Showing posts with label Confirmation Bias. Show all posts
Showing posts with label Confirmation Bias. Show all posts

Tuesday, 17 April 2012

EURUSD - Not a Head & Shoulders Pattern


This is NOT a Head & Shoulders Pattern.


Nor is this:

Yesterday I was sent some research from a Tier 1 investment bank that went as follows:

EURUSD: Breached the head and shoulders neckline at 1.3037 and is also threatening to take out important support at 1.3004 (breached intraday already). A close below 1.3004 would confirm this bearish break and suggest a move down to the 1.25 area. 

Many years ago, I worked with a brilliant Technical Analyst who would react with fury on an almost daily basis to traders coming up to her telling her that they had found a Head & Shoulders pattern and had loaded up a position on the back of it.  When she looked into the pattern she would advise them that in nearly all cases these were not H&S patterns, and that their rationale for the position was incorrect, in most cases she was proved right. The biggest mistake she would point out was the Head & Shoulders patterns are ‘End of Trend’ patterns, they occur in the wake of a significant trend, and are usually relative in size to that trend. (There are of course other qualifying criteria this article here covers this nicely).

I was I admit quite surprised to see this investment bank sending out a report with such a basic error in it. – I do not dispute the overall bearish nature of the EURUSD at present, however the tendency to look for patterns in markets to supports one’s view rather than observing price action and data objectively are fundamental analytical flaws, more generally known as ‘confirmation bias’, it is also lazy analysis. 

I am not being pedantic here (ok maybe I am slightly), but from my experience mis-labeling of patterns can create a false sense of confidence. I do not disagree with the fundamental premise of the analysis, a close below 1.3004 would be bearish, however the readers attention would have been drawn to the initial comment about the neckline of the Head & Shoulder being broken, this may have led them to false conclusions more so than the qualifying comment which followed it. 
 
To finish I would like to present some good examples of Head & Shoulders patterns: The chart below shows the SP500 over the past few years, with three major ‘Head & Shoulders Patterns’, two successful and one failed. – The two successful ones proved spectacularly effective at signalling sharp moves ahead, the other one not so, however even failed patterns can prove useful, in that their failure can often signify that the prior trend is likely to re-assert itself. And note how the failed pattern itself ended in a perfect inverted Head & Shoulder Pattern.


When Head & Shoulders patterns work they can be spectacularly successful signals for market direction, however the world is littered with failed traders who spent too much time looking for patterns which aren’t or weren’t really there.

Friday, 10 June 2011

Confirmation Bias - A pernicious threat to your own trading.


Do you recoil in anger at a news story, item, article or blog that differs strongly in opinion to your market view or position?
Do you perhaps find that when a blog has an opposing outlook to you on the market, you stop visiting it?
If the headline to an article is the complete opposite of what you think, do you blatantly disagree with it, even without reading it?
These are all signs of ‘confirmation bias’; a bias which poses a serious risk to your trading success.

When traders have a view, hypothesis or opinion about market direction, there is a danger this can become deeply entrenched. The placing of trades and strategies concurrent with this view deepens this attachment, as do mixing and sharing this view or opinion, with other like-minded individuals. The discussions will veer towards confirming with each other how right you are, and any opposing views opinions will be quickly dismissed. Typically then what happens you pick up on evidence that supports your hypothesis and ruthlessly ignore or dismiss any evidence that opposes it, even without giving it due consideration, often without even reading it or listening to it.

Over the years I have seen the 'confirmation bias' literally destroy traders. I myself have been victim to it, though thankfully without disastrous consequences and before it became too entrenched. I think humans probably have a natural tendency towards it, perhaps linked to a deep-rooted behavioural trait or habit that is natural to us in some form and which served its purpose in ensuring our survival when we roamed the ancient open plains, savannahs and jungles.

My own way of dealing with it, or overcoming it, was to be rigorous and thorough with my analysis and research. I would make sure I tried to visit, read and listen to opposing views. On many occasions, where I had a strong technical analysis opinion, or had seen some research or data which lead me in one direction, I would seek to disprove it, deliberately looking for flaws in my arguments or conclusions (sometimes the market would do the work for you). - As these flaws or holes would disappear or not stand up, my conviction would grow that I was on the right path, though I would always try and keep an open mind. - Even then I was still prone to lapses, but I think you have accept that it important to try and strike a balance between having a degree of self-belief and conviction and questioning that belief, otherwise you risk ending up with no view and never taking a risk.

It would be interesting to hear other people's opinions, thoughts and experiences on this.

AlphaMind podcast #107 A US Navy Seal Commander, A Mindfulness Expert, and Self-Compassion

In the brutal world of trading and markets, we can often turn in on ourselves, and end up becoming our biggest problem. The ability to stay ...