Diversified Trend Trading Approach

Copernicus

In post 42 above you mention the following


1. EURUSD - enter at breach of 1.2317
2. EURCAD - enter at breach of 1.3825
3. NIKKEI - enter at breakout confirmation of 17855 and so on

Unless I am missing them, these levels don't appear to line up with past Donchian Channel levels. If this is the case how were they determined?

Thanks

Hi pakeha

On the D1 'wait and see' charts I set predictive breakout lines (green dotted horizontal lines) based on simply a visual take on zones of clear blue sky a sufficient distance from the current price for a D1 trend to emerge. They may coincide with D1 donchian channels or they may just represent a sufficient distance for a trend to emerge before I get interested. With the number of instruments I am watching, these green lines stand out like proverbials to me to keep me focussed. Have a look at Post 42 and the screeshot to see those green predictive lines.

When price continues towards these lines I am just keeping a visual check of the D1 trend to ensure it is sufficiently consistent for me to get interested. If it 'shouts' out trend on D1 and price is approaching the predictive lines I then zoom into H1 for a Donchian breakout on the H1 timeframe.

What this ensures in reality is that as soon as I switch to H1 timeframe a Donchian breakout should be occurring and I will jump in immediately, or may have already occurred, which means I may jump in immediately on a high momentum bar, or if patient (which I never am at this stage) wait for a future H1 Donchian breakout on H1. What this actually means to me is that entry technique is not critical for this strategy provided you are trading the bar on H1 of a lower low or conversely higher high.....but better traders than me might argue the point on this.

The good thing however is that the strategy is sufficiently robust to accept a sloppy trader like me :-)

Bottom line is this. I am discretionary on the D1 timeframe but intensely mechanistic on the H1 timeframe. I know this doesn't really help for the D1 timeframe but unfortunately those who trade this strategy need to do some significant work on identifying trends that they are happy to trade.....time at screen and backtesting....and backtesting.....and .........appears to be the only way of doing this without a superb EA at your disposal to do the trick.

Don't change a thing, its not worth it for a few pips unless you are scalping and u need these 5 - 10 extra pips and ur profit target is 10 - 20 pips away which is not the case, drawdowns is a part of our trend following approach to trading and we embraced it as a cost of doing business .

best,
B.R.

Much obliged kind sir. A much needed confidence booster. Second guessing yourself is something I excel at :-)

Hasta la vista ASX200 :-(

Close on touch of std dev channel.

Note the progressive trailing stop from entry associated with the upper channel (for a short) defined in yellow of the std deviation channel of 1.0.




The ASX200 now reverts to D1 on 'watch and wait' mode with predictive lines set for a trend resumption on breakdown or sufficiently far away for a new long trend to possibly emerge (in this case just above D1 Donchian for a long).

A bit of trade management for the day.

CADJPY is now 2.5R in profit. Adjust std dev channel to 0.5 width to narrow band and reduce potential to leave profit on the table.

Copernicus

Thanks for the spreadsheet - not many put in-depth back-forward tests on FF so this is appreciated.

Re: USDCAD, there are OPEC meetings this week in Vienna and associated news events (Thursday) which may impact this pair a little in your favour (i.e. downward move in USDCAD if cuts announced). Iran is apparently proposing production cuts of 1 mil b/d, but other OPEC nations will need to agree, especially the Saudis.

Thanks for the update on USDCAD pakeha. I will keep a close eye out on Thursday.

With regards to your suggestion to consider GBPJPY, I have been thinking about this and will probably soon expand the portfolio as I have a good handle on the 24 pairs and trading them is currently not a significant hurdle once you get your head around the strategy.

Given trade frequency on 24 instruments is currently about 0.5 trades per day and max trades on any one day over the last 2 years so far is 9 trade actions (eg. entries or exits), I think I can comfortably handle some more to further diversity the portfolio without turning into an octopus. As you may have seen in the spreadsheet profit results, the diversification has been a significant benefit to me so far......PS This is exactly what Curtis Faith from 'the Turtles' has also found.....and I can see why. Further diversification in my opinion is only going to enhance the result.

