Diversified Trend Trading Approach

Warning...he or she that dares to enter this thread will be sent on a perilous journey into science and the mind. This is a journal of trend followers. A strange divergent species that exist in the trading world who think to much, luv the scientific method, listen to too much music and like the sound of their own voices. Enter with care. You have been warned!!!!!!!

To get the ball rolling....this video is mandatory.






Update - 2018 Resolution - A Change of Direction into the Land of Fun

Ok guys. The New Year brings a few changes to the direction of this thread. I would prefer to see this thread become much more of a discussion thread regarding all things trendy (divergent) and diversified. The gloves are now off and we can create a bit of chaos here..,...so let's see if we can raise the tempo with participation and make this less of a one way diatribe from a few die-hard trend followers.

Over the 2018 year I will be winding down my efforts a bit on this thread and increasing the sporadic brain farts and anything that catches my eye that might be a bit relevant to the general topic of diversified trend following. I would prefer that this thread remains active and is not relegated to the archives from a general lack of participation.

The only rule that we should abide by is a general respect for all participants here and the need to keep things positive and fun. Of course fun posts of general nonsense will be encouraged as well as mandatory music inclusions. Let's turn this into a fun place as opposed to a public library :-)

Cheers C

.....and the best for the New Year to all of you.

Below is the original intro to this thread.....but things have moved on considerably in this thread's evolution from it's original intent.



Greetings traders

I have been a lurker on this forum for quite a while and have been looking at the pro's and con's of sharing my trading approach with everyone here. Bottom line is that I wish to collaborate with the community as I have seen some very worthwhile contributions in some of the threads I have visited, particularly those from the trend following camp who have spared the time to spread the jubilation that sustainable trading for a living can bring.

I am particularly excited with the robustness of the strategy referred to as the Diversified Trend Trading Approach or DTT, as I have been involved in trading for the better part of my life both in the operation of a small fund and in areas of compliance relating to fund management and have seen and experienced first hand a lot of false hopes. Experience has taught me to avoid Martingale's no matter how they are hedged (I was a past Martingale addict and once you are one, it is incredibly hard to pull yourself out) and I also avoid aggressive pyramiding into trades. On Martingales, the key is a regular harvest and the need to ensure you only allocate a small percentage of your trade capital towards them, as you just never know when or even if that day will arrive, no matter how clever you structure your progression. The market has this unnerving ability to sniff out the weaknesses in your creation.

If I was asked, what are the merits of this approach and why does this strategy deserve any attention, then the simple response is this.....and trend traders have all heard it before. A single instrument may experience only a small number of significant trends over a defined extended period. On a D1 chart, maybe as little as 2-4 times a year......A trend trader wants to be on these trends as this is his/her bread and butter......but they also want to avoid or minimise false starts .......so to increase your probability win rate, you need to focus on confirmed trends of substance, and keep control of your trigger finger.......and if your living is dependent on your approach, you want to control your behaviour to prevent overtrading as soon as you sniff a trend.To improve trade probabilities, reduce lumpy cashflow and diversify risk, you also need to keep watch over a large number of diversified instruments so you can afford to be selective in only choosing those trades that 'scream out to you for action'......and you want to strictly control your risk to keep you in the game, pay those bills and keep your family happy. This approach seeks to address these important issues faced by a trend trader.

Warning!!! Ifyou are new to the game of trend trading then I strongly suggest you spend considerable time backtesting, forward testing, trialling a demo account on a single instrument before you launch into multiple instrument trading..and before you even think of going live..not that multiple instruments pose any significant different nuances....but rather that it is essential to understand how to trade trends first before you attempt to diversify. Managing multiple instruments is only going to compound your problems if you are not fluent at trading trends on a single instrument. Your hands will be flying everywhere, your screens will bear witness to grimaces of pain and torture....and your pets will be steering clear of you. This thread should not put you off participation but I strongly advise that you do the hard yards and earn your certificate first....for you could very well be swearing at me in later posts....and I can always refer you back to this original section of the first post, without having to say I told you so directly.

Note regarding the use of correlation and trend strength measures as filters for the strategy. Despite the power of these tools for trend trading strategies in general, I tend to disregard these filters as their inclusion significantly reduces the number of available trades. I recognise the importance of these measures if trading a narrow range of instruments, but given the diversified nature of this strategy, my backtest concludes that they are not necessary....at the moment but may alter in the future as this strategy is honed (as discussed later). What is the critical determinant as a filter for this strategy is whether or not a trend exists with the instrument.

