On Using Technical Analysis

There’s been a broad discussion in recent weeks about the efficacy of technical analysis in investment strategy (see here & here).  I’ve touched on this briefly in the past (see here), but my position is rather simple and I think it’s a position most people should adopt, because, obviously, you should adopt my views of the world!  Just kidding of course.  You’re free to do with my thoughts as you wish and I am only here to provide one guy’s perspective and not pretend to offer some holy grail view of the world.

Anyhow, much of the confusion on this discussion starts with definitions. Technical analysis often gets a bad rap for being labelled as charting.  But they’re two different, but related things.  Charting is the use of chart reading in various ways to formulate strategies.  Technical analysis, on the other hand, is simply the use of past data to analyze future market direction.  Of course, a chart is merely a picture of past price action so charting is a subset of technical analysis, but does not comprise the universe of technical analysis – some of which can be extremely complex and sophisticated.

I find that understanding the past is an essential element in any good form of portfolio construction.  Perhaps you study the past to conclude that market timing is silly.  Or perhaps you study the past to conclude that buy and hold is silly.  But in the end, what most of us end up doing is essentially a branch of portfolio construction that begins with understanding past performance.  Of course, past performance is not indicative of future returns and while history rhymes, it rarely repeats perfectly.  So understanding past price performance and the history of price action is merely one building block in the development of a portfolio.  But that doesn’t mean it is a useless piece of the puzzle.  Whether you’re a trader or a buy and hold investor it’s useful to understand technical analysis and past price performance in order to better understand how one should go about attacking the future.

Of course, I am a very fundamentally driven analyst so I take the view that technical analysis is a good complement to good fundamental analysis, but to each his own.  In sum, keep an open mind.  There’s no holy grail to the world of portfolio construction and understanding and embedding many different approaches into your own will only make you a more well-rounded and informed investor/saver.

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services.

More Posts - Website

Follow Me:
TwitterLinkedIn

  • GLG34

    I still find it so odd that there are so many definitions in the world of economics and finance that are vague or not completely agreed on. I’ve always thought of TA as charting, but it’s not.

  • http://www.deltafinancials.com Delta Financials

    I’ve always thought of it as charts providing context and a series of tools – both technical and fundamental providing the meat. I’ve gone through the drill of the CMT program (which is quite an excellent way of getting beyond silly labels) and much like the author of the reformed broker piece – I can’t for the life or me understand what ELSE short-term traders use to base their decisions. I’d be fairly lost without my charts and sentiment, momentum, money flow, MAs & economic indicators.

    I get the feeling academics often want to “explain” reality (even through theories as inept as EMH) while practitioners are happy with a framework to anchor themselves to reality. Technical analysis is ultimately a toolbox – make of it what you will.

  • http://www.conventionalwisdumb.com Conventional Wisdumb

    Two key things I learned when I was a financial advisor:

    1) Always read the footnotes first
    2) Data is beginning and endpoint sensitive

    With those two rules in effect you can pretty much create any story you want with financial data.

  • http://pragcap michael schofield

    People who dismiss TA generally think it’s supposed to be a way to predict the future. But TA more often than not is wrong, which causes the uninitiated to see that as a fatal flaw. TA is a way to understand the risk and reward. Good R/W analysis comes as a result of understanding the possibilities, which is what a technician deals in.

  • Boston Larry

    An example is that most technicians would say that equities have come up too far, too fast, and are overbought now. TA would “predict” a correction. I agree with Schofield that one should use it to understand that the risk of a correction is higher now than it mormally is. As Cullen says it is one piece of a puzzle, and one tool in the toolbox. On the other hand, we have seen the reversal of a long-term trend of cash flows into bond funds and out of equity funds. Since the prior trend was going on for so long, it may take a long time for this new trend to peter out.

  • http://www.nowandfutures.com bart

    *rimshot* -g-

  • Tom Brown

    You write “But TA more often than not is wrong…” regarding predicting the future… well if that’s REALLY true, then it sounds like TA can at least be used to rule out one possible future (the one it predicts) more than 50% of the time. ;)

  • http://pragcap michael schofield

    HAha. R/W is the key, a potential profit well in excess of the loss on a reasonable stop. It pays to be picky. But I’ll bet you knew that.

  • Lukey

    Anyone using TA to predict the future is asking for trouble. My current investing philosophy is fairly TA dependent, but it is focused only on the present – what is it telling me that the market is doing today. Then that is used to determine my level of risk on vs. risk off. And, so far, I must admit that I’m quite happy with the results.

  • jt26

    Obviously, some aspects of “technical” are rooted in psychological fundamentals (i.e. trend following/momentum, 3′s, pre-conditioning) which are a market inefficiency that can be traded. The other aspect of “technicals” is the feel for the market which traders in all markets from collectibles, Impressionist art, to stocks, have to have a feel for. There is some evidence (i.e. the trend following, CTA crowd) that claims persistent long-term profitability; I’m not convinced. The latter, “feel for the market”, is always VERY important (unless of course you are in a 20-year bull market!).

  • Tom Brown

    So if TA can’t be used to predict the future, how do the “rational agents” of neo-classical macro theory (in their “micro foundations” model) gain their clairvoyance? ;)

    That clairvoyance is required for the theory to hold together. Sumner says it’s a waste of his time to even talk to people that reject the “rational agents” model. Hahahaha!

