How to fight the markets and your own biases

Sep 07, 2012, 04:29

Share with
friends:

Comments (16)

Before I cover gold, I would like to discuss a very important factor in trading the financial markets. It’s one which many traders pay absolutely no attention to, but can cost them dear if it is not addressed.

I am talking about trader biases. Don’t worry – we all have them. The key to operating effectively though is to recognise your own biases – and whether yours are so extreme that they blind you to taking wrong trades time after time.

Psychologists have investigated the human tendency towards bias for a long time. We’re predictably biased in a number of ways.

A typical bias that affects traders is ‘information gluttony’: The more I read and gather information on the economy (or the particular market we are studying), the more confident I will be in my forecasts – and I will be a better trader.

This seems like common sense, does it not? Most people believe this. But in the markets, this is a dangerous belief!

So much so that Daniel Boorstin said: “The general obstacle to discovery is not ignorance – it is the illusion of knowledge”.

Some time ago, a famous experiment was conducted using experts in their field - eight experienced bookmakers were shown a list of 88 variables that affect the performance of racehorses.

Each bookie was asked to rank each piece of information by importance.

Then, the bookies were given this data for 40 past races (with results hidden) and asked to rank the top five horses in each race.

Each bookie was given this data in increments of the five, ten, twenty, and forty variables the bookie had selected as the most important.

Thus, each bookie predicted the winner four times in each race. For each prediction, the bookie was asked to state the degree of confidence in each prediction.

The results were surprising! They showed that instead of accuracy and confidence increasing over time, while confidence increased markedly, the accuracy remained flat.

With just five pieces of information, accuracy and confidence were closely related. But with forty pieces, accuracy was still only around 15%, while confidence soared to double the initial reading!

The implication for us is clear – more information does not automatically translate into better forecasts. That is why I like to keep things as simple as I can – and why I stick to my trading methods, which are the essence of simplicity.

Gold is the asset investors love to love

OK, let’s turn to gold. Surely, I am not alone in noticing a barrage of bullish articles on the yellow metal? The Telegraph, in particular, has been running this article repeatedly: "Gold to hit $1,800 by year-end – predicts analyst".

This forecast is pretty tame compared with others I have read.

Gold may or may not hit this target, but one thing is certain – that when gold rallies, bullish articles come out of the woodwork. Where were they in May, when gold was trading $160 lower?

Meanwhile, since I last covered gold on 29 August, the market has caught a bid and has broken above my upper tramline:

Gold spread betting chart

(Click on the image above for a larger version)

So, is this the start of a big rally towards the $1,800 area that has been widely touted as a year-end target? If so, I had better mend my ways and start trading from the long side.

But there are a few reasons why I am reluctant to switch camps just yet. In the above chart, note the momentum reading is at levels previously identified with tops. The market is approaching being overbought. This is not necessarily fatal to the rally, since high readings can exist for a long time in a strong move, but it is flashing a yellow.

Incidentally, my short trade described last time resulted in a loss. Here is the chart:

Gold spread betting chart

(Click on the image above for a larger version)

For a while, my trade was looking good as the market dropped to the sub-$1,650 level. But then there was huge buying which resulted in new highs for the move.

If I was trading on the very short term, I had an opportunity to take a profit before the key reversal kicked in.

My loss was well within the 3% rule limit, so that will go down as one of my losers.


MoneyWeek Trader logo

Claim your FREE report: The six-step game-plan for
spread betting profits


A great mental test

Now, the market has been powering up and I have been caught wrong-footed. This is always a great mental test for a trader. I have found that the best policy in this situation is to do nothing. It is all too easy to panic into a wrong move at the wrong time. You want to recoup your loss quickly, don’t you?

In gold especially, I have found it normally a poor policy to chase the market. Mostly, you will be getting in just as the market begins a sharp counter-move. You may find yourself at the wrong side of one of the many spikes!

