Forex Factory (https://www.forexfactory.com/forum.php)
-   Trading Discussion (https://www.forexfactory.com/forumdisplay.php?f=11)
-   -   Renko Trading Discussion (https://www.forexfactory.com/showthread.php?t=561530)

Hercs Oct 30, 2019 9:59am | Post# 1081

1 Attachment(s)
Hi Matt,
The first and only trade not taken by the EA can be seen here.
The reason for that long was that I changed the HalfTrend to Amplitude 1 (from 2 previously and defaulted) where that long was triggered correctly.
I came to realise that these shorter TF's require a quicker response or else I have to go back to my M6/7 templates.
So, I closed that long for +2 points, but look at what happened next:
The HalfTrend is short and @ 16h30 the RTH drops through the 50-level without a trade being triggered, but it could have been
that I missed the visual myself for whatever reason at the time.
So, I will not hold it against the EA as the trades on the EURUSD, were all purrfect (as in cat)
No doubt that you are in a class of your own.
Thanks again.
Hercs.
Click to Enlarge

Name: Matt re US30_M2 missed short @ 16h30 today 30 Oct 2019.PNG
Size: 41 KB

Hercs Oct 30, 2019 10:24am | Post# 1082

1 Attachment(s)
Matt, maybe just a quick look tonight at these last 2 entries where the short registered 9,5 and the long 91,29 for the long if you have time, that is. Otherwise, the EA performs well.
Thanks in advance.
Hercs.
Click to Enlarge

Name: Matt re EURUSD_M2 entry 30th Oct @ 17h05 and 17h13.PNG
Size: 45 KB

MathTrader7 Oct 30, 2019 10:24am | Post# 1083

Hi Matt, The first and only trade not taken by the EA can be seen here. The reason for that long was that I changed the HalfTrend to Amplitude 1 (from 2 previously and defaulted) where that long was triggered correctly. I came to realise that these shorter TF's require a quicker response or else I have to go back to my M6/7 templates. So, I closed that long for +2 points, but look at what happened next: The HalfTrend is short and @ 16h30 the RTH drops through the 50-level without a trade being triggered, but it could have been that I missed the...
Hi Hercs!

Thanks for reporting the missing trade. I think I know what happened in the EA code that it didn't take that SELL trade. I will fix it and release a new version.

P.S. 10 points (1 pip) Renko?? Are you kidding me?

Best,
Matt

MathTrader7 Oct 30, 2019 10:28am | Post# 1084

Matt, maybe just a quick look tonight at these last 2 entries where the short registered 9,5 and the long 91,29 for the long if you have time, that is. Otherwise, the EA performs well. Thanks in advance. Hercs. {image}
Are you sure that the EA's settings of RTH and the settings of the RTH indicator that you have attached to the chart are the same?

Hercs Oct 31, 2019 3:37am | Post# 1085

{quote} Hi Hercs! Thanks for reporting the missing trade. I think I know what happened in the EA code that it didn't take that SELL trade. I will fix it and release a new version. P.S. 10 points (1 pip) Renko?? Are you kidding me? Best, Matt
Good morning Matt,
Thank you for the reply and comments.
Best wishes for today.
I have moved up to M3 TF with amplitude @ 2 for today.

Hercs.

PS, yes, the RTH settings correspond with EA and oscillator.

MathTrader7 Oct 31, 2019 4:28am | Post# 1086

{quote} Good morning Matt, Thank you for the reply and comments. Best wishes for today. I have moved up to M3 TF with amplitude @ 2 for today. Hercs. PS, yes, the RTH settings correspond with EA and oscillator.
Hi Hercs!

In this case, the RTH indicator might repaint?

Best,
Matt

@those who are asking where are the indicators for this EA to download. The indicators are available at this link.

Hercs Oct 31, 2019 8:23am | Post# 1087

1 Attachment(s)
{quote} Hi Hercs! In this case, the RTH indicator might repaint? Best, Matt @those who are asking where are the indicators for this EA to download. The indicators are available at this link.
Hi Matt,
In that case, would you mind using this ordinary RSI, but it is an "NRP-version" and the developer MrTools, does not build repainting indicators.
You will notice that I changed the levels to 85/15 as this one does not give you acute angles.
Thanking you most sincerely.
Hercs.
rsi nrp (alerts + arrows).mq4

MathTrader7 Oct 31, 2019 10:17am | Post# 1088

{quote} Hi Matt, In that case, would you mind using this ordinary RSI, but it is an "NRP-version" and the developer MrTools, does not build repainting indicators. You will notice that I changed the levels to 85/15 as this one does not give you acute angles. Thanking you most sincerely. Hercs. {file}
If it is the same as ordinary RSI, why we should use that? I can use the standard RSI indicator of MT4.

