Backtesting the Most Popular Strategies on FF | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Backtesting charts for newly developed strategies 1 reply

Distributed Processing or Backtesting of Forex Strategies Using MQL5 0 replies

Backtesting multicurrency strategies? 18 replies

Any interest discussing the Hedge strategies that were popular? 3 replies

Backtesting multiple timeframe strategies on mt4 3 replies

Attachments: Backtesting the Most Popular Strategies on FF
Exit Attachments

Backtesting the Most Popular Strategies on FF

  • #1
  • First Post: Edited 2:48pm Oct 10, 2020 1:39pm | Edited 2:48pm
Quote
Do you think there is a profitable strategy on forex forums?

This is the question that one of my friends asked me and took me to the old days when I would read every topic popping up on forex forums such as forex factory, I think it was 2007.

Like most people around here, I was looking for a profitable strategy... actually no, I was trying to find the holy grail, a system with no flaw, is that familiar to you?

Anyway, that's another story.

When my friend asked that question, the first thing that came to my mind was, I don't think so.

Have you tried all of the strategies there? , I asked myself, maybe there is one, or at least there is a strategy with a potential of profitability.

I remember, every now and then I would find a thread with high-quality content. The OP of the thread knew what he/she was talking about to a great extent and the strategy could have been profitable if I had used and tried to find its strong suits.

That's why I paused when he asked the question.

Then something came into my head!

Let's backtest some of the strategies on FF or maybe other forex forums to find out the answer.

I'm not actually talking about backtesting using EAs because first many strategies either can't be turned into an efficient EA or if it's possible, the creator hasn't made one for it and my programming knowledge is just at this level: buy when MA 6 crosses MA 12; actually, I can't remember how I did that.

Plus, I want human feelings and decisions to be involved

That's why I decided to backtest the strategies manually by using some simulators made for this purpose such as fxblue simulator which is a free MT4/MT5 add-on provided by fxblue for free.

For the first strategy, I picked "trading made simple" thread from FF because first it's been around since 2011 and has more than5000 posts and second, it's still active so people probably see some results that make them committed to the strategy. So basically I think it can be a good start.

I've backtested that in three separate years and I'm gonna publish the results, probably tomorrow because I need to find some time to write what I found about the strategy.

In the meantime, let me know your thoughts on this subject.

What do you think about trading strategies on forex forums? Do you think they are profitable?

Do you have any suggestions for the next backtesting? Is there any specific strategies that you want to be backtested?
  • #2
  • Oct 11, 2020 3:55am Oct 11, 2020 3:55am
Quoting daivid40
{quote} This is the question that one of my friends asked me and took me to the old days when I would read every topic popping up on forex forums such as forex factory, I think it was 2007. Like most people around here, I was looking for a profitable strategy... actually no, I was trying to find the holy grail, a system with no flaw, is that familiar to you? Anyway, that's another story. When my friend asked that question, the first thing that came to my mind was, I don't think so. Have you tried all of the strategies there? , I asked myself, maybe...
Hi David40

The best method in deciding whether a strategy is profitable on Forex factory is to look through the list of posters that were around at the start of the thread, see if they are still there years later. Looking at the Trading made simple thread, only 1 long term trader remains there - that should answer your question about that strategy.
About two years ago I build a EA to see if the strategy was any good, here is what I found.
1.. It has a profit factor of about 1.18 if you trade the M15 or Day charts. That is not good enough to bother trading.
2.. The TDI offers no value to the strategy, removing it from the strategy improves results.
3.. Without the stochastic in this strategy, it has no chance.

I will be interested to hear about your back tests.

Cheers
Trading thin liquidity at the boundary of the charts
  • #3
  • Oct 11, 2020 9:28am Oct 11, 2020 9:28am
Quoting daivid40
{quote} For the first strategy, I picked "trading made simple"
Do you have any suggestions for the next backtesting? Is there any specific strategies that you want to be backtested?
This strategy is not profitable. And many others too. As you can see there are always a word "SIMPLE" in the heading and "NO EA HERE, EA PROHIBITED" warnings in every "successfull" thread. And then 5000 pages of pure luck and showing only good entries. Bad entries are not shown, because it was Monday, trader ocasionally was out for coffee and same bullshit. All that "this was not so good TDI angle"... Just too many rules and you always can say why you should NOT enter the trade.

Strategy must have 3 very important thing to work together: entry, exit and stop. All 3 must be mathematically calculated. As you see only entries are calculated but exits with stops are not using maths, they are using magic for that. Like "feelings", or 1:3 RR.

