Systematic Trading research and development, with a flavour of Trend Following
Au.Tra.Sy blog – Automated trading System header image 2

Profit Targets and Trend Following

September 30th, 2010 · 22 Comments · Backtest, Trend Following

target-iksuokaz

OK, Profit Targets. You’ll tell me this is the anti-thesis of Trend Following, with one of its strongest mantra being:

Cut you losses short, let your winners run…

Indeed, Trend Following relies on a few positive outlier trades (from a fat-tailed distribution) to derive most of its gains. Stopping trades from running as high as they could might not sound like a good idea.

But it is also a tempting idea to try and exit the trade before the typical Trend Following reversal, if only to try and smooth the equity curve.

I know about it: I have been there before with my discretionary trading. It is so easy to fall into these two traps (let your losers run and cut your winners short!). That is one of the reasons I decided to trade in a systematic fashion.

Anyway, I digress… Despite hearing many times over that take-profit orders are detrimental to a Trend Following strategy, I wanted to test this for myself (always double-check your sources…).

Profit Target Test

I ran a quick test using a dual Moving Average cross-over system (the medium-term one from the State of Trend Following report)

The test involved stepping through two different parameters:

  • Target profit level (expressed in multiples of entry ATR).
  • Percentage of position to sell upon hitting the target profit.

There are three main cases driven by the position reduction percentage:

  • 0% – no profit target: the position runs as per its normal course (ie. normal system exit signal)
  • 100% – position is fully exited at the profit target level
  • a value between 0% and 100%: position is partially exited at the profit target level

When partial take-profit orders are executed (eg cut 50% of position after price has risen/fallen by 6x ATR), the rest of the position gets liquidated when price rises/falls by another ATR-multiple.

The Results

Below are the results in graphical 2D and 3D form. Each surface chart represents the metric output (CAGR or MAR) for each stepped parameter pair.

CAGR results in 3D and 2D – click to expand the charts:
CAGR-3d

CAGR-2d

MAR results in 3D and 2D – click to expand the charts:
MAR-3d

MAR-2d
 
Both CAGR and MAR clearly display their maximum at the 0% level for position reduction – which means no profit target.

This is probably just one more example of the benefits of letting winners run.

Of course this is just a one-off, very simplified example and should not necessarily be used for generalisation… But it should not be a surprise that Trend Following’s performance is negatively impacted when breaking one of its fundamental principles….

There might be still some benefits in trying to identify potential reversal points – as is sometimes being advertised by some CTAs (BlueTrend comes to mind) – in order to cut positions early. But I believe the approach needs to be more specific/sophisticated than a blanket ATR-multiple Take-Profit order.

EDIT: Updated Test Results

I realised after writing the post that there was a “trend” in the results with performance seemingly improving as the profit target moved higher. I decided to complete the test by running it all the way to 40x ATR as profit target. Below are the results:
CAGR-40x-3D
MAR-40x-3D
 
We can see some small pockets of over-performance on areas outside of the 0%/no profit-taking orders, but given the variability within the results, they may well be down to random variation…
 

Target picture credits: iksuokaz via flickr (CC)
Related Posts with Thumbnails

Tags:

22 Comments so far ↓

  • Tom

    I’ve come to the same conclusion in my testing. Time based stops are a different matter.

  • Jez Liberty

    Hi Tom – thanks, somehow time-based stops sound intuitively less “logical” to me but, hey, who said there was logic in the markets…
    Thanks for the idea, I’ll probably run a follow-up post on this.

  • Jez Liberty

    Hi Tom, was just running some tests on time-based exits applied to the same system as tested on this post and could not see any improvement in performance – actually performance (CAGR and MAR) increase proportionally until the point where all positions exit before the time-based limit (something like 800 days).
    Were you referring to time-based stops applied to Trend Following systems?

  • Pretorian

    Jez, maybe you remember it, but Curtis Faith seems to be a big fan of time based stops and he offered some backtest results in one of his books. However these were based on strategies WITHOUT stop loss, so they may perform in the long term but accepting some extra risks.

  • Nick Radge

    Jez,
    Enjoy your thinking and thanks again for the insights. I am primarily a stock trader now but the premise remains true as commodities – upside volatility is a lot more, especially in hot sectors such as resources and energies. Mid cap and lower cap stocks also tend to increase that upside coefficient, albeit with some extra volatility but the rewards are there.

    Thanks again.

  • Motomoto

    Curtis Faiths latest book “trading instincts” has some interesting comments on this.
    There are a few other things that also need to be looked at if looking at taking profits…..
    what is the reentry criteria?
    Is it better to apply a profitable counter trend trading system, than muddle up the current system?
    As per Curtis’ book – is there an element of discretion (unable to be back tested) that can be used?
    With the exit rate varying between 0-100% I have seen….but dont know exactly how it works, a constant allocation to each instrument, based on levels between -10 to +10. But I think this sort of portfolio weighting system is very detailed.
    Otherwise….always a good read Jez. thanks

  • lperepol

    I have done a similar test and found that taking profits at high levels does not hurt. The test used limit orders to sell and the test got back into the same instrument on the next day. Sells limit orders were placed when the instrument made X consecutive up days. The test showed that buying after X down days did not have much of an impact.
    See Price Persistency by Gordon Gustafson Stocks & Commodities V. 20:1 (44-47)
    Gordon shows that system that buy on X down days and sells after X up days has some merit.

  • Howard Bandy

    My research and experience indicates that profit targets are effective exits when the holding time is relatively short and the amount won by winning trades has a narrow distribution. (The same reasoning applies to maximum holding period exits.) When the distribution has a few large winners, use of profit targets lowers overall system performance. It is difficult to set the level of the profit without peeking at the trade-by-trade results and creating a curve-fit system. An alternative to profit targets for trend following systems that works well is a trailing stop — the parabolic portion of the SAR, or chandelier.