To be fair to the backtest results, if I include GBYJPY alone, this will obviously significantly bias the result in the positive direction given the excellent recent trending history on all JPY pairs. To be fair I will await a further screen as 'she who must be obeyed' flogged one of my screens when her's broke down and I am now down to two. With another screen I can up the anti with the portfolio on watch with another 6 to start with and I will choose a diverse range including GBYJPY to keep the backtest fair, then run an arduous backtest across the additional 6 from 2013 and include the results in my analysis. It might be a while away...but a worthwhile exercise at least.

Regards

C

Below is an annotated chart of the DAX which I would like to use as a current example that reflects the dynamic nature of the standard deviation channel. This explanation hopefully will give you a deeper appreciation of the merits of the std dev channel as a trend management device.



Please refer to the original standard deviation channel (henceforth SDC) used to define the trend volatility and momentum. Trend volatility is reflected by the width of the SDC, whereas trend momentum is defined by the regression trend of linear elements of the channel. The central channel actually represents the line of best fit of bar closes contained within the entire channel from commencement to end.

The width of the original SDC is used as a basis to determine position sizing where R (distance between entry and stop) is defined by referring to the entry price and the extremal channel of the SDC. For a 'long trade' you are looking at the lower channel. For a 'short trade' you are looking at the higher channel. If the prior price movements of the original trend have been volatile (namely volatile closes), the wide SDC will reduce your position sizing as R increases. Furthermore the extended width will give the trend more room to breathe and takes into account prior volatility. You can see that the SDC achieves this without you actually having to think about it.

While the std dev channel takes care of volatile closes, you still have to look at the bars contained in the trend movement from channel commencement to channel entry to consider the range of each bar as this can influence your decision to trade. For instance, I frequently avoid entering a trend where prior price movement on most prior bars has been volatile. My preferred trend of choice is a consistent and steady trend.

Now have a look at the chart again and note how in the redraw following entry where new highs were being made, that the std dev channel is 'ratchetting upwards' as detailed on the chart. The additional closes from new highs are increasing the trend slope to reflect increased momentum. The beauty of the std dev is that it is dynamically changing to reflect new market conditions without you having to do anything at all. The increased slope of the trend means that you reach breakeven and profit territory quicker if trend momentum is in your favour.

The above is also found in Bollinger Bands that actually are composites of the end points of these standard deviation channels but I actually find the linear expression of the std dev channel more helpful as a trend trader. I guess it's because of my linear mindset/ tic.

As more examples come to bear, I will update you with further strengths of the std dev channel as a trend management device.

Regards

C

Hasta La Vista GBPUSD. Close of trade on H1.....*sobs hysterically*





Back on watch and wait mode on D1.

Cop., my experience tells me that the H1 is not reliable as the H4, on the expense of some less pips (or more in the case of a fake out) the H1 TF tend to display much more fakeout than the H4 so a word of advice will be to consider closing your trades ( or entering) on a H4 closing basis and better yet D1.

I speak from experience and a lot of pain.

Best,
B.R.

Thanks for the heads up B. :-)

If I go to the H4 or D1 timeframe with this technique as you suggest it will be far more manageable if I want to up the instruments in the portfolio. I will have a think about it for a bit. I will, for the purposes of this backtest/forward test and this thread, continue on with H1 entry but if I get this approach going on H4....and I can see your point, I will start a new thread for the H4 entry or even D1 entry to keep things seperate and compare both approaches.....but I can see some work ahead for me. It might however be well worth it....unless there is some interested trader who understands the approach and is willing to run the H4 test or combinations of H1 entry and H4 close etc... over the same period. I can supply systems to assist such as Dbase and analytical spreadsheet and provide assistance in operating these systems. On second thoughts...that's a big ask as it's probably best to allow time for interested folk to get the swing of this approach.....so when I have some time up my sleeve I will give it a crack. :-)

C

Farewell USDJPY. It was great knowing you. This is why I am glad I am a trend trader. Looks like all my JPY pairs are running out of puff. Probably more to follow soon.