So since there are no rules it can't be tested, just a waste of time.
Observer effect
  • #4
  • Oct 11, 2020 12:53pm Oct 11, 2020 12:53pm
Quote
The best method in deciding whether a strategy is profitable on Forex factory is to look through the list of posters that were around at the start of the thread, see if they are still there years later. Looking at the Trading made simple thread, only 1 long term trader remains there - that should answer your question about that strategy.

Good point

Quote
1.. It has a profit factor of about 1.18 if you trade the M15 or Day charts. That is not good enough to bother trading.

Mine is way worse

Quote
2.. The TDI offers no value to the strategy, removing it from the strategy improves results.

But as I understood, the whole strategy is based on that
  • #5
  • Oct 11, 2020 12:59pm Oct 11, 2020 12:59pm
My take:

1. Manual backtesting is really stupid in this day and age. It's like doing business accounting with pen and paper instead of using automated excel spreadsheets.

2. If a strategy can't be automated because the rules are unclear or not defined at all, then the strategy most likely is garbage and it's not a strategy at all. Better flip a coin. It's much simpler and you will have the same expectancy.

3. The so called "Discretionary Trading" aka "Manual Trading" cannot be tested. I know some people claim they make money consistently this way, but I don't believe it. I think they are liars or they simply talk bullshit because they got some temporary lucky streak and now think they have mastered the market. No. The market will take back your profits soon.

4. The only way to win in this game is to have real systematic automated strategy with proven statistical advantage. Maybe for some this opinion sounds extreme, but I think I know what I'm talking about. But hey, don't' listen to me and my free advice..... Listen to the market. The market will teach you these lessons and you will pay for them.
  • #6
  • Oct 11, 2020 12:59pm Oct 11, 2020 12:59pm
Quoting ryuryu
{quote} This strategy is not profitable. And many others too. As you can see there are always a word "SIMPLE" in the heading and "NO EA HERE, EA PROHIBITED" warnings in every "successfull" thread. And then 5000 pages of pure luck and showing only good entries. Bad entries are not shown, because it was Monday, trader ocasionally was out for coffee and same bullshit. All that "this was not so good TDI angle"... Just too many rules and you always can say why you should NOT enter the trade. Strategy must have 3 very important thing to work together:...
I agree with most parts of your posts, however, since this strategy is a trend follower, the only way that it might be improved is by finding a good RR
  • #7
  • Oct 11, 2020 1:56pm Oct 11, 2020 1:56pm
Try to create backtest for Laura thread. This is really good one. Yes it is manual, but she has exact rules. And her strategy very similar to mine, I spent years to come to it. I think it can be backtested. Don't waste time for "no ea" threads and for threads with "simple" in heading. They are fake.
Observer effect
  • #8
  • Oct 11, 2020 3:30pm Oct 11, 2020 3:30pm
Quoting ryuryu
Try to create backtest for Laura thread. This is really good one. Yes it is manual, but she has exact rules. And her strategy very similar to mine, I spent years to come to it. I think it can be backtested. Don't waste time for "no ea" threads and for threads with "simple" in heading. They are fake.
Thanks for the suggestion, I will.

If anyone else has a thread in their minds, I'd like to hear that.
  • #9
  • Oct 11, 2020 4:19pm Oct 11, 2020 4:19pm
Quoting JamesSmithFX
My take: 1. Manual backtesting is really stupid in this day and age. It's like doing business accounting with pen and paper instead of using automated excel spreadsheets. 2. If a strategy can't be automated because the rules are unclear or not defined at all, then the strategy most likely is garbage and it's not a strategy at all. Better flip a coin. It's much simpler and you will have the same expectancy. 3. The so called "Discretionary Trading" aka "Manual Trading" cannot be tested. I know some people claim they make money consistently this way,...
No, I don't think you know what you are talking about.
You are correct about only trading a strategy with a statistical advantage but this can also be done with discretionary trading. You are often better not automating a strategy because any back tested strategy will always confirmed to you its best traded only during favourable market conditions.
Many professional Algo traders turn their EA's off and on as the market conditions demand, that is a little known fact outside Algo trading circles.
Good luck running a trending strategy EA during a long period of consolidation.
Lastly, I suggest you get yourself into a good live trading room and see what's possible with discretionary trading.
Trading thin liquidity at the boundary of the charts
  • #10
  • Oct 11, 2020 4:35pm Oct 11, 2020 4:35pm
Quoting RickM
Many professional Algo traders turn their EA's off and on as the market conditions demand
Many professional delivery companies start to ride horses and leave their big trucks on the parking, because of... ( traffic jam?). Yeah. In the beginning of 20 century... not in 21th.

And yes. Professional Algo Traders KNOW when to turn the algo off. Yeah. Like every "Simple" strategy know where is a trend and where is a flat.