The other difficulty is this: if you are fortunate and the market continues going your way, where do you take profits? Psychologically, you are so relieved to be clawing back some of your losses that you are tempted to stay at the party too long – your bias here is over-optimism.

I confess I suffer from this bias, and so I must always check myself when in this situation.

Gold approaches the ‘danger zone’

Back to gold – here is another reason why I doubt the longevity of this rally:

Gold spread betting chart

(Click on the image above for a larger version)

The rally has overshot the Fibonacci 62% level of the move down off the 29 February high to the 16 May low. This could be significant.

Here is the action in close-up:

Gold spread betting chart

(Click on the image above for a larger version)

The market has tried to move above the 62% level, but has been knocked back down to it as I write.

The pink bar is my idea of the danger zone for the bulls. A move below this zone should herald a deeper retreat – and a possible short entry.

I am still treating the May-September rally as a bear market rally – so far. Am I exhibiting an irrational bias?

Incidentally, I spotted a great tramline trade in a related market – the AUS/USD, which has at least partially offset my gold loss. I may cover this off-piste market next week – I know many traders are interested in this market.

• If you’re a new reader, or need a reminder about some of the methods I refer to in my trades, then do have a look at my introductory videos:

The essentials of tramline trading
Advanced tramline trading
An introduction to Elliott wave theory
Advanced trading with Elliott waves
Trading with Fibonacci levels
Trading with 'momentum'
Putting it all together

• Don't miss my next trading insight. To receive all my spread betting blog posts by email, as soon as I've written them, just sign up here . If you have any queries regarding MoneyWeek Trader, please contact us here.

Comments (16)

Share with
friends:

Comments

  • 1. Chris

    (07 September 2012, 06:18PM)  Complain about this comment

    I can't help thinking that Gold is the currency of choice now that the Fed and the ECB are indicating the money presses are warming up. I think this rally could seriously gather momentum over the coming months. Good luck everyone and thank you so much to John for sharing his trading expertise with us.

  • 2. joseph Gillen

    (07 September 2012, 06:20PM)  Complain about this comment

    No Comments yet?
    perhaps the bears are speechless,

    Susan may now wake up and wont be so critical of advice next time.
    What happened to the big fall in Gold (and the Dow)???
    I told you about the H+S on the USD. Whats the target?
    Dont be engrossed in very short term charts.

    I think what you are doing is very useful but the novices out there have a lot to learn. They should study hard - it is not as easy as being suggested. For example go read this link
    http://www.forexstreet.net/profiles/blogs/the-traders-journey?id=3252082%3ABlogPost%3A345877&page=1#comments.

    Soon I am going to be too busy on a new job to contribute again, but I will be watching your posts to help me learn more.

    Joe

  • 3. Grosvenor

    (07 September 2012, 08:26PM)  Complain about this comment

    Thanks again John for your insightful comments. I have just been looking at my chart and note that I still have the fib retracements from 6 and 26th September last year when Gold plunged from the high of 1921 down to 1532. Based on this we have (as I write 20:19 Friday) popped above the 50% level (1726) to 1736 - would these fib values that are a year old still be valid? I note that the 61.2% level (1772) has been reached twice in November and February - looks like we are heading there (and above) now?

    Gold
    Top 1921.00 06 September 2011
    Bottom 1532.70 26 September 2011
    Drop 388.30


    23.60% 91.64 1624.
    38.20% 148.33 1681.
    50% 194.15 1726.
    61.80% 239.97 1772.
    76.40% 296.66 1829.


  • 4. Beasties

    (07 September 2012, 08:38PM)  Complain about this comment

    Yawn. Yes it does seem that trader bias is indeed in play here. But it's not news is it? John's been in denial for quite some time on gold.

    If you go back about 4 years on the charts there are some pretty obvious upward sloping tramlines screaming at you (if you use the log scale). These would indicate that 1790 is the next target in the near future.

