EMA 25-50 Trading System (Detailed) | Page 94 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

10 Ema and 200 Ema cross system 6 replies

EMA crossover (ema 14 & 34) 4 replies

Please help code an EMA Crossover 12ema x 24 ema cross 1 reply

EMA's and MA's confiming small time correctoin ema's crossing 1 reply

EMA Close Crossing EMA High/Low Help Please! 2 replies

Attachments: EMA 25-50 Trading System (Detailed)
Exit Attachments

EMA 25-50 Trading System (Detailed)

  • #1,861
  • Jul 15, 2008 4:42am Jul 15, 2008 4:42am
GBP/USD

L 39
1 51
2 32
3 16
4 17

Tot +17650 pips

waaa now I want some reeeest
Remember that L voice contains both complete Loss and Stop and Reverse Loss, so all the results are a little LESS than the real ones

Anyway I think the way to finalize this system is to use the ATR. But we have to rethink completly the money management imho. Because the profit are really too narrow...

An Idea:

SL & TP1 triggers at ATR
TP 2,3,4 triggers at ATR*2

this can be sweeer imho, great way to:
1) cut early the losses
2) save early the gain (a operation with SL at BE is risk free)
3) let the profits run

Rooootz new work for you. We have to try this
  • #1,862
  • Jul 15, 2008 5:48am Jul 15, 2008 5:48am
Quoting ronmatzkin
Well, I am in the hi-tech field - we had a long period of 9-5 working hours but the last four weeks were hell so I dont have time to do anything but work.
Hopefully we will finish this crisis period and I will have time to invest more time in the indicator/EA programming (trust me it is more interesting then work).
Don't worry, we understand.
Myself, I 'm sort of in the hi-tech field too (semiconductor, 24/7 factory, technical area), and this period from june to september is always pretty hard (holidays -> less people at the plant -> more work for each one)
  • #1,863
  • Jul 15, 2008 5:55am Jul 15, 2008 5:55am
Quoting ronmatzkin
please just change the name of the indicator and the version so I can continue my developments in my version line... sorry for the bother but I need to keep track of my own versions.
That was imediately done:
Version's name added in the indicator's name;
Version's description added in the introduction of the indicator;
Some English and logic errors in the description field corrected (I took the liberty, hope you don't mind).
ATR(24) stayed but wasn't indicated in the description field (forgot about it).
  • #1,864
  • Jul 15, 2008 6:02am Jul 15, 2008 6:02am
Quoting ronmatzkin
Hi Tonirai, I will integrate some of your changes to my version 1.6 and send you the result.
I will merge the following:
1. 24 periods ATR insted of a 100 ATR.
2. Multiple TP in one candle just in a different way - the most profitable TP will be displayed. In your version you will see one TP signal over the other.
1. OK, I think 24 period adapts better to changes in the volatility
2. I had thought about that, but since I was just testing the changes I made, I was thinking to look at that today - It's done, great

Just to make sure (haven't loaded it yet), when you are live, looking at the last candle, if it reaches TP1 it shows it, but if the same candle reaches TP2, it will remove the TP1 icon and replace it by TP2 icon, right?
  • #1,865
  • Jul 15, 2008 6:22am Jul 15, 2008 6:22am
Quoting Corre
I started counting somewhere with my charts and the ATR sys. In my opinion there's something wrong with this idea (the indicator works perfectly)

The point is that a loss is far far far more difficoult to save than before. A loss here would mean about a -120 pips (let's assume 30 atr). To have a gain of 120 pips with this system we need to reach at least the 3rd profit...

The profit/pips calculator here would be great . Just by nose it's hard to tell if this way is going to be more profitable than the other.

Anyway now I'll start counting the results from Jan to today with the fixed SL_TP and the 1.6 Indicator
Wowww... everybody is working hard on the indicator today :nerd:
And I was... (not a working day for me, and went to bed at 4:00 am).

I don't understand how a loss is more dificult to recover from than before?
The proportion between SL and TP is always the same!
And I think it's recoverable at TP2:

Before you used 50 pips TP/SL, right?
So a loss would be 200 pips - and to recover from it, you'd need TP2 (50+100+50+50) = 250;

With 30 pips, a loss is 120;
TP2 is (30+60+30+30) = 150.
  • #1,866
  • Jul 15, 2008 6:29am Jul 15, 2008 6:29am
yes, it seems your right about that. I probably miscounted this morning, sorry.

