SL/PT optimalization – by Ziggy

I am a little reluctant to put this up simply as I am aware of the dangers of optimization, and despite all warnings I know that some folks will always try to optimize. I’d really caution against doing so as in 1 months’ time the optimal level is likely to be somewhere else, but at the same time we can use available data to find a “sweet spot” area.

First I should explain that my data is not “3D”, it is 2 dimensional only. The way I record my data is as follows.

1. I record how far against the level price goes before getting 10 pips up, and then how much further it goes before coming back to the level.
2. I record how far against the level price gets before getting 20 pips up, and then how much further it gets before coming back to the level.

So I can say how Dales levels have performed so far on the basis of moving the stop to BE once 10 pips up, & I can say how it’s performed versus a 20 pip PT – and I can do both of these things at various stop levels.

As I always try to remind, my data is either it does or it doesn’t and pays no regard to any actions that might be taken mid trade.

So is it better to leave a trade alone or to move stops, well there is a lot written about moving stops and basically if you do you are trading your P&L rather than the trade itself, & on every test of every approach I’ve ever looked at you get “better” results leaving your stop alone and letting the trade play out – but having said that we are all human and who is going to let a trade that is way in front come back and take you out for a loss. Certainly I don’t even though I “know” the best approach is to leave it alone.

So here is a table that shows the results of various stop & PT levels on the basis of moving the stop to break-even once a trade has got itself 10 pips in front. Now spread is massively important, as what is +10 for me may not be for you – these are the spread levels I have applied, and I promise you if your spreads are worse than this it can have BIG impact on the results. EU 0.7, AU 0.8, UJ 0.7, UC 1.0. Certainly in the UK there is no reason to accept worse spreads, I have no idea about other countries.

 

Stop	PT	10	12	14	16	18	20	22	24	26
10		230	318	398	430	468	450	392	438	416
11		302	396	478	524	576	552	462	516	502
12		304	408	472	524	582	516	458	516	506
13		385	493	571	621	691	665	591	613	585
14		410	516	588	644	720	700	588	612	586
15		470	588	672	740	828	800	676	708	690
16		542	660	754	832	930	912	798	840	832
17		653	761	865	953	1061	1053	949	1001	1003
18		638	750	844	934	1044	1038	936	990	994
19		627	743	841	935	1049	1047	949	1007	989
20		680	792	882	980	1080	1060	962	1020	1002
21		741	861	945	1049	1137	1101	1005	1065	1049
22		714	836	922	1028	1100	1064	968	1028	1012
23		689	813	901	1009	1083	1049	955	1017	1003
24		632	756	844	952	1026	992	898	960	946
25		610	801	812	920	994	960	866	928	914
26		662	782	876	990	1070	1042	854	1022	1014

 

So to be clear about what this table is showing, the stop level is down the left, the PT level across the top, and if the trade gets to +10 the stop is moved to break even, so trading with a 15 pip stop, moving your stop to BE at +10 and going for 20 pips PT gives a total of 800 pips.  Just to reinforce the spread issue, if your spread is worse you can get stopped out earlier, it will take you longer to get to +10, once there you can be taken out at BE earlier, and you may not actually reach the PT level, bad spreads hurt on so many levels.

Anyway, what you can see is that the “best” PT levels are constantly in the 16-20 region no matter what stop level is applied, and the “best” stop level is constantly in the 17-23 region no matter what PT is applied. On current information this box gives us a sweet spot area.

Now that in itself isn’t the whole story as we wouldn’t want a situation where all the profit is made during a purple patch and the rest of the time it does nothing. Short of me posting up hundreds of charts you’ll have to take my word on how it looks, and anywhere in that sweet spot range I mentioned is ok – but you still have to accept ups, downs, & sideways in the results.

So, I’ll give you a graph using a 20 pip PT with various stops applied – all with moving the stop to BE at +10.

*Results were generated using a Simulated Trading Account and are Hypothetical.

As you can see not much in it, a bit of a flat spot in the middle but pretty damn good. That flat line at the end is coz I have the calcs set in my spreadsheet waiting for new data.

So I mentioned that in terms of pips performance it is generally better not to move your stops. And this is a chart comparing a 20 stop, 20 PT with & without moving the stop to BE, but there are other things to consider – like how often you actually lose. Obviously at 20/20 it’s win or lose, except to say if all that stops me from making a PT is spread I will move my stop to anywhere between +5 and +10.

By leaving the stop unchanged the win rate is running at 63%, so 37% lose rate, by moving the stop to BE at +10 the win rate drops to 45%, but 31% BE and only 24% lose. That may be psychologically important to you.

Results were generated using a Simulated Trading Account and are Hypothetical.

In terms of drawdown there isn’t a great difference, but of course the sample size isn’t that great, 2 charts below.

How you use this information is up to you, I can tell you I trade with a 20 pip stop & 20 pip PT and generally don’t move my stop unless it is only my spread that has prevented me getting a TP. I do however take heed of pending news & if I’m in profit prior to the news I tend to either bank it or move my stop so that I’ll at worst come away with BE.

I personally believe that you have to find a way of trading that suits your own needs / personality – there is no point in changing what you do after a losing period, because this approach does work very well – for us retail traders it is hard to make an income from FX, I am totally convinced that by adopting a consistent approach to Dales levels is key.

-Ziggy