And of course if algo lose manual win. Because of luck and magic happening all the time.
Observer effect
  • #11
  • Oct 11, 2020 4:53pm Oct 11, 2020 4:53pm
Quoting RickM
{quote} No, I don't think you know what you are talking about. You are correct about only trading a strategy with a statistical advantage but this can also be done with discretionary trading. You are often better not automating a strategy because any back tested strategy will always confirmed to you its best traded only during favourable market conditions. Many professional Algo traders turn their EA's off and on as the market conditions demand, that is a little known fact outside Algo trading circles. Good luck running a trending strategy EA during...
This "logic" does not make any sense! And I see too many flaws.

If the strategy is backtested and proven without on/off function in the tester then why use on/off during live trading?
Why do you turn the EA on and off?
How do you know that you have to turn the EA on or off?
If you are so sure that it is a period of consolidation then why not program the EA to detect this pattern?
How do you define consolidation?
If you are so sure the strategy will lose during consolidation why don't you invert the strategy instead of turning it off?
  • #12
  • Edited 10:04pm Oct 11, 2020 9:54pm | Edited 10:04pm
Quoting JamesSmithFX
{quote} This "logic" does not make any sense! And I see too many flaws. If the strategy is backtested and proven without on/off function in the tester then why use on/off during live trading? Why do you turn the EA on and off? How do you know that you have to turn the EA on or off? If you are so sure that it is a period of consolidation then why not program the EA to detect this pattern? How do you define consolidation? If you are so sure the strategy will lose during consolidation why don't you invert the strategy instead of turning it off?
Hi James

Building in a consolidation switch to turn on during a ranging period is unworkable, it’s even in the too hard basket for professional Algo traders.
You would have too many external functions requiring so much optimization, you would only achieve a curved fitted EA that works on demo data on only one time period.
That is not the way to do it anyway. You would build two seperate EA’s that works on each market type and run them together. So when in a trend, model A is making money and when in consolidation, model B is making money. The whole idea is to run a portfolio of EA’s that work together, not one EA that does everything.
Another reason is large coded EA’s often fail to open trades because the logic gets loss somewhere in the 1200 lines of code.
Lastly for traders who have never build a EA, it takes a few days to build a decent version, a week to debug a model, 3 months of demo testing looking for errors before ever getting to a live chart. If you were building a very complicated model, you can kiss one to two years goodbye before it ever got to a live chart. Most likely, it would get binned anyway.

Cheers Rick
Trading thin liquidity at the boundary of the charts
  • #13
  • Oct 12, 2020 12:24pm Oct 12, 2020 12:24pm
Why do they change me to a commercial member? I haven't promoted any products and I will never do this in forex forums....I just put my blog link into my profile around one year ago. If this is the case, why do they change it now? Anyway, I removed that link so the moderator may move back the thread to the forex discussion section, however, I like black ID.

By the way, what are the conditions of the commercial section?

Anyway, I'll put the results in a few hours
  • #14
  • Oct 12, 2020 1:24pm Oct 12, 2020 1:24pm
Quoting JamesSmithFX
{quote} This "logic" does not make any sense! And I see too many flaws. If the strategy is backtested and proven without on/off function in the tester then why use on/off during live trading? Why do you turn the EA on and off? How do you know that you have to turn the EA on or off? If you are so sure that it is a period of consolidation then why not program the EA to detect this pattern? How do you define consolidation? If you are so sure the strategy will lose during consolidation why don't you invert the strategy instead of turning it off?
lots of questions, so why not use then some more?

what happens to the demo backtest result when in actual live trading at fast moving market you will encounter slippage?
do you believe slippage at this day and time is totally fair, or is it inherently if not as obviously still biased to favor the broker?
is your backtest at any point in time or size or pair gets into performance actual execution?
what will happen with your broker if trade a minor cross at 24/7, or trying to execute 10-20-50 lot around the clock on the market? in test vs in reality?
how do you propose to discretionary code an EA about ongoing Brexit talk, or a (natural) disaster, an act of terrorism, etc? is there a reliable way to do so?
do you believe it is that simple to flip a strategy to make money? ever heard of news straddle? ever seen both legs taken out on the same trade?
can it really happen that oil price fall into negative? is that a chance an EA some point believe that is a possibility as unbelievable as it sounds? it was deemed a far fetched theory up til mid 2020, would you think anyone coded such scenario?

we agree, discretionary manual trading is a lot more work than have an imaginary EA making money month after month, with little no supervision, after all, supposedly everything was coded into.
for example i was buying oil, until the point when it just didnt feel safe, at around 7$ suddenly that 0$ wasnt that impossible, and the best was to not be involved...worst case lose outright, or risk that the broker cancel the profit if they lose...no upside. not sure an EA could come to that conclusion, but will be rooting that you come up with one that beats discretionary trading on the long run.

happy trading!
there is always, always another trade!!
  • #15
  • Oct 12, 2020 7:20pm Oct 12, 2020 7:20pm
I backtested the trading made simple strategy for 3 separate years: 2013, 2017, and 2020.

