Technical Analysis (Part 11): Direct Price Analysis (DPA): Bands

19th Jun '17 by Stephen Hoad

15 comments 9563 reads

In the last article, we looked at Moving Averages. In Part 11, we take the evolution of direct price analysis, in technical analysis, a step further in its application. In this article, we are looking at 'bands' created around price, for the purposes of trade management and identifying trading opportunities. We are going to look at four of my favourites: Bollinger Bands, Donchian Bands, Keltner Bands and Envelopes

Why use bands?

Price bands plot a price range around the original price, above and below. They overlay the price on the chart. This range can be based on percentages, statistical measures, volatility etc. These bands will move and adjust with the flow of the price. A band will help the analyst better determine if the trend is likely to be broken; if it is continuing within the band, it can offer statistical measures of where to possibly place your stops or take your profits. These tools can be used as trading systems by themselves; however, I find they work better in collaboration with other technical tools and approaches. The four we are going to look at are arguably the most commonly used in technical analysis:

Bollinger Bands

These bands encapsulate the volatility of the underlying instrument. They were created by John Bollinger and use a statistical concept to capture price movement. By default, the tool plots 2 standard deviation bands away from a 20-period moving average. Because of this set up, it tells the user that between the two bands, 95% of the 'normal' price action should occur. When markets become more volatile, the bands widen and when less volatile, they shrink. The inputs - the standard deviations and moving average length - can be adjusted for the purposes and objectives of your analysis and trading.

Some trading thoughts for Bollinger Bands

Price that hits +/- 2 standard deviations can signify price reversals back to the mean moving average.
Bands are better used to determine the beginning of a trend rather than trading between them. Trading between the bands on first glance looks the obvious solution, but this can be tricky.
A breakout from the bands that contains approx. 90% of previous price action suggests the general trend of the previous price action has changed in the direction of the breakout.
When bands tighten, this can often signify an explosion in price going forward - use other indicators to line up…

Unlock with Email

Unlock with Google

Already have an account?
Login here

aflash

19th Jun '17

Stockopedians have very little interest in Technical Analysis, apparently.
I use an American service that ONLY analyses charts and twice a week takes requests from subscribers.
The problem is that UK stocks on New York have lots of gaps are difficult to analyse except on a weekly and monthly basis. I wonder if Stephen would consider filling the 'gap' (ha-ha)? I notice he has already offered opinions and 'charts of the day' here. It seems the StopHunter only offers one-on-one in person analysis.

Stephen Hoad

Hi,

Uptake hasn't been too bad - guess I have to do a lot of convincing! ;)

I might be being a bit thick in understanding your question so apologies if I have got the wrong end of the stick.

One of the facets of my company THE STOP HUNTER is to provide training and education in Equities, Indices, FX and Commodities. In THE STOP HUNTER we also run a traders club where I analyse technically, fundamentally and behaviorally those four markets producing individual reports every week for our members. In equities for example I analyse both UK and US stocks, sectors etc. The Indices report I analyse most of the global markets. If you're interested in the traders club, if you can hang on until the first week of July I will be running a one month free trial offer, so you can take a look without the commitment and then if you like it it is £49 per month after that or just cancel.

Other parts of the business, I offer 1 to 1 coaching and mentorship. I also do a lot of consultancy work in the City and teach technical analysis for the Society of Technical Analysts and Universities, write books etc etc etc. (I like to keep busy!)

I also trade for myself running my own fund which trades all those 4 markets which is predominantly technical analysis based using Japanese charting techniques blended in with the best of the West. (Current returns FYI 20% per month - results published in Traders Club!!) - as well as building other trading strategies, systematic systems etc.

Where possible I will try to analyse stock markets technically and post out - firstly that will be in THE STOP HUNTER world and then in Stockopedia for example. From my pov Im also looking at FX, Commodities etc so often I am busy in those worlds.