I have a very low risk tolerance and to be perfectly honest with you all, I am not a very good trader.....however my strength, in my opinion at least, lies in risk management and system design. I have been through many trading phases in my life looking at various systems including working with a programmer on a host of EA's, but inevitably find that conditions in the market change thereby rendering these different approaches to the dustbin...meaning that these approaches failed the robustness test in that a truly robust system needs to weather all market conditions.

Given that in the next few years I will be returning to full-time trading for a living my focus has been on the development of sustainable trading systems that focus on capital preservation and income yield....and here we are today.

To be clear, I have not commenced trading this strategy as I lack sufficient available capital at the moment to commit my livelihood towards this venture. Furthermore, despite the relatively low trade frequency, this strategy is for a team that are prepared to operate 24/5 in shifts. I have time on my side to get this all cracking if this strategy stands up to intensive scrutinisation.

What I have done however over the past year or so is extensively backtest and forward test it over the past 688 calendar days commencing 1 January 2013.

Some guidelines for the Strategy:

1. Obviously different trading strategies suit different personalities. This strategy may appeal to those seeking a trend trading strategy backed up with a solid risk management approach.
2. I use IG Markets as my preferred platform as opposed to MT4 due to the excellent advanced charting package provided that assists in managing a diversified portfolio, plus the vast array of products which IG offers that can be traded from currency pairs to indices and commodities. The strategy is currently configured to actively monitor 24 products and a user friendly charting interface with comprehensive trading alerts is essential. It works well on MT4 but can become a pain in the neck when speed of execution is required....but of course it is your choice. I just thought I would mention what I use.
3. I have not been successful developing an EA for this strategy as the determination of the trend is not timeframe dependant which makes programming difficult. Significant decision making needs to be made at D1, W1 and M1 levels Maybe there is a talented individual out there that might want to see if they can convert this into an EA. My feeling is that this approach will only ever be semi-automated at best given the discretionary nature of trend determination with the mechanistic nature of trade management.
4. Despite my lack of success in developing an EA for this, I have removed human discretion at H1 TF where possible.
5. An initial stop is always placed on each trade but immediately following stop execution, the standard deviation channel becomes your exit mechanism so you rarely hit your initial stop. As a result, time in trend is your friend as each bar of progression reduces your loss factor. The standard deviation channel is used to manage the trade at all times.
6. My maximum risk tolerance is 0.5% risk per trade but this does not stop others from magnifying this amount if they are prepared to strike for a higher return. For example, a 1.0% risk per trade doubles returns but also expect an associated doubling of associated risk parameters such as the drawdown. I actually feel that this approach could handle a bit of aggressive risk management so based on current results, a 1.00% risk per trade would be regarded as perfectly acceptable as it is unlikely to keep you awake with cold sweats.
7. For the DTT I use a daily chart to determine if a trend is in place with a standard deviation channel (1.0 std dev's) to confirm the trend and it's significance, and then I go to the H1 timeframe to redraw the trend from the D1 commencement point again using a standard deviation channel of 1.0 std dev's, and then wait for a Donchian breakout, or jump in immediately if it has just occurred using a 40 period Donchian on the H1. I know I don't spend much time discussing support and resistance levels which is a key requirement for trend trading, but this is where the Donchian comes into play by visually confirming that at least in the short term, following a breach, you will be trading in blue sky. Identification of support and resistance is a key step in your analysis on the D1 timeframe in identifying entry points and periods of congestion, but I don't think I need to go into this much in this thread as there are many other threads that do a better job of this.
8. Upon (or just before) breakout of the Donchian on H1 timeframe, immediately redraw the standard deviation channel from the commencement of the trend (on D1) to the bar immediately preceding the entry bar. The lower or upper standard deviation channel (dependant on trend direction) determines your initial stop distance = R.
9. If I achieve 2.5R profit I redraw the trend on the H1 timeframe using a standard deviation channel of 0.5 standard deviations to further confine the trend boundaries and ensure not too much profit is taken off the table before an exit.
10. Note that this strategy does not pick the highs and lows of a trend and tends to execute when the trend is well established and never achieves maximum profit potential.
11. Using these rules you will ensure you are always in the direction of the daily trend.
12. Note that I do not use price action signals such as pin bars, engulfing patterns etc. I simply use the Donchian breakout on H1 to keep things simple for me. This is necessary as I watch a large portfolio and cannot afford to be too detailed in my trend assessment.
13. Also as I am not using traditional price action patterns I therefore use a bar chart as opposed to candlestick charts to keep things clean, decision making fast and avoid letting my brain override the technique.
14. Note that I use the standard deviation channel to define a trend and do not use the classic trend definition methodology of a sequence of highs or lows. The reason for this is that a standard deviation channel is defined using the entire data set as opposed to say 3 or more points in a trend. The basis of this methodology is that I am looking for consistent trend patterns across the entire timeframe I am using. This helps me to validate those trends moves of more regular nature as opposed to those trend moves that may be biased by a few bars in the set. Furthermore the use of the standard deviation channel removes human discretion that is often present in drawing trends. For example, give the same chart to different people and they frequently draw different trend lines. This is avoided through the standard deviation channel method provided you draw the channels from the lowest point in the trend to the highest point in the trend. Another reason the the std dev channel is that I use it as a linear regression tool to define the slope of the trend.
15. When in a trade, always redraw your standard deviation channel when a new high is made (for a long) or a new low is made (for a short). It is essential that when new data is added, the standard deviation channel is redrawn as this manages risk on your trade as it progresses.
16. Note that use of the standard deviation channel effectively acts as a trend volatility measure like an ATR. For example volatility prior to entry will result in a wide std dev channel prior to entry. As a result your R will be wide reducing your position size for your level of risk tolerance.
17. Losses are a fact of life so expect them. Fortunately this strategy strictly manages those losses and through diversification attempts to reduce drawdowns, so let the strategy take care of them.....do not attempt to second guess. Just keep in your mind that you only trade strong trends on a longer term timeframe so momentum from fundamental factors are most likely driving these movements as opposed to random market behaviour. We never predict the future and simply jump on board the current daily trend......until of course it bends.

