Not a review of The Art and Science of Technical Analysis

22nd Dec '19 by HumourMe

2 comments 1322 reads

Not a review as:

I haven't finished the book yet
I can't do it justice
In the sceptical spirit it encourages, I'm seeking contrary views

This post then describes why I'm finding a particular TA book useful, in case anyone else might find it, the free course or podcast useful too. Be warned that if you think you have a working system you might find the material challenges your current perceptions. If this makes you uncomfortable, quickly move on.

This year I've spent some time investigating technical strategies with the goal of finessing entry and exit. I've been trying to develop a breakout strategy and managed to lose money slowly (~150 trades with an overall 1.5% loss - so at least discipline and money management worked). During the same time period my (mostly neglected) joke Fantasy portfolio, which initially held near the maximum amount of cash allowed, went up 19%. A useful experience, this points the way forward for me (technical investigations off-line, while mainly pursuing a NAPS like system in real life).

During the year I'd sought evidence for, amongst other things, common moving averages strategies, pattern strategies and % based stops. Back testing, while it may have its flaws, can tell you what hasn't worked recently. I didn't find much evidence to support the more common assertions, but I did enjoy the programming challenges!

I did come to the conclusion (after thinking about them since reading Bulkowskis Encyclopedia in 2000 and revisiting updates this year) that all patterns were just variants on a theme; consolidations by any other name, they represent a loss of momentum and a potential change of direction. With no need to classify them as Table Top, Batman or 'Paint-pot Leaking' they can be analysed with far less effort. Moving averages, as popularly promoted (crossovers, order, slope etc..), don't work for me (I've still a few ideas to test, now focusing on distance from the average and direction). Others may have evidence that these popular strategies work. If so what else do you use them with? With stops I need to think this through further but my direction is towards volatility based.

During the year a popular thread came to my attention. https://www.stockopedia.com/co... This was demonstrating a methodology called 'Wyckoff' and is worth reviewing if you've managed to ignore it. After 30+ years investing on and off, I…

Unlock with Email

Unlock with Google

Already have an account?
Login here

Nick Ray

22nd Dec '19

I confess that did I read a few TA books early on, and they cost me a lot of money!

My personal arc went something like this fwiw:

1. I started with Technical Analysis, partly because the web is awash with it. You see all these techniques like MACD and RSI with various claims about how to trade them. In my case after finding that they did not seem to work I coded up some backtests and discovered that all claims for indicators that you find on the web are false!

The books often talk about an "edge" but also about not hesitating when deciding whether to buy or sell, and generally blame any losses on a lack of confidence or the lack of an "edge".

2. Eventually you realise (more clearly than when you start) that everyone else is trying to outwit the market too. And you start to wonder: how does a system where everyone is trying to find patterns to beat the market actually perform? Of course it is not an original thought and eventually leads you to articles about the ideas of "efficiency" and "anomalies" or alpha versus beta.

Taleb calls the market "anti-inductive". In other words it resists attempts to find inductive rules for how it behaves.

3. The third part of the journey consisted of trying to work with the market instead of against it. That meant factor-based or smart-beta approaches; and in particular, designing portfolios with specific properties, rather than stock picking.

Slightly to my surprise it also led me to the Sharpe Ratio as a remarkably good building block for this task. The reason this is a surprise is that the Sharpe Ratio is pure TA - it is just based on price action.

4. There seems to be agreement about what really works once you get away from those trying to outwit the market (and apparently under the impression that you can write books to tell everyone how to beat the market - an obvious mathematical impossibility).

It is sometimes called GARP: high growth or quality, but don't over-pay. Positive price action with low volatility. Choose Quality/Stability/Profitability metrics better than the median value. "Value" is much more difficult to assess because it is not as simple as "cheap". I find that low volatility is a useful proxy for good value.

The set of stocks which meet some variation on these rules is about 10-20% of the market. The set is so "obvious" that it just keeps on popping up on the top-ten lists on the home page and endlessly discussed in the forums. It slowly changes as new stocks enter and old favourites leave.

It will generally out-perform the index or match it. In fact it might even be the case that the performance of the "top ten percent" reflects the true market performance and the poorer performance of the whole market index is some kind of secondary effect. (So you cannot "beat the market" but you can under-perform it!)

HumourMe

23rd Dec '19

Nick you downplay your personal arc, however, you share much of the same reasoning as Grimes with 1 & 2. It may be a necessary path some people just have to work through. It was necessary for me, when I became seduced by the idea that there must be something in it, if so many people seemed to think these things worked.

For 3 / working with the market, for Stocks, Grimes implies that reversion to the mean is generally stronger than trend (although drift presumably determines which particular point, price is 'reverting' to); all is obscured by random price movements due to 2. His 'Technical' response is to buy pullbacks and 'support tests' (as two of four trade types). The reasoning here being that this is where the buyers, whether technical or fundamental in motivation, were previously prepared to pay. The 'Fundamental' answer to 'fluctuations' is to not overpay and wait until someone is prepared to overpay sufficiently. Both approaches are attempting to be systematic.

Buying a basket of shares, based on a set of criteria, known to, on balance, demonstrate price appreciation (e.g. Factors), tilts the balance of probabilities in ones favour and is one way to create an edge. A portfolio approach also makes much of the risk management considerations of TA, designed for individual trades and larger individual position sizes, irrelevant. Overall a portfolio approach just seems easier and much more aligned with my goals.

4 G/QARP is where I'm heading now that I'm convinced that other paths are not for me. Distinguishing Value & Cheap is certainly a challenge. The F-Score, a large part of the Quality factor, was I believe, originally designed to improve Value selection. It does show that on balance improving fundamentals is good for price appreciation although the effect will be logically stronger the cheaper the start.

Not a review of The Art and Science of Technical Analysis

Unlock the rest of this article with a 14 day trial

About the Author

HumourMe

2 comments

Unlock the metrics for all these stocks and more...

Unlock the metrics for all these stocks and more...