Also depends on the demand and what is going on globally. If I turn my attention to stocks in much more detail it would have to be worth my while. The remit was to firstly provide education on technical analysis for Stockopedia to go alongside their new platform and then if there was the interest possibly to build this out further. I will have a further chat with management to see if they'd like me to contribute more technical analysis and ideas.

I personally think technical analysis combined with fundamental analysis provided by Stockopedia could make a highly potent and successful trading approach. Technical analysis is also especially important for risk and trade management. I have several strategies in the equity space that work that way. Those that dont embrace both worlds I believe are missing out.

Regards

Stephen

20th Jun '17

If you're interested in the traders club, you can take a look without the commitment

I did look at the Traders' Club section on the website. As far as I could understand the Traders' Surgery, they are meetings in the U.K. Do you make a recording for those unable to attend?

Then those reports you mention are articles, illustrated with charts, on situations you find technically important or webinar type recordings where we look at your software moving?

I shall certainly sign up in July as I agree that is the way to go. Stockopedia is much more useful for medium and small cap stocks where the chart patterns are hampered by spreads and absence of liquidity.

Morning - doing a major overhaul of the whole lot and hence the July offer to kick start things. Website will also look different. My net has spread so I'm introducing webinars, recorded stuff etc from July to meet just the needs/ questions you have. Reports - exactly that. At the moment my software is 'post' event due to FCA rules etc - but I publish the results and analysis around what I have done so that through the club you can learn what I do and build your own system (s).

Totally correct on your observations - technical analysis works best in more liquid markets, but there are techniques that many aren't aware of to counter this for the mid / smaller size stocks or in other markets such as commodities which suffers the same issues eg Lean Hogs vs WTI crude oil..

Regards

Stephen

monions

Thanks for the extra information Stephen.

I would like to see much more information and discussion around TA. As i said on one of the earlier posts, I see modern TA used together with fundamentals to be the obvious way forward. This is basically Minervini style investing which many people on Stockopedia favour.

I'm still not sure why the term 'trading' is seen as a dirty word around here. To me, long term and short term are both valid time scales. you use the one most suitable for the situation at the time.

The more we can learn and discuss the better. I'll certainly take a look at your revamped Stop Hunter site, I've been looking around it since you started these articles and found a lot of useful information there.

My Equity trading is all about finding the right stock fundamentally then trading it technically. My trades can last anywhere from less than a day to a couple of months. What technical analysis can do that fundamental analysis cant is measure the behaviour / sentiment of the market - I also employ a lot of behavioural science into my trading. No point buying a stock on fundamental analysis if the market is selling it to death - you're wasting your time and money. If the stock does turn in favour of your fundamentals - wait and use technical analysis to time your entry and get the most bang for your buck - that's what its all about isn't it?

It's a grey area between the term trader and investor. You can have long term traders that you'd maybe otherwise class as investors. I think many investors could be more proactive with their portfolios to maximise their returns and adopt a more 'trader' mindset and that is where technical analysis comes in: optimal entry / exit timing, risk and trade management eg stops etc. I've coached a traditional share club for example quite entrenched in the fundamentals to include technical analysis as an overlay to their fundamental analysis to improve entry and we introduces technical analysis to provide an exit / stop loss strategy.

Regards

Stephen

chillyhill1

Hi Stephen,

I've enjoyed reading your articles but the problem I find is there are so many different techniques which make it difficult to decide where area to focus on. Currently, I just tend to use moving averages, trading ranges and breakouts but would like to expand my knowledge but in which area?. I've been reading technical analysis for dummies but in my opinion it's not the best book. If you have any other good starter books and suggestions on areas to focus on that would be much appreciated.

Thanks

While waiting for Stephen's reply I pulled out my copy of Technical Analysis for Dummies 2011 edition by Barbara Rockefeller.
It is very good.
She takes no system very seriously and explains that No technique works all the time and No technique works on every security.
Not only is there an overview of most systems there is a lot of good advice and some savy explanations of the phenomena.
Take responsibility for your trades, define your type of trading, my highlights and annotations are frequent.
You can use the internet to examine patterns. Just type in what you are interested in. There are so many sites that give detailed explanations.
To give you an idea on authors there is Steve Nison on Candlesticks Robert Prechter and now John Burford on Elliot Wave.