If this approach sounds interesting to you then I welcome you to this thread. As with standard warnings on other threads, this is for interested members only, who wish to learn or positively contribute.

I am not going to post backtest results (ok I lied...refer to Post 13) as it is a pointless exercise given no live trading has occurred. This will be a task for those who wish to trade this strategy themselves. Despite this however, from my own testing results, I could not be happier and I think those who get involved will also come to this conclusion.

What I can tell you however is this. Over the last 688 calendar days, there have been only 450 trade opportunities across the portfolio which gives you an idea that the strategy is fairly selective in its entries namely due to the condition that a good quality daily trend needs to be in place before a trade entry is signalled. For those who lack the patience to wait for these opportunities, this strategy may not be for you.

If you are happy with all this then I will shortly start posting a few charts to get things cracking.

Regards

Copernicus

Note to Readers
During the course of this thread the approach has been dynamically adjusting to respond to market conditions and strategy additions have been made to supercharge performance results. This approach described above which I refer to as the Diversified Trend Trading Approach DTT represents the core module to this technique and it is a necessary pre-requisite to fully understand this approach prior to moving onto supercharging trend trading additions which can be found later in the thread.

1. Posts 1 to 230 relate to the basic DDT approach;
2. Posts 231 onwards relate to a supercharged variant called the Enhanced Diversified Trend Trading approach of EDTT. Comprehensive details regarding the EDTT are found on post 231.
3. Pay particular attention to post 360 onwards relating to a methodological recap of the EDTTI. RedLineFred has kindly compiled this recap into a pdf for easy reference.

Unfortunately to understand this approach in its entirety, you need to do a bit of reading for Posts 1 to 230 are still compuslory for understanding and get's you a beginers certificate before you can advance to supercharged variants. This thread is not a free ride for a trader and demands a bit of effort on the readers part if they wish to participate or trade this approach. There are many nuances that must be understood before committing your hard earned dollars to this technique. I wish you all well, but please do your homework first.

I will use a nice long trend that I am currently riding now as an example which is the USDJPY. I have taken two bites at this trend but the following example defines the most recent trade example which I am still riding. Note that trends of this magnitude are not common but it helps to use this for explanatory purposes.

The Daily Chart is used to identify if a trend exists (effectively a trend filter) that can be well defined by a standard deviation channel as displayed below. Note that the trend commences on October 15 and reached a high on 26 October before I got interested.