Anyway I don't like all this cut of the profits.. a whole operation till the 4th quarter is going to let us far lesser than before.

And since we're doing many more operations, we're going to pay much more spreads.

mmm
  • #1,867
  • Jul 15, 2008 6:46am Jul 15, 2008 6:46am
Quoting Corre
yes, it seems your right about that. I probably miscounted this morning, sorry.
That's what we're here for, helping each other and getting this indi to be the best!

Quoting Corre
Anyway I don't like all this cut of the profits.. a whole operation till the 4th quarter is going to let us far lesser than before.

And since we're doing many more operations, we're going to pay much more spreads.

mmm
Hummm... you've got a point, but imagine this:
A whole operation till the 4th quarter is less frequent, the bigger is your SL/TP...
And with a smaller one, you'll dig deeper in the smaller trades.

But for more operations, more spreads, that's a fact.
Only comparing several SL/TP for the same pair can clarify it.

And if I understood it right, multiplying factor was introduced to the code, so you can always use 2*ATR (for EURUSD, it means 54 right now)...
Other pairs will probably have better performance with different factors (1, 1.5...)
  • #1,868
  • Jul 15, 2008 8:03am Jul 15, 2008 8:03am
ok, but the point is that if we use ATR as SL and TP1 and we save earlier our investments from the loss and then we let the profit run with a TP=ATR*2 (which as you said is similar to the 50fixed)...

we can take all the advatanges from both strategies, can't we?
  • #1,869
  • Jul 15, 2008 9:05am Jul 15, 2008 9:05am
I get what you mean, but I honestly can't predict if it's a better strategy

Clarify one thing:
You mean an initial SL and TP1 = ATR, and other TP = 2*ATR...
What about the trailing SL above TP1?
  • #1,870
  • Jul 15, 2008 9:43am Jul 15, 2008 9:43am
Quoting tonirai
I get what you mean, but I honestly can't predict if it's a better strategy

Clarify one thing:
You mean an initial SL and TP1 = ATR, and other TP = 2*ATR...
What about the trailing SL above TP1?
it's just an idea anyway:
tp1=sl=atr

if sl is reached we close, ok.

if tp1 is reached we trail to BE. If tp2 is reached (which now is Atr*2) SL trails to tp1.

when we reach TP3 (always atr*2) it goes on TP2, and so on.
  • #1,871
  • Jul 15, 2008 1:12pm Jul 15, 2008 1:12pm
Quoting ronmatzkin
Hi Tonirai, I will integrate some of your changes to my version 1.6 and send you the result.
I will merge the following:
1. 24 periods ATR insted of a 100 ATR.
2. Multiple TP in one candle just in a different way - the most profitable TP will be displayed. In your version you will see one TP signal over the other.
By the way, I'm now starting to take a look at the new version, and apparently nothing has changed, regarding to point 2.;
In my version, TP signals were not overlayed, they were spaced (stacked) - now they are the same.
What I thought you meant was eliminate lower TP, and leaving only the most profitable one visible.
But that's not important... it's a detail anyway
  • #1,872
  • Jul 15, 2008 1:39pm Jul 15, 2008 1:39pm
I did a change in the code (thought it was included):
I defined external variable ATRMultiplier as double, instead of int.
To allow you to use 1.5*ATR, for example.
No need to attach it, it's an easy change to do... if another version comes later, make sure to include this little change.
  • #1,873
  • Jul 16, 2008 4:00am Jul 16, 2008 4:00am
Quoting tonirai
By the way, I'm now starting to take a look at the new version, and apparently nothing has changed, regarding to point 2.;
In my version, TP signals were not overlayed, they were spaced (stacked) - now they are the same.
What I thought you meant was eliminate lower TP, and leaving only the most profitable one visible.
But that's not important... it's a detail anyway
Yeah, you are right,
I left it as it is since I came to the conclusion it is better...
Thanks for the output.
  • #1,874
  • Jul 16, 2008 4:01am Jul 16, 2008 4:01am
Quoting tonirai
I did a change in the code (thought it was included):
I defined external variable ATRMultiplier as double, instead of int.
To allow you to use 1.5*ATR, for example.
No need to attach it, it's an easy change to do... if another version comes later, make sure to include this little change.
I have incorparated the change five minutes ago
here is the final version - if you would like to fix stuff in it let me know so I will add it to the original version.

