is a question not limited to technical analysis.  I reviewed a quantitative research report last week, the kind that shows various metrics as they are today versus where they have been in the past.  The bottom line on the report was “strong buy,” based upon a time-tested algorithm, yet when I looked at the individual variables, I found myself saying, “No longer valid, temporarily elevated, distorted by transient factors, etc.” as I went through them one by one.

Quant portfolio managers are struggling too, according to Pensions & Investments, and making changes in an attempt “to reverse their performance drought since the start of the financial crisis.”Pensions & Investments | Entitled, “Quant firms tinkering with factor weightings,” the article is restricted to subscribers. You see it everywhere:  The old relationships have broken down.

So, perhaps this is the time for fundamental analysts and portfolio managers to shine.  After all, if the downside of that approach is that it is easy to be led astray by your emotions, the upside is supposed to be the ability to navigate uncertainty by spotting opportunities not yet apparent in the charts and the formulas.  Yet, fundamental analysis is also tied to history and decisions are formed in light of it.  Look at twenty years of data on margins or sales growth or whatever, and tell me which levels are normal or valid or reasonable today.

If you work at this blogging thing long enough you repeat yourself a lot, and I knew when writing this that I was revisiting old ground.  Seventeen months ago, I wrote about our analytical desire to get “back to normal.”the research puzzle | The posting included a graphic that remains appropriate today. Despite a better economy and higher prices for most things, normality seems out of reach.

I see levels, but what do they mean?