Note that this is the most discretionary part of the strategy. The volatility range of the daily bars within the standard deviation channel is what I use to identify if a valid trend exists. I am not after significant volatile swings within the standard deviation channel or exceptional bars that dominate the trend. What I am after is consecutive orderly bars within the channel that signify that a trend was present throughout the entire data sequence.

Some of you might suggest that slope of the standard deviation channel should be used as a filter however in my experience, this is not actually necessary. Obviously those trends with a higher slope will tend to have shorter duration but using slope as a filter can eliminate some significant trends or alternatively put you into the end of a trend where significant volatility is common. Experience teaches me that trends end in a crescendo and some of the biggest but short term moves are during what I term the capitulation period for those traders who become significantly exposed in positions against the trend. My preference is to already be established in a trend before this happens. The reasons for this is that during the more orderly phases of the trend the standard deviation channnel is narrower and as a result your position size is larger. Once you are riding a trend with a good position size and entering into the significant profit zone, the tail end of the trend, where often the exciting movement happens in the trend direction really serves to 'ice the cake'.

Using a standard deviation channel (1.0 std dev's wide), define the trend from the low to the high. Note the green horizontal line which was placed on the daily chart to make me drill down to the H1 as price approaches it.



My next post will display the entry on H1 timeframe.

Having established that a well defined trend using the std dev channel was in place using daily chart (above) then drill into H1 chart to redraw trend on H1 and wait for Donchian breakout (40 period). Note that I entered before the green horizontal line and on H1 the Donchian had been breached.

The entry was late however (after the breakout of the Donchian) so not perfect, however the dotted yellow line represents R and is used to determine the position size on the trade. Remember that I use 0.5% risk on trading capital so for a $200K account that represents a maximum risk on entry of $1000. The dotted line is therefore converted in the position size (after taking commission spreads into account). In this example I acquired 11.8 mini lots.

The entry was at 108.47 on 29/10/2014 at 18:00 hrs with a stop placed at 107.685.

The chart below displays the entry.



The next post will display how to redraw the standard deviation channel as the trade progresses.

In a long trade, as new highs occur, redraw the standard deviation channel (every time a new high occurs) from the commencement of the trend to take in the new data (refer to chart example below). The converse for a short trade.

Keep in mind that when you get to 2.5R profit a different standard deviation channel is drawn as with more data the channel naturally widens and you do not want to take too much profit off the table.

Note that if the trade goes against you the exit condition is when it touches the lower standard deviation channel. Set alerts on this channel by overlying a line (yellow) over the channel to take action when this occurs.

The trade is progressing well and is deep in profit at approximately 2.5R.

At this point redraw the standard deviation channel with 0.5 width to narrow the band.



The next chart will display where we are at currently deep in profit.

We are currently deep in profit with no sign the trend will end. Note that this is a spectacular trend so don't think this will happen all the time. As you can see however we do not have a profit target with this strategy as we let the profits run and we wait till the standard deviation channel is breached before exit.

Let me know when you want to see further examples or if you have any queries.

The reason why I refer to this as a diversified approach is that I trade 24 instruments. The principle behind this is to only trade significant trends yet assist in achieving a regularity of income stream.

The consistency of income and minimal drawdowns is confirmed by the following equity curves achieved since backtest commencement to current day/time (excluding those trades that are currently live). Note that you should treat these with scepticism until you can confirm similar results from your own testing.



I will post next a screen shot of the chart setup I monitor to reflect the diversified approach.

The chart setup is crucial to the strategy. I hate missing an opportunity so trades can be executed at any time at all hours. Comprehensive alerts are therefore critical to the strategy. For many of you a 24/5 lifestyle is out of the question but a team of traders can manage this effectively.




Note that those charts in hourly mode have trades running. Those charts in daily mode are in wait and watch mode. Currently of the 24 instruments, 9 trades are live with the balance in wait and see condition. Note that the last 4 months have been excellent trending conditions and in particular October following Japan's quantitative easing announcement.

Please absorb for a while and if you have any questions give me a hoy!

So basically we trade on trend beginning and uses divergence as exit point?

As with all trend trading approaches the intent is the same, to get into the trade once a trend is confirmed and let the profits run. The use of the standard deviation channels however is for trade management purposes predominantly, to validate that a trend is in place at all times, when to exit the trade (when the standard deviation channel is touched) and a device to remove human discretion.