Attached File(s)
File Type: mq4 EMA_25_Indicator.mq4   26 KB | 424 downloads
  • #1,875
  • Jul 16, 2008 6:02am Jul 16, 2008 6:02am
Guys,

i just added profit calculator to the indicator...
But the results are waay lower then when u calculated manually.
May be I did something wrong? Please comment...
Attached File(s)
File Type: mq4 EMA_25_Indicator.mq4   30 KB | 352 downloads
  • #1,876
  • Jul 16, 2008 7:13am Jul 16, 2008 7:13am
Quoting svlad1983
Guys,

i just added profit calculator to the indicator...
But the results are waay lower then when u calculated manually.
May be I did something wrong? Please comment...
OK svlad, first of all, thanks for trying

I tried it on EURUSD now, with the start date of 17:00 GMT of yesterday (I put 19:00, because that's the hour my server has for 17:00 GMT).
The point was to make it count only one trade (the last one).

I used ATRforSL=off (to be sure of SL/TP value at the moment), and SL=TP=28.

That short trade reached TP1, and then SL with profit, meaning profit for the quarters was:
28+0+0+0=28;

But the profit calculator indicates:
win/lose ratio is 2/0;
profit for period is 9 pips.

So there's obviously something wrong with it... but keep trying, if you can
  • #1,877
  • Jul 16, 2008 8:43am Jul 16, 2008 8:43am
Quoting tonirai
OK svlad, first of all, thanks for trying

I tried it on EURUSD now, with the start date of 17:00 GMT of yesterday (I put 19:00, because that's the hour my server has for 17:00 GMT).
The point was to make it count only one trade (the last one).

I used ATRforSL=off (to be sure of SL/TP value at the moment), and SL=TP=28.

That short trade reached TP1, and then SL with profit, meaning profit for the quarters was:
28+0+0+0=28;

But the profit calculator indicates:
win/lose ratio is 2/0;
profit for period is 9 pips.

So there's obviously something wrong with it... but keep trying, if you can
thx for testing it
can't do testing at work:S
will try toimprove tonight
  • #1,878
  • Jul 16, 2008 9:36am Jul 16, 2008 9:36am
Guys, I noticed a detail that may affect the results and the backtest, and may be related to the problem in the svlad’s pip counter:

Moments ago, I was watching GBPUSD;
Manual Mode=false, ATRforSL=true, multiplier=1, and ATR=33 at the moment.

Suddenly, it gave me at the same time signal to enter short, and TP1!

I thought that, after the cross of the 25 EMA (blue line), it was waiting for RSI and MACD to line up – and as it did at 1,9989, maybe it is considering the open or the high of the candle to calculate TP1…

But ATR was 33, so 1,9989 + 33 pips = 2,0022, which doesn’t match with open, neither high of the candle…

Finally, a new candle came, and the previous one closed above signal, so it’s not there anymore… but if it did, looking at it later, doing backtest would lead to serious miscount of pip profit

See the screenshots:
First is right before signal;
Second is when the signal arrived.

I might be less active here for the next days… but I leave this observation for you to discuss.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: right before.gif Size: 31 KB
Click to Enlarge Name: after signals.gif Size: 32 KB
  • #1,879
  • Jul 16, 2008 12:13pm Jul 16, 2008 12:13pm
Correction:
It's a short trade, so what I meant above was:

But ATR was 33, so 1,9989 - 33 pips = 1,9956, which doesn’t match with close, neither low of the candle…

The low at the moment was 1,9969...

So I think once the entry signal is given (even at the middle of a candle, like in this case), it's calculating TP1 taking in account the open of the candle - we have to make it count from the entry signal.
Without that, we never get the right distance between entry and TP when we get live signals.

But even fixed for that, it will still be impossible to count correctly the pips when backtesting...

ideas ?
  • #1,880
  • Jul 16, 2008 1:30pm Jul 16, 2008 1:30pm
Hi Corre, Tonirai and Ronmatzkin,

Thank you for sharing your ideas and hard work with the people on this thread. This is a useful system and even without these changes, still works. I am sorry that I cannot help with the level of details that you guys are going through. But I am waiting patiently for the final EA. I am sure it is going to be a great system to use. Thank you all!
0 traders viewing now