Hercs Oct 31, 2019 10:23am | Post# 1089

{quote} If it is the same as ordinary RSI, why we should use that? I can use the standard RSI indicator of MT4.
Yes I am aware of that, but the addition of "nrp" guarantees that there is no repainting. Nevertheless, feel free to use the standard RSI. Setting @ 5.
Best,
Hercs.

Hercs Nov 1, 2019 7:56am | Post# 1090

1 Attachment(s)
Hi Matt,

Attached for your perusal and comment is the NRP version of the RSI_TripleHull indicator. I would prefer to have this one to replace the existing one as I have all my experience with RTH. But I still leave the choice to you.
Thanks very much.
Best wishes,
Hercs.
rsi_triplehull 1.1 nrp.mq4

MathTrader7 Nov 1, 2019 8:13am | Post# 1091

Hi Matt, Attached for your perusal and comment is the NRP version of the RSI_TripleHull indicator. I would prefer to have this one to replace the existing one as I have all my experience with RTH. But I still leave the choice to you. Thanks very much. Best wishes, Hercs. {file}
Just rename "rsi_triplehull_1.1_nrp.mq4" to the previous RTH indicator's name. If buffers of this indicator are the same as previous indicator, The EA should work fine.
I am busy with other projects at the moment. I don't know when I would find some free time to modify the EA...

Hercs Nov 1, 2019 12:00pm | Post# 1092

{quote} Just rename "rsi_triplehull_1.1_nrp.mq4" to the previous RTH indicator's name. If buffers of this indicator are the same as the previous indicator, The EA should work fine. I am busy with other projects at the moment. I don't know when I would find some free time to modify the EA...
Matt, I don't want to put any additional pressure on you, but could you please let me know whether the change you wanted to make prior to you discovering that the RTH "might be repainting" is still on the cards? # Post-1083 above on this page.

Kindly let me know as soon as conveniently possible as you did not share what you thought that problem could be.
Much appreciated.

Hercs.

Just checked these 2 indicators for buffers; the one on the EA has 2 buffers and the new NRP one, 5.

Hercs Nov 7, 2019 10:25am | Post# 1093

{quote} Just rename "rsi_triplehull_1.1_nrp.mq4" to the previous RTH indicator's name. If buffers of this indicator are the same as the previous indicator, The EA should work fine. I am busy with other projects at the moment. I don't know when I would find some free time to modify the EA...
Hello Matt,
I can now confirm that the EA did not "take kindly" to the suggested changing of the name of the "rsi_triplehull_1.1_nrp.mq4" as you suggested above. It is probably related to the number of buffers as I mentioned above. Fxgenial very kindly offered to confirm my initial findings as he is way above my level of understanding of what differences that would make to the EA.
It, therefore, seems that the only solution is to change the coding in the EA and I come to you in humble submission requesting your expertise to make that change.
Continued success to.
Best wishes,
Hercs.

MathTrader7 Nov 7, 2019 1:19pm | Post# 1094

1 Attachment(s)
{quote} Hello Matt, I can now confirm that the EA did not "take kindly" to the suggested changing of the name of the "rsi_triplehull_1.1_nrp.mq4" as you suggested above. It is probably related to the number of buffers as I mentioned above. Fxgenial very kindly offered to confirm my initial findings as he is way above my level of understanding of what differences that would make to the EA. It, therefore, seems that the only solution is to change the coding in the EA and I come to you in humble submission requesting your expertise to make that change....
Version 1.07 of MT7_Renko_Hercs_EA released. In this version,

1. The indicator RSI_TripleHull_nrp is used instead of RSI_TripleHull_COL;
2. The two indicators RSI_TripleHull_nrp and HalfTrend_MA are embedded in the EA code;
3. The signal logic is modified to take it valid cases like in Post-1083.

Best,
Matt
MT7_Renko_Hercs_EA.ex4

Hercs Nov 8, 2019 12:28am | Post# 1095

{quote} Version 1.07 of MT7_Renko_Hercs_EA released. In this version, 1. The indicator RSI_TripleHull_nrp is used instead of RSI_TripleHull_COL; 2. The two indicators RSI_TripleHull_nrp and HalfTrend_MA are embedded in the EA code; 3. The signal logic is modified to take it valid cases like in Post-1083. Best, Matt {file}
Good morning Matt,
You are indeed a true spirited person. Thanking you from the hidden recesses of my heart.
May you be blessed for your kindness accorded to the community.
May you continue to enjoy the blessings of your work for many years to come.
Hercs.