Given the daily trend entry filter we will never be able to enter a trend at its beginning. We only enter when a trend is confirmed on D1. Also given the exit condition on touching the extreme standard deviation channel you will never exit a trend at its peak. That's just the nature of the game. :-)

Alright.

IT is a great system with low risk and Trade in Daily Trend direction after confirmation.

You produced equity cover. If possible do you have details of these trades i mean pair, buy or sale, entry time, and exit time, entry price closed price etc.

I want to understand the pattern on Daily TF and I hr TF for these trades.


Thanks

Hi mkfx

Please find attached backtest history. Note that Trade Capital Account is $200K for this backtest. All profit is taken off the table and not re-invested. No account compounding. Note also that some trades are still active hence no close details.

Regards

Copernicus :-)

Copernicus Thanks for details of trades.

What is brokers server time now Is it GMt+2hr

Thnaks

Hi mkfx - Trade time for the backtest and Chart time is GMT+1.

Regards

Copernicus

I gone through stats and very promisiing. I have dropped you a PM (message )

Looking forward to participate in this system.


Thanks

Hi again

I thought I would commence running a trade journal with updates hopefully everyday to get those interested into the swing of things.

I will therefore give a synopsis of current live trades and their entry conditions and then update the journal periodically.

Here goes.

Below is a view of the 24 instruments I watch. Currently I am in 9 active trades with the balance of wait and watch. Those charts in daily mode are in wait and watch mode. Those charts in hourly mode represent a current trade in progress.



Details of all active trades are as follows.



I will give an overview of trade entries in the following posts.

Regards

Copernicus

Starting with AUDJPY. Note that this is the second trade undertaken on this trend as the prior trade which commenced on 23 October was closed out in profit following a breach of the 0.5 std dev channel.

Anyway, following the prior trade close this trend traded horizontally between 31 October to November 11 signalling that it was taking a breather. The green dotted line represented a point I was happy to consider entering if the trend resumed. On 11 Nov 2014 the daily bar commenced its breakout towards my zone of interest.






Switch then to the H1 timeframe to consider an entry on a Donchian Breakout. The following post will provide details regarding the entry on the H1 timeframe.

Entry was taken when the Donchian on H1 was breached with a new high signalling a breakout in the direction of the Daily Trend.

A standard deviation channel was then drawn from a prior significant low (or swing low) the the bar immediately preceding entry to determine the initial stop distance (R) and position sizing. Refer to the trade entry details in Post 17 to determine the values.

Now you may be asking a question at this stage about why the std deviation channel prior to entry was not extended back to the swing low on November 16? In my prior trade discussed above, this was where my standard deviation channel was actually drawn from however following a successful and very profitable trade I was not willing to extend the channel to this commencement point as a significant component of this trend was over. If a channel had been drawn back to Nov 16 the wide channel would have resulted in a far greater R reducing my position size accordingly. Now this isn't really a problem but following considerable profit on the first move of the trend I was prepared to up the anti. In doing this however you need to ensure that the width and slope of the standard deviation channel is sufficient to carry this risk......don't worry, as you progress in trading this system this will become clear to you.

......so in this instance I commenced my channel from the low of 30 Nov 2014 and extended the channel to the bar preceding entry.

As the trade reached new highs I continuosly redrew the standard deviation channel to the latest high and you can see that over time the channel widened. Note that this trade has not yer reached 2.5R at which point I will use a 0.5 standard deviation channel in place of the 1.0 std deviation channel currently used.

Note that the channel extends to the highest high on 16 November. Until a new high is made this is where the channel sits. The yellow line that is superimposed on the lower channel represents the exit point. Set an alert on this yellow line to ensure you close the trade when triggerred. It is likely that this exit point will be touched very soon. At this point therefore close the trade and set the chart back in daily watch mode.

Now to CADJPY. Once again this currency pair has been on a great trend and I have taken numerous bites on this cherry with a significant win result on one of them.

The preceding trade was entered on 11/11/2014 at 07:00 GMT+1 but a glance of the lower std dev channel knocked me out at $2.40 in profit on 13 Nov 2014 so I waited again for a re-entry if the trend continued, with the intent on widening the std deviation channel to give room to move by extending the channel back to the low of 4 November 2014.



Note the green dotted line that was placed to confirm a trend continuance on breakout. As price moved towards this entry point, we hone in on the H1 chart.