    Yes, he's identified some short trades over the last few months, but the trend is still very much intact, and he's been clutching at straws for a couple of weeks now to stick to his guns. Look at my comments under his article "the big question for gold traders" from a fortnight ago and you'll see what I mean.

  • 5. Susan

    (07 September 2012, 10:52PM)  Complain about this comment

    Hi John , thanks for another very informative and frank article. I am very much with you on the May to September still being a bear market rally as well as being reluctant to switch camps as yet. Thanks to your tramlines I was able to take a very nice 285 pips on Tuesday shorting the Dow. I took profits at the lower tramline running since 4 June at 12975. It didnt make it this time, but I'm sure it will not be too far away!
    Many thanks and have a great weekend

  • 6. Bronco Bill

    (07 September 2012, 11:28PM)  Complain about this comment

    Believe me John I and many other gold "bulls" or gold "bugs" have not bought and held gold since 2001 because we are biased and are in love with the yellow stuff but because with negative low interest rates, money printing etc etc its been and is the sensible thing to do and even more so today.
    Yes, I know I know its all about time frames well here's my take on it today for what its worth.
    Daily chart: Overbought zone but now well above 144 and 252 MA's.
    Weekly chart: Not overbought and now above 30 MA and for the first time since 2009 after dipping below last March finished week above 52MA again.
    Monthly chart: Not overbought and most importantly for the second month running above a rising 9 month EMA.
    For short term traders a pullback to retest the daily moving averages around the 1660/1640 (todays level) would not be unusual and in a way for us long term holders would not be a bad thing.

  • 7. Susan

    (07 September 2012, 11:38PM)  Complain about this comment

    @Beasties How do you workout that we are in an uptrend? The highs from Sept 2011 has been lower highs. The lows have been lower lows. Sure there has been a run up this last few months, but until price exceeds a few of the major highs or indeed the Sept 2011 high, there is absolutely no confirmation that we are in an uptrend, more than likely just a correction of the existing downtrend.

  • 8. Joe

    (08 September 2012, 03:42AM)  Complain about this comment

    Continued from comment 2.
    The USD index has about 2% further downside on the head+shoulders.
    And has clearly broken the uptrend from last september. The Euro mirrors it.
    The Dow has now moved into what I see as the fifth wave of the major 5th wave in this bull market since march 2009 (comments before).
    Target about 13500 on major tramlines. It has seriously lost momentum and may not get there next week.

    I do agree that the big fall is now very very close.
    How will we judge confirmation?
    That should be a major topic for all of us.
    FYI I believe that first confirmation will be break of short term trendline support now at about 13100 AND break the 10week ma at about 13000. This is serious and likely to happen very fast.
    Down to the long term uptrend support at 12500, to extreme as low as 12100.
    Below that is the bear market in earnest. Likely to be a rally first, but it will be October and as John rightly points out this is more likely atime for the fall.

    Joe

  • 9. Beasties

    (08 September 2012, 09:47AM)  Complain about this comment

    Susan, not a problem. It is just my opinion though!

    If you go onto a daily log chart going back to 2007, draw a tramline from the Mar 08 top, go through the Dec 09 top, May 10, June 10 tops, May 12 low. Also it passes through the Jan 11 low vaguely. This isn't a perfect touchpoint, but you should always be aware that the USD strength/weakness has a bearing on exact gold prices, to the extent that you should NEVER expect tramlines to be perfect in the gold market. I've pointed this out before on other threads.

    Happy with that tramline? Firstly there are other steeper tramlines that you can draw which are valid, but this is the one I have most faith in. Right then, draw another parallel tramline which touches the Nov 10 high and Dec 10 high, Sept 11 low (and various other points in between), Apr 12 low, June 12 high, and Aug 12 high.

    Ok finally draw another parallel tramline touching Apr 11 high, 22 Sept low, and 12 Mar uptick. More below...