H1 Exit



.......and now back on watch and wait on D1 timeframe.

DI Watch and Wait




Regards

C

C

The dotted lines of the D1. I realize they are discretionary above/below the daily high/low, but is there some logic to their placement (e.g. x pips above the high or below the low or round numbers)?

Thanks

I will give you an idea of why I placed them where I did for the example above. I wish the D1 element of this strategy was non-discretionary but unfortunately I have found that the human brain is required for a host of subjective decision making at this stage.

The upper dotted line is well clear of the highest high to give room to avoid anticipated fake-outs on smaller time scales. There is likely to be significant institutional activity at these levels and the likelihood of 'sucker' rallies will be high. Clearly it is in blue sky at a point where I am confident that the bull run will continue with momentum. We have already had a great trend and there will be lot's of traders of all time durations wishing to enter for a resurgence or take profit if already in a trade with inevitable shorters lining up with the rationale that this has got to have a correction sooner or later. Inevitably on trends like this there are zones of significant uncertainty and if possible I want to avoid them.

.....also remember that we have benefitted from a great trend so I am starting to go in defensive risk management mode to protect the hard earned profit. Immediately after trends like these considerable periods of uncertainty where the market may be range bound for a considerable period and you never know how long these will last. Lock those profits in and hold onto them with your life.....that sort of logic also influences my decision making at this stage.

Of course these subjective judgements may all be BS........however I have been in this game a very long time now and have experienced moments like these before. My think time in this strategy is invested in D1 'Wait and Watch' while on the H1 timeframe most of the trade management is left up to the SDC, apart from the period immediately before entry (discussed before) when an assessment of price action is important.

The lower dotted line simply represents a sufficient distance away from current price for any reversal to take form in the shape of a trend. This lower line is continually amended based on price action during the interval it is in watch and wait.

As time unfolds on a horizontal range bound price history, the significance of these predictive bars, and their placement might change.

I will try to keep all interested aware of my decisions with regards to these predictive lines so they can apply their own logic to their placement.

C

Goodbye JPY pairs.......as the soldiers fall I salute you /tic :-)

NZDJPY H1 Exit






NZD JPY D1 Watch and Wait




CADJPY H1 Exit




CADJPY D1 Watch and Wait




AUDJPY H1 Exit



AUDJPY D1 Watch and Wait

C

CAD retail sales and US GDP news today at US open should determine the fate of the USDCAD trade today. Pair at 1.1300 resistance at the moment and closing in on the upper SD line, so its possible pre-news activity may do this trade in. Do you see the same thing?

Hi Pak

It's not looking good and never has following the sloppy entry I did. I deserve to get a slap around the chops with this one. :-(

Sounds like you are getting the hang of the approach :-)

Here is the current status. Is this what you get? A loss is shaping up of about 0.8 to 0.9 R.




Regards

C

PS AUDCAD is shaping up quite well on D1 and my trigger finger is getting itchy. Your thoughts?

Why the interest in AUDCAD? Below are reasons why this approach likes the emerging setup.

Below is D1 overview



Note the consistant price action of this emerging trend. The slope is relatively high for the type of trends I typically take with this approach but this signifies the potential for a short but sweet ride and are perfectly valid trends to ride. It is progressively getting closer to that D1 predictive line of 'blue sky' which is also a D1 Donchian breakout so I will be watching for a breakout with momentum if it breaches.

Over to H1 for a squiz



A nice consistent trend that works well with this approach with only one significant volatile section. It's looking good for an H1 donchian breakout at the predictive line. A relatively narrow channel will allow for a good position size on entry. The higher angle of slope means that we possibly reach breakeven and profit territory quicker but in general these type of trends will produce a 1-2 R profit so don't expect too much.

Entry on AUDCAD.

H1




As it turns out the channel width restricted my position size using R to 9.9 mini lots which is a bit smaller than normal with risk at 0.5% of trade capital for the trade. This was contrary to my comment above.....but trust the SDC....not my ramblings. :-)