Hi,

To start with getting an overview of whats available through my articles is a good start - I will cover off pretty much the spectrum of content that will be all you need to know (until you dive deeper into the subject). At the end of each article I recommend further reading and learning. Most of these books are on the reading list for the MSTA technical analysis exams. You'll notice 2 or 3 books that run throughout my series of articles. I taught myself initially from these books. I am hopefully releasing a book on technical analysis in the late summer as well as 2 courses on technical analysis - beginners/ intermediate and advanced - will keep all posted!

Technical Analysis for dummies isn't on my reading list so can't comment on the quality of the book and what it teaches.

Barbara Rockefeller isn't someone I've really got into (She comes from an FX background) but sounds like she talks sense. Unfortunately also which clouds the answer to this question is that there are soooooooo many books, videos, websites, You Tube channels, all offering their spin on things. My suggestion is to find someone who has qualified in technical analysis (MSTA, IFTA, MTA) and had practical trading / technical analysis experience working for a bank / fund etc as often these people have a very broad understanding of technical analysis as well as often being experts in a particular sub-section Most of the teachers in technical analysis you'll find are ex-market professionals. With their more rounded view they are more likely to be able to help.

Like you say technical analysis is a huge field. I've been lucky to have been playing with technical analysis for over 20 years so over that time I've refined what I like to use. It is like horses for courses - you'll find a part of technical analysis that suits you. Unfortunately sometimes this can only be done from trial and error and experimentation.

One of the most important aspects is that you want to learn and study technical analysis, learn the art and don't think of it as just a money making golden ticket - by doing that you will more quickly find technical aspects that suit you rather than trying to find the holy grail - it doesn't exist! There is no one size fits all and that is one of its problems and also one of the things that makes it so great. There are some parts of technical analysis I just haven't got the time to learn properly or use and they could be fantastic and I could be missing out but I've come to the conclusion that I am going to have a broad knowledge of everything and refine into areas that suit my personality, trading approach and strategies, lifestyle, trading tools etc etc eg Elliott Wave, Gann, Market Profile (there's an article coming up on them btw)

I've tried most things but now just have 4 set ups that I use based on Kagi, Renko, Line break and Ichimoku/Heikin Ashi: heavily Japanese which I've blended in just 2 or 3 western indicators. My policy is to keep it simple - you can end up over analysisng and getting the dreaded paralysis of analysis! (there will also be an article coming up on them as well)

You may find some of the blogs and resources on my website www.thestophunter.co.uk useful. Towards the end of my 25 part series there will be a couple of articles where I talk about putting it together, refining your technical analysis, helping you find what may suit you.

Hope that helps.

Regards

Stephen

21st Jun '17

Thanks Stephen and aflash for your comments

Nick Ray

These bands encapsulate the volatility of the underlying instrument. They were created by John Bollinger and use a statistical concept to capture price movement. By default, the tool plots 2 standard deviation bands away from a 20-period moving average. Because of this set up, it tells the user that between the two bands, 95% of the 'normal' price action should occur.

There are several things wrong with this paragraph. Bollinger bands do not use the volatility. The volatility is the standard deviation of the returns (i.e. change in price) not the share price.

The price of a stock on different days is not random and independent - it is highly correlated. So the standard deviation of stock price does not provide any useful information. It is certainly not the case that 2SD will cover 95% of the normal price action. That would be true if the prices were Normally distributed but they are not. They are not even close. For example the chart below is the last 200 days of price action for VOD and you can see that it is nowhere near a Normal distribution. The use of the standard deviation by Bollinger bands isn't motivated from any underlying mathematical reason.

22nd Jun '17

Essentially, a point not worth arguing as its the perception / the behaviour of how the market interprets and utilises the tool and that is how they think about it. It is how it is taught and technically 2sd's are 95% - why don't you take a look at a few charts and see how the price mostly falls between the two bands. You're also missing the point - if something is non-normal and the market perceives it as normal then that is a trading strategy - using the bands could be a good reverse strategy as the data may not fit! I will ask John Bollinger though what his motivations were just to clarify things - he follows me! Also a bit like Fibonacci - shouldn't work but it does.