  • 10. Beasties

    (08 September 2012, 09:48AM)  Complain about this comment

    Ok finally draw another parallel tramline touching Apr 11 high, 22 Sept low, and 12 Mar uptick.

    Do you not find that interesting? Those look like three excellent tramlines to me, with equal distance between all three. Hence my comment that 1790 is the next target. We've been bumbling along between the middle and bottom tramlines since May 12, but have now broken up into the upper tramlines. I'm not saying we'll get up to the 1950 area yet, but it looks to me like we're in a definite upswing now.

    My own personal view is that we're going to continue bouncing about between the upper 2 trams for a while, but will ultimately burst up through the top one.

  • 11. Susan

    (08 September 2012, 11:31AM)  Complain about this comment

    @Beasties Fully respect your views. I learned all my basics from David Jones, Chief Market Strategist at IG Index. He's very much about keeping it simple. Rather than drawing in all these parallel lines, I'm just looking at the gold chart and what I see at the moment is a top in September 2011. As I said, to convince me otherwise I would need to see the current price exceeding the major highs in the last year. It may well do that, only time will tell, but until then my belief is that we are still in a bear market.

  • 12. Beasties

    (08 September 2012, 12:41PM)  Complain about this comment

    Susan, fair does. Personally I feel most comfortable trying to make money from reasonably large moves rather than lots of micro moves with Elliott Waves and the like. Buy low and sell high is much easier to understand! I got badly burned by silver in 2004 when the market went from 850 to 550 in horrific time, and have used that experience as a basis for fear. This year has been a periodic pause by gold, one of several on the way up.

  • 13. Susan

    (08 September 2012, 11:01PM)  Complain about this comment

    @Beasties - Oh Dear not Good! Only time will tell what the outcome with Gold will be and I am sure we will all continue to have differeing opinions. Healthy debate is good, but lets keep it at that!

  • 14. vintrader

    (09 September 2012, 10:42PM)  Complain about this comment

    Have to agree with grosvenor ; does depend on the h/l points you take with Fib and consider 1920 ish as a valid high down to the considerable support@ 153o and so there is still upside towards 1800.With QE being a matter of when and not if , this will add weight to the bulls out there.

  • 15. Zooks

    (10 September 2012, 12:21PM)  Complain about this comment

    Hi John, first of all your newsletter is great. However with this report I think you have missed something- the multiyear support line which was broken in May 2012. Since then, the huge 'wedge' formed, which apparently the price has broke up through from 31st Aug. Now however, at the current price of $1,732 we are at the underside of the previous multiyear support line, so a more significant 'pivot point' that the one at $1,690. I was excited to find this is also at the 76% Fibonacci level on the second from last chart in your article. This level is very significant resistance- if it were to break out from here, it would likely signal a return to the multiyear gold bull market which apparently broke down in May. Such a signficant level though is unlikely to go without a fight and the alternative is this is the top before resumption of a gold bear market for the time being. Events this week (such as the nature of the Feds response) could be seen as a trigger either way.

  • 16. Bronco Bill

    (10 September 2012, 08:37PM)  Complain about this comment

    Forget your fibs, waves, tramlines etc in this gold bull market. All you need are the the 144 day MA and 252 day moving average. Whats happening now is just about identical as end of 2008 going into 2009. Gold dipped below the two averages and then shot up above....as today. If gold can remain now above the MA's while the 144 crosses from below to above the 252 the next stage of this wonderfull bull market will have begun.

Leave a comment

This will be the name displayed with your comment.

This helps us verify comments are genuine. It will not be displayed anywhere on the site and is stored confidentially.

Please keep your comment within 1,000 characters and relevant to the main topic. We encourage healthy debate, but we don't allow insults or bad language. Anything off topic or unpleasant, we'll remove. Enjoy the conversation! Thank you.

captcha To prevent spam-related comments please enter the characters shown in the 'Captcha' box to the left.

By leaving a comment you accept our terms and conditions.


>