In Price Action Trading Strategy

“What’s the best place to put a take profit or stop loss?”

If this post doesn’t clear it up, I’m not sure what will. Let’s forget about price action for a moment and let the numbers speak to us. Many materials available to FX traders (including those discussed on this website) talk about certain price obstacles and other things that get in the way of profit. While they're very important to know, this post is an attempt to clear up a lot of the ambiguity about bringing home the best results simply by knowing yourself and looking at your historical data.

One of our readers, Steve McGrechan (a.k.a. Dingo) went out on a limb one day, and about 6 months ago did a thorough analysis of the trades posted on this site to find out which take profit / stop loss variables made the most money.

Any programmer of automated trading strategies will tell you that the number one thing about boosting any system’s performance is through simple optimization. Whether it be trading rules or parameters, you need to optimize an automated strategy with recent data to ensure it makes money. It’s a requirement.

But for manual systems, ones you buy and ones found all over the internet, you rarely (if ever) see any type of “optimization”, so to speak. The automated strategy programmer will tell you that you could take a basic stochastic crossover system, and optimize it with recent data to perform well, whereas standalone, it’s a complete loser. Manual trading is no different.

For us, we have our systems, we fade the current trend or trade right along with it; those are the two basic conditions. All of the trades that have ever been posted on this site I can easily categorize into one category: fading (going against price action) levels at areas of very heavy support or resistance. So we assume we are using one strategy in the case of optimizing the SL/TP of this one.

The Test

Below is Steve’s analysis, take a good look before you proceed:

The obvious winner of the bunch was the 10 pip stop, 100 take profit.Cleary, the 1:10 risk/reward paid off the most, and in the other analysis, expanding the take profit, while doing the same with the stop loss never improved things.

Steve only went as far as a 100pt take profit, but I knew we could do even better if we keep expanding the risk / reward, maxing out at 250 pips take profit. Here were my findings:

And this is what the equity curve looks like at the most optimal level (1:25 risk/reward), trading 5 contracts:

Now the risk/reward does indeed jump around in terms of what’s better or worse. So it’s not 100% linear, but close. And at one point there will be a situation where things max out, so to speak, but optimization will tell us where that point is. Also, your win/loss ratio will certainly decline, but your account balance, of course, does the opposite.

So how do you apply this information in your own trading?

Everyone trades in his or her own unique style.They execute differently, manage differently, etc.Additionally, people take trades based on numerous strategies.

For example, assume you have 5 different strategies you use for entering trades:

Trend following:

  • Breakout level
  • Thrust bar continuation
  • Buy/Sell trend pullback in line with macro trend

Countertrend / Cyclical:

  • Major support or resistance countertrend
  • Intraday price exhaustion scalp

So for each of these five types of setups, you would need to optimize your SL/TP separately using historical entry execution, and follow through accordingly.You could even include trades you were about to take, but didn’t for whatever psychological reason.It doesn’t matter because the ultimate goal is to just find out, based on your past actions, what the most optimal place is to put your SL and TP.That’s it.

We segregate optimizations based on entries for the simple reason of diversification. Like anything else, diversifying our strategies allows us to be nimble and execute based on various conditions. A diversified trading plan typically equals adaptability, so long as the means for entry are not closely correlated, of course.

If you’re demo trading, you’re fine, if you’re live trading, get a demo and take every easy trade you see. Gather together a substantial-enough history to take a look at plop into excel at some point.

For now, our numbers tell us that careful planning and excellent risk/reward wins the race, which is to be expected. But you should always know ahead of time what kind of setup you are taking, and expect the unexpected, whether it be good or bad. Your ‘optimized’ SL and TP for your own trading style will lead you to the rest.

Again, this is nothing more than a very simple, mathematical representation without regard to price action; simply the facts of knowing what your ideal choice of entries are, and how you execute. The parameters above might suit my particular strategy but certainly not all. Everyone trades differently.

Steve got my thoughts running when he emailed me out of the blue one day with this among other analysis; if you have any thoughts on the topic or can speak fromexperience, please feel free to comment below.