For all backtests, I used $10000 as an initial balance and since there aren’t any specific MM rules, I used 1 lot as a fixed lot size for all the positions.

For the first backtest I picked 2013.

I opened 171 positions on H4 and I only traded when the green TDI crosses the red one. I mean I didn’t trade the bouncing setups. And of course avoid consolidation which is detected in this strategy by ema, when it’s flat.

Well, to be honest, I expected a lot worse. The strategy lost only 1.5%. It kind of broke even. You can find the stats on fxblue from the following link. Just one thing, I accidentally did something that messed up the results so for seeing the correct results, set the filter to September twenty-third, 2013 as the end of the result.

https://www.fxblue.com/users/tmc2

For the second attempt, I picked 2017 but this time, I decided to take both types of positions, crossing and bouncing and I backtested on H1.
Well, that didn’t turn well. I lost around 60% of the account after 398 positions. Actually, I don’t think the bouncing setup is a good idea because most of the time when the green bounces off the red, the trend is about to end so not only can’t you get any pips but you also lose some.

You can find the result from here:
https://www.fxblue.com/users/tradingmadesimple

The last backtest is for 2020. This time I decided to go only by the crossing rule on H4, no bouncing.
That didn’t turn out well either, however, it’s better than the previous one. This time I lost around 33% after 204 trades.
It wasn’t so bad for the first 4 months and we even made some money but then it started to lose money and then it took a nosedive.

Here the result:
https://www.fxblue.com/users/tms2020


Conclusion

well, the results were somehow disappointing but not unexpected because as one of the friends mentioned, when you say trading is simple as crossing two MAs, nothing profitable is going to come out of it.

However, after hours of backtesting, I think 10 hours, and looking at the strategy in different situations, I can say that the biggest problem of this strategy is its stop loss or RR.

Since it’s a trend follower, it starts losing money in the ranging phase, like most trend followers. You might say, ok I won’t enter a position when I see flat ema, which is the consolidation detector in this system, but the problem is, you don’t know where and when the consolidations end so you probably enter late when the ema gives you the end of consolidation signals.

To top it all when you catch a late trend, the TDI most of the time shows you an exit signal after one or two candles because the price gets back to regain the energy, pull back, bye kiss … whatever.

Given that, you need to find another method for exiting. For example, a fixed 1:2 or 1:3 RR using ATR indicator…

…or instead

Use TDI but this time exit when it crosses, not hooking over or flatting. That way at least you stay in a long trend for a long time so you can catch more pips.

Anyway, I hope you find something useful from this backtest. if you want to see the whole process or see the above explanation visually, you can watch the following video.

Inserted Video
  • #16
  • Oct 13, 2020 9:47am Oct 13, 2020 9:47am
Quoting ryuryu
Try to create backtest for Laura thread. This is really good one. Yes it is manual, but she has exact rules. And her strategy very similar to mine, I spent years to come to it. I think it can be backtested. Don't waste time for "no ea" threads and for threads with "simple" in heading. They are fake.
Do you mean this thread
https://www.forexfactory.com/thread/...o-read-markets
  • #17
  • Oct 13, 2020 3:50pm Oct 13, 2020 3:50pm
Yes this one
Observer effect
  • #18
  • Oct 13, 2020 4:49pm Oct 13, 2020 4:49pm
Cool, great thread. This is a great way to test things out.

Another suggestion: https://www.forexfactory.com/thread/...o-trade-simple
Daily Open Strategy. Should be simple enough.
  • #19
  • Oct 13, 2020 5:05pm Oct 13, 2020 5:05pm
Now I can't even like the posts. I don't know what to say. Anyway...

Quoting ryuryu
Yes this one
Thanks

Quote
Cool, great thread. This is a great way to test things out.

Another suggestion: https://www.forexfactory.com/thread/...o-trade-simple
Daily Open Strategy. Should be simple enough.

Thank's I'll look into that
  • #20
  • Oct 13, 2020 9:42pm Oct 13, 2020 9:42pm
Quoting daivid40
Now I can't even like the posts. I don't know what to say. Anyway... {quote} Thanks {quote} Thank's I'll look into that
You need to talk to Twee, otherwise being a commercial member will prevent you from enjoying forex factory.
Trading thin liquidity at the boundary of the charts
0 traders viewing now