Some further education: Unfortunately the worlds financial markets (and I have had more experience than most in knowing this) in both risk and front office functions start from the point of view that the market / world acts normally. It is very obvious that individually it doesnt and not everything is normal but as an assumption this is where the markets start from and tweaks - from my commodities background I had to price both electricity options and gold - electricity is non-normal, mean reverting, seasonal, half life and gold on the other hand does act more normally. In risk most of the Nobel prize winning fundamentals of measuring market and credit risk start from the assumption that most things behave normally - and is where you get the concept of Value at Risk from for example. This is where the quants come in and build their models to tweak to a particular market - I can tell you equities in comparison to most markets acts normally but you are correct that some individual stocks dont act normally say those with less liquidity and this is where technical analysis can fall down.

I could also mathematically prove that Vodafone acts normally and that Bollinger Bands are a sensible tool to deploy. But if you dug into understanding the Bollinger bands you could mathematically change the inputs to make it fit - but then data fitting is also a cardinal sin of why people fail at trading. I'd ask, WHAT IS NORMAL? I could create and manipulate a distribution to approve or disprove any concept.

Glad I am creating debate though - that is what it is all about!

The price of a stock on different days is not random

but you told us the other day the Patterns are.

- See more at: http://www.stockopedia.com/con...

Please develop

Just to clarify some points raised.

The Normal Distribution is the name for a very commonly occurring probability distribution. It is sometimes called the "Gaussian" distribution. It doesn't mean that things are normal in the colloquial sense. It is a very precisely defined term. If your distribution is not "Normal" ("Gaussian") then you can still compute means and standard deviations but you can't use the properties of the Normal distribution to work out what they mean (such as 2SD covering approximately 95% of cases).

Value at Risk (as well as volatility, Black-Scholes, Sharpe ratios, and risk-adjusted momentum, etc) are based on the probability distribution of stock returns not stock prices. That is the distribution of Price(today)/Price(some previous time) -1. This distribution is fairly close to a Normal (Gaussian) distribution (but there is a lot of literature pointing out ways in which it differs (and we risk getting into the niceties of "Normal" versus "log-Normal" too.))

It is the work based on the mean and standard deviations of stock returns which has won Nobel prizes.

As for whether Bollinger Bands have some empirical use because the stock price does stay inside them most of the time, well that is fine. I have no problem with the claim stated in that way at all. Short-term trading is so far away from the way that Mean-Variance analysis works that the worlds do not really meet.

Personally, I do think Bollinger Bands are over-complicated. I prefer to use what the article calls "envelopes" where a fixed percentage is added/subtracted to the SMA as per example below.

As I've mentioned in an older post somewhere, one simple estimate of the mean daily return is (2/N)*(P/SMA(P,N)-1) for any value of N (say 100 or 200). So if say you want to be alerted if a stock's potential yearly return is falling below some minimum value 'm'% we convert 'm' into a daily return (assuming 250 days in a year) so we want (2/N)*(P/SMA(P,N)-1) >= m/250/100 which means that we want P >= SMA(P,N)*(1+m/250/100*(N/2))

If N=200, m=25, that gives for example P >= SMA(P,200)*1.1 so I'd use a 10% offset - but note that I would want the stock to be above the SMA+10% not inside a channel of SMA +/- 10% in that case. It doesn't look as if Stockopedia can plot envelopes (can it?) but you can set up a screen like this:

%200d MA >= 10

Obviously this particular use of envelopes is not aimed at those doing short-term trading.

Technical Analysis (Part 11): Direct Price Analysis (DPA): Bands

Why use bands?

Bollinger Bands

Some trading thoughts for Bollinger Bands

Unlock the rest of this article with a 14 day trial

About the Author

Stephen Hoad

15 comments

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

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