MathTrader7 Nov 9, 2019 7:24am | Post# 1096

2 Attachment(s)
I performed optimization and backtests of MT7_Renko_Hercs_EA with 100 points Renko on EURUSD from 2016-01-15 until 2019-09-01. The results of optimization shows that HalfTrend's best input settings is 2! Any other number would result in overall loss. So I used 2 for HalfTrend and run the optimization on RSI_TripleHull (RTH). Below are the results,
Click to Enlarge

Name: opt_rth.png
Size: 25 KB
The best RTH period is 5! This is interesting as these numbers (2 and 5) are suggested by Hercs who uses this system on 10 points (1 pip) Renko. It is actually consistent with Fractal market theory (assumption). Below is the backtest result of using RTH period = 5,
Click to Enlarge

Name: report.png
Size: 46 KB
However, as one can see the DD is high and the profit factor, which is 1.12 is not good at all. With 1 pip Renko, spread and commission fees might even make the profit factor worse. Since the system is profitable in presence of trends in the market, adding a trade time filter may improve the results.

Hercs Nov 9, 2019 8:41am | Post# 1097

1 Attachment(s)
I performed optimization and backtests of MT7_Renko_Hercs_EA with 100 points Renko on EURUSD from 2016-01-15 until 2019-09-01. The results of optimization shows that HalfTrend's best input settings is 2! Any other number would result in overall loss. So I used 2 for HalfTrend and run the optimization on RSI_TripleHull (RTH). Below are the results, {image} The best RTH period is 5! This is interesting as these numbers (2 and 5) are suggested by Hercs who uses this system on 10 points (1 pip) Renko. It is actually consistent with Fractal market theory...
Hello Matt,
Thank you for the insightful tests that you have run.
I use HalfTrend on a setting of 2 and have discovered that a Renko Brick Size of 1000 (NB) set by the MathTrader7_RenkoChartCreatorEA (Non-gap) and a Renko TF of M6, is best for my manual trading.
The RTH period of 5 is also standard on my charts.
Attached is a screenshot showing you 4 trades, 3 of which were successful (in a very difficult market I might add). I could have closed the trades manually as I would normally do but decided not to interfere and to let the HalfTrend work its "magic" - hence the smaller profits.
14h55 entry had a potentially dangerous situation at 16h39, but the HalfTrend visually kept me in the trade for the run down. But notice that during that time the EA actually closed the trade @ 16h39:37 and re-opened another short @ 16h58:51 and because I did not interfere, closed the trade 17H01:58 on the reverse signal.
Which brings me to the point, did the HalfTrend perhaps "repaint"?
I would love to hear your view on this strange phenomenon plse.

I do appreciate your expertise even more and wish you continued success.
Best wishes,
Hercs.
Click to Enlarge

Name: Screenshot1.png
Size: 121 KB

MathTrader7 Nov 9, 2019 2:40pm | Post# 1098

{quote} Which brings me to the point, did the HalfTrend perhaps "repaint"? I would love to hear your view on this strange phenomenon plse. I do appreciate your expertise even more and wish you continued success. Best wishes, Hercs. {image}
Hi Hercs!

A quick look at the source code, I can confirm that HalfTrend repaints the buffers Up and Down in some cases.

Best,
Matt

jo24 Nov 10, 2019 5:16pm | Post# 1099

Hello MathTrader and everybody. I'm making some backtests with renko No-Gap, and I would need real tick No-Gap renko data...as you probably know renko data are reliable only when they have been running forward because they are "created" with real ticks...so, since I have only 1 month of real data, I was wondering if here someone has a No-Gap Renko running from some months and could help me sharing a .csv with that renko.

Follow this thread to extract csv: https://www.forexfactory.com/showthread.php?t=572375

Thank you.

Hercs Nov 11, 2019 1:22pm | Post# 1100

Hi Matt,
Would it be possible to add the option of taking the trade on the OPEN of the next candle rather than the close?
Could that possibly assist in this "repainting" issue with the HalfTrend?
The first time that I read about this, was last year when I guy made $1,0 million in ONE day - I kid you not - but the downside was the draw-down during that day and he used the OPEN candle. This evening FxGenial also suggested that and now my question to you is could it make a difference or not?
Looking forward to hearing from you at your convenience.
BTW, North America was closed today as you know, but in spite of that, the EA took 4 trades without any loss. The one trade (Long) was interesting where the retrace was 40 pts while the HalfTrend stayed LONG and I must admit that I may have missed the "repainting" during that time which could have resulted in a close and re-open similar to last week.
Best wishes and continued success to you.
Hercs.


© Forex Factory