P.S. This is the first post I have made in quite some time, but I have more materials on the way; quite a bit, actually. This site continues to undergo technical updates; more features will be added in the upcoming weeks. Emails are not updated on every post to the blog in order to limit spam, but Facebook and Twitter are. So if you have interest in following I look forward to seeing you on the list!


4.86/5 (7)

ParaCurve is free and has been since 2008. Help us out and rate our content!

Showing 7 comments
  • vlad

    dang; thanks a lot.

  • Marc

    Hi Steve,

    what kind of metrics do you suggest to use for a backtest of intraday profit targets? Since we are looking for a multiple x of our initial risk it all comes down to the Stop Loss which leads to a few different options:

    a) A static hard stop as used in Steve’s calculation above. But might not be sufficient enough since volatility changes over the time.

    b) A volatility stop like ATR. But I found them not that great if you are trading very small timeframes like M5 and they are lagging (e.g. when you are trading during London Open your ATR is based on the less volatile Asian Session)

    c) A discretionary SL (e.g. above or below a major high/low)

    Since we are trading discretionary solution c makes the most sense and maybe I am overcomplicating things here a bit. But for further aspects my trades vary so much that I sometimes have trades with 10 pips SL and then 35 pips SL. And there are other factors like time of the day, news driven movements, etc… So I totally get lost when I think about it and I guess you may need a rather big dataset of trades to break down all the different contexts?

    Is a mechanical backtest (haven’t done any yet myself) for profit targets the best idea for discretionary trading at all? It occurs to me that you quite often use measurement moves as a profit target which actually creates solution ‘d’.
    What you know today, would you still say run a mechanical backtest over your entries or go the hard way and learn your market and use more discretionary exits (i.e. an achievable profit target derived from the current market context like a gap fill, measurement move, etc.)

    Thanks and all the best.

    • Steve W

      Hey Mark, this post is all about the ratio game when it comes to RR, and that’s really the only point it’s trying to make. A few posts around here that talk to the effect of letting losers wane and profits snapped at in seconds – this is just here to reinforce the opposite. In system trading it all depends on how you’re exiting. Example, the last system I worked to build counted lower lows to get out, or higher highs in relation to X previous bars as one means of exiting. Of course the loser has to get out at less bars than the winner to sustain anything like the above, so it’s ultimately relative to the logic being used. And that of course varies. Also this post was based on trades I used to post, which I stopped doing a while back (see About page as to why), and they were all longer term, based on S/R. Vol stops, etc, no matter what it is, it’s all about relation to the logic. I always say start with the take profit, then work your way back. If there’s not enough yield to justify the loss, then don’t do it. Plain and simple. Hope this helps. Thanks.

      • Paolo

        This is pretty good stuff. Thanks for sharing, Steve. Yours is definitely one of the best contributions to my trading, and, hopefully, to my future life and projects.
        Keep it up.

        Greetings from Italy 🙂

  • roel

    Hi Steve,

    I can use ‘take profit’ alone without using ‘stop loss’ option in my trading?


  • Jesper

    I probably missed something, but Steve McGrechan has a “run” with a stop of 20 pips and a target of 250 which results in a 24% winrate. You then make a run with a 10 pips stop and a 250 pips target and get a winrate of 31%. Decreasing the stop would not increase the winrate…

    I’ve been trading for 8 years and have never seen or heard of anybody who trades with a 25-30% winrate. I may come in the wrong places (all twitter trades has a win rate of 60%+, of course), but even if these traders do exist, a winrate of 30% is not for everybody. Only winning every 3-4 trades and having losing streaks of 10 trades on a regular basis is not easy.

    It would have been great if had taken the psyochological aspect into account, which in my opinion is much more important than purely optimizing the numbers of a system. If you can’t trade that system, it is useless.

    • Steve W.

      This article is ancient but yes, I completely agree. I talked in length about this actually in a more recent entry to this blog because keeping such a low ratio is nearly impossible, discretionarily, for most people. I’m not sure where you see the other numbers but frankly the point was to display the basic math. I remember copying and pasting all these years ago.

Leave a Reply

Start typing and press Enter to search