  • http://www.linkedin.com/in/franksmietana Frank Smietana

    The age-old argument between technicians and fundamentalists misses an important point. Any legitimate trading/investing approach must be objective, quantifiable, and testable. Even under those stringent criteria, the curse of non-stationarity eventually renders even well-grounded trading approaches obsolete. At least for awhile, until they become profitable again. Traders/investors, whether fundies, techies, astrologists or dart throwers, would be well served learning basic statistics and programming to see if their approaches stand a snowballs chance in hell of attaining long term profitability. Cullen — keep up the great writing, you’ve given me plenty of testable hypotheses over the years.

  • Edmund Karner

    Given its centrality to the human experience as we know it, the vagueness and unscientific nature of the twin pursuits are totally bizarre and unacceptable.

  • AdmChesterMynutz

    You have prices and quantities displayed over time. Draw as many lines and formulate as many indicators as you need to make sense of the data. But its still P, Q, over t. Unless your analysis results in a mechanical trading system that is testable, the method may have slightly more empirical content than pure human intuition.

    I can determine where a price is or maybe where its going, but where it is, where its going and where it will be, that is quite a leap of faith.

    If you had to choose between a CMT or a physicist to build an algo, which would it be?

  • Crossover

    Im wondering if it makes sense to use fundamental analysis if you’re a day trader.The way i see it, one should combine TA with FA according to his style.Rely more on TA for short term investments,especially day trade, and combine TA with FA for longer term investments.

  • Lukey

    I think we’re talking about two different “futures” here. Just because how the market is behaving today makes me think it is more likely to move higher than lower over the next few days or few weeks does not mean that I can’t also believe that trillion dollar deficits as far as the eye can see means I’m going to see less GDP growth (and therefor lower stock prices) at some point down the road.

  • Tom Brown

    So regarding GDP growth and stock prices “at some point down the road” … do you see yourself as a clairvoyant rational agent? :o

    Sorry, couldn’t resist that.

  • Gary_UK

    The silence is deafening on this slow news day.

  • AWF

    Technical analysis is akin to weather forecasting
    When was the last time the weatherman got it right?
    I’ve seen Snake oil saleman with better records

  • Android

    hahaha, good one tom. rationality can be mathematically modeled, and the theories can be (necessarily will be?) deterministic, especially under time translational invariance. irrationality can also be mathematically modeled; however, the theories are not necessarily deterministic, depending on initial or boundary conditions, but nevertheless the probability of some vaguely defined end point can be estimated within some confidence intervals. A close parallel to predicting the future is asimov’s psychohistory (i am reading the Foundation series now) where you can not predict individual behavior over a short time frame but can predict the behavior of many people over a long time frame. there is an analogy to this in statistical mechanics which asimov, as a biochemistry, was surely aware of.

  • Tom Brown

    I’ll make a case for it being more difficult: At least the weather doesn’t change because of a forecast.. whereas, if anybody starts to publicly make even slightly accurate predictions, then that causes a feedback loop as more people start to pay attention to those predictions.

  • Tom Brown

    I have not read Asimov. Sounds interesting. Most of what you said, while I understand the words, I’m not getting the full meaning, e.g. “time translational invariance” … I’m used to a “time invariant” system, but I’m not sure what you’re getting at there. I’ll Google it!

    A couple of thoughts:

    1) Once you start predicting what the markets will do with any accuracy, and you do something public with the results (rather than just keep them locked up and not act on them), then you create a feedback loop in the system… one that grows as more and more people start to tune into what you’re up to. This makes predicting the markets inherently different (perhaps harder!) than predicting the weather. At least that’s they way it seems to me!

    2) “Adam K” (a commenter here) mentioned Dan Ariely in a post. He studies irrational decision making. Look him up if you’ve never seen him. It’s pretty fascinating: he speaks to your statement about modeling irrationality, although I’ve only seen his talks to laymen, so I don’t know if a mathematical model is involved… but his examples are really great! They suggest a level of determination in the process. One of the things I recall was that humans tend to depart from rationality if they have too many options available. He gave a scarey example of medical doctors being given some information which led them to believe the patient in question would need an invasive surgery… but then when presented with more evidence that something far simpler could be prescribed, most called off the surgery and went with the simpler option. However, when evidence for a 3rd possibility was introduced, most Dr.s got overwhelmed (apparently) with too many choices, and stuck with the “default” position: surgery! What I’ve seen from him ties into economics.

  • http://None Midas II

    The”feedback loop” is always in play. Everyone is watching, and they all try to discern the same trends. It sets a pattern that looks like “follow the leader” although it is actually parallel thinking.

  • http://www.asiachart.com Gregory Barton

    Technical analysis is a black art, more art than science. At best it is a blunt axe, not a scalpel, and it is only as good as the axeman who wields it.

  • Dennis

    I am “buy and hold” except that I try to sell in May and save up for October. This is why See Federal Withholding & Payroll Tax Deposits Rolling 4 – Week “Standard Periods” http://www.shadowstats.com/charts/employment/federal-withholding-taxes

  • Lukey

    Not “clairvoyant” so much as, perhaps, perceptive. Say, Was Tiger Woods moving out of California a “clairvoyant” move?

  • http://make300aday.com Steve Wise

    The biggest problem with technical analysis is the susceptibility to the inevitable Black Swan events which not only reverse earlier gains but also wipe out initial capital.

    I agree tech analysis is useful to know, but its weakness is portfolio destroying.

  • http://www.asiachart.com Gregory Barton

    On the contrary, good trend analysis would be one of the few indicators of an impending black swan.

    Relying on TA to the exclusion of fundamental analysis can be capital destroying. But technical analysis is a trading tool, not an investment tool and a portfolio is a collection of investments, not trades.