    When multiple exits are used, each exit should occur often enough so that it can be judged on its own.

    Thanks for listening, Howard

  • Nizar Mahri

    I have done some testing on this topic some years ago and my findings have been similar to Howards.

    Nizar.

  • lperepol

    Paul Whitfield, from Investor’s Business Daily — The Big Picture (2010/10/03) commented today on top rated stocks and their draw downs in single day:

    ” Green Mountain gapped down 16% on Wednesday alone after it disclosed that the SEC was conducting a probe.

    Stocks that gap down that much in one day usually have further to go. Historical examples include
    Taser International (TASR), Shuffle Master (SHFL), Travelzoo (TZOO), Smith & Wesson (SWHC),
    Crocs (CROX) and VeriFone (PAY). Each stock fell further after one-day plunges of
    15%-46%. None of the six has ever surpassed its previous high.”

    An end of day trend following system that follows top rated stocks can have draw downs in excess of 15-46%.
    Especially, trend following systems that use moving averages.

    Paul Whitfield comment gives one some thing to think about!

  • RiskCog

    Thanks Jez, this is a good topic. I have resisted adding profit targets to my automated systems so far. But when I see all the profit I expect from a trade run up at the start of a holding period, it is tempting to just bank it…

  • Tom

    Jez,

    Yes it is a trend following system, though a little different to the ones you test here. The time based stops benefit the portfolio rather than the individual trades.

    Here’s the backtest results with time stop :
    Trades 779
    Kelly Criteria 0.29
    Longest Drawdown 1.15 yrs
    Sharpe 1.52

    And without:
    Trades 521
    Kelly 0.32
    Longest Drawdown 1.44
    Sharpe 1.26

    Spreadbetting spreads applied to the results.

    I don’t have a problem with the logic. After all, how long does it take even the slowest investor to get on board a trend? And for when the good (or bad) news keeps coming, or a bubble develops then I have a longer term system running.

  • aaronh

    Interesting comments so far in this post…..

    I found that adding a profit target and taking 50% off a position once the profit target has executed, < CAGR & CAGR & < MaxDD results. Due I feel to the increased accuracy and putting some profit dollars back into the trading account. This was a pleasing result for 2 of my medium term TF systems.

    I'd be interested to hear how others found their monte carlo results affected by taking partial profits….

    cheers

  • aaronh

    …some text dissapeared in my post above..it should have read :

    I found that adding a profit target and taking 50% off a position once the profit target was reached resulted in Monte Carlo results with a higher CAGR and lower MaxDD. Compared to Monte Carlo results without using a 50% profit target where they resulted in a lower CAGR and higher MaxDD.

    This effect I believe is due to the increased accuracy and taking of partial profits. This was a pleasing result for 2 of my medium term TF systems. Anyone else find similar Monte Carlo testing results with partial profit taking ?

  • Jing

    It is maybe an interesting idea to use a very tight trailing stop after the price has reached 6 times ATR instead of profit taking exactly at 6x ATR. If we use a very tight trailing stop, we may consider reentry after the price makes new high/low. The combination of very tight trailing stop after making a few times ATR and reentry the trade may have better total profit than simple profit taking and more smooth equity curve.

  • Lawrence

    I agree with Jing. Find a device that identifies when an instrument is trading at out sized return ( be it 6 times ATR of what ever) seems like a good point to to put a sell limit order at. Then if the instrument starts to trend in the favorable direction get back in. Simple profit taking does not work well and can reduce overall performance.

  • Jing

    Hi Jez, I just found in currency markets, quick profit taking/re-entry and tight exit can improve the total profit and reduce volatility/maximum drawdown. Take simple channel breakout for example. “Quick profit taking” simply means after price advances 6 times ATR, sell long position if today’s low below previous 2 days’ lowest low. “tight exit” just means using 10-day low to exit long position. After adopting above 2 changes to channel breakout, most 7 major currencies’ profits improved, especially for weak-trending/choppy markets, such as 2000-2010 GBP/USD, USD/CHF, USD/JPY. Thus, in choppy/weak-trending markets, quick profit taking and tight exit combined with reentry not only improve total profit/profit factor, but also reduce volatility/maximum drawdown in the currency markets from my testing. And I think it makes sense because it avoids the problem of giving up a large portion of open profits faced by many traditional trend strategies.

  • Stefan

    Hi Jez,
    What is the portfolio structure? Is it somehow balanced (the same number of markets per group)? Finally do you normalize position size and what sing rule do you use?
    Thanks,
    Stefan

  • Jez Liberty

    Stefan, the “State of TF” posts (latest: http://www.automated-trading-system.com/state-of-trend-following-in-may/ ) should have the answer to your questions in the Appendix section (I used the same portfolio for this test)

  • Stefan

    I do not see stock indexes in the portfolio. Sometimes developers avoid them since performance is not great with old TF systems. Is it the case since there are symbols I am not familiar with?

  • Jez Liberty

    There are quite a few actually – some of them quite “more” exotic (TOPIX, Taiwan Index, Bovespa, etc.) but also more established ones (FTSE 100, DJIA, EuroStoxx 500).
    The link to the list of instruments (with symbol AND description) is here:
    http://www.automated-trading-system.com/wp-content/uploads/2010/08/Instruments.html

  • Steve

    I use discretionally adaptive trailing stops. I’m harder on losers than I am on winners. Any trade that starts to lose gets the stop tightened on it by the amount of the negative move sice the last point of observation (typically 4 hrly during a trading day), and winners have their stop tightened back to their original 2.9XATR on daily chart, once a day. Works well.

Leave a Comment