Holy Graily Bob 'n Dotty 'n SuperSlope

EA's inspired by nanningbob's work here, especially those based on his 240 Moving Average trend detection filter.
Post Reply
dclayw
Trader
Posts: 139
Joined: Wed Nov 13, 2013 6:53 am
Location: Australia

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by dclayw »

It is still very early days for me testing this EA. My initial test results are somewhat invalid because the EA was not doing what I expected so I've had to make some slight changes. I'm pretty much understanding the main EA code now, the include file is a different story, there are some nifty pieces of code in there and hats off to all contributors.
traderduke » Thu Apr 20, 2017 9:54 pm wrote:
I tried 14-50 & 14-100 several times but didn’t trust them because if you try 16,17-50, the pairs flip flop. I’ll give them another go.What TFs do you think work the best with the 14-50
The setting is really somewhat arbitrary ofcourse, different settings working better in different TFs and in different conditions. But a line has to be drawn somewhere sometime and 14,50 is what I chose after eyeballing a lot of charts on various TFs. 14,50 definitely seems to work better then 7,50 on most TFs. I'm using H1 for initial testing and 14,50 seems to work pretty well there.

I intend using a threshhold and testing may even reveal that the stock 7,50 with an appropriate threshold is all that's needed. Main idea as I said is to get on those good moves and stay on them, and avoid the chop as much as possible, 14,50 helps with this. Initial testing is faster on H1 but I intend moving to higher TF if this all works out, probably D1. Have done enough short term trading so I'd like to wind down on a higher TF with a bot, if that makes sense (and what better bot to do it with!) 8-)
traderduke » Thu Apr 20, 2017 9:54 pm wrote:
I have used thresholds before but the dips always seem to cause false exits and entries especially in lower TFs. I’m not sure what you mean by “then trades will exit on yellow SS colour”?? Can you elaborate?
Yes that's what I was referring to with one of the slight mods I made. I'm always using CloseTradesOnColourChange so unless I employ some other kind of stop that CloseTradesOnColourChange is what will close the trade.

If you have a threshhold set then OOTB the EA will exit on a SS yellow (i.e. on candles where the currency lines are within the threshhold). That is quite a legitimate thing to do, it's just not what I want. E.g. I don't want a long trade to close until it hits SS red colour. The thinking is that most often SS yellow are just reactions on a good move, (not always but often).

So I make sure no trades close on SS yellow. In my version of the EA the slight mod is at lines 851 and 874 (I'm only using trading TF atm). E.g. line 851 is: if (SsColour[0] != green), change to:
if (SsColour[0] == red) PM me if you need more info.

I have made some other slight changes in that area of the code but not relevant to this particular discussion.
pakeha
Trader
Posts: 78
Joined: Tue Nov 06, 2012 3:20 am
Location: Canada

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by pakeha »

dclayw

You may want to look at threshold values of 1.0 on the higher TF (H4, D1) and 0.5 on the lower TF (M15, M30, H!).
trader689
Trader
Posts: 674
Joined: Thu Nov 17, 2011 12:53 am

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by trader689 »

looks great thanks for the share Traderduke
traderduke » Wed Apr 19, 2017 7:22 pm wrote:Hi dc
Thanks for the insight, I'll give it a try.

Here are my results for this week sooooofar!
I got in on the GBP PM Mays loose lips mess. The low TM244 SL-1.0 & TP-1.5 Help out a lot.

Using H1 Trading time sheet & H4
238 trades-Losses 85 wins-153

Best of Luck
Ray

PS; any Ideas for improvement!

here is the tradereport. The +$95 report if for last week Apr 12-14 using steves H4 set
The bottom is of course the Apr 16 to 19 H1 and still running, see you at the end of the week
If you discount the 3 GBP winners -$3500 you still have a $900 gain and growing.




Here is the set file for SS



Here is the set file for TM244
User avatar
traderduke
Trader
Posts: 306
Joined: Mon Nov 28, 2011 7:30 pm
Location: East Coast USA

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by traderduke »

dc
How did you turn off everything except the single trade on SS. Any chance you could load the set file.
thanks Ray
dclayw
Trader
Posts: 139
Joined: Wed Nov 13, 2013 6:53 am
Location: Australia

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by dclayw »

pakeha » Fri Apr 21, 2017 6:27 am wrote:dclayw

You may want to look at threshold values of 1.0 on the higher TF (H4, D1) and 0.5 on the lower TF (M15, M30, H!).
Thanks for the tip. I had a decent look at thresholds of 0.8 and 1.0 on D1. Not a huge amount of difference overall. Will stick with 0.8 on my initial testing and see how that goes.
traderduke » Tue Apr 25, 2017 12:25 am wrote:dc
How did you turn off everything except the single trade on SS. Any chance you could load the set file.
thanks Ray
PM sent
dclayw
Trader
Posts: 139
Joined: Wed Nov 13, 2013 6:53 am
Location: Australia

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by dclayw »

By definition, with D1 testing you only get one shot at a trade per day, so it's important not to miss it. I find that quite often even though a trade signal is given the trade is not taken because of a spread event at the start of the trading day. I was expecting the trade to be taken once the spread went back to normal, but that's not what happens.

I will do more testing but will this problem be fixed if I set EveryTickMode=true. It would seem that setting is much more important on the higher TFs. How are people using this setting on various TFs.

Seems to happen on lower TFs too but is less of an issue there.

Edit: EveryTickeMode=true fixed my problem by the way.
Last edited by dclayw on Thu May 18, 2017 12:42 pm, edited 1 time in total.
andwen
Trader
Posts: 62
Joined: Tue Jun 03, 2014 4:07 am
Location: Melbourne, Aust

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by andwen »

Hi Guys.

I am on board with this , not a lurker, just directed here recently by TJF...Thanks TJF. :hi:

Working my way through the thread as I like what the CSS can offer particularly on the daily time frame, and trading the 4hr. I prefer longer time trades as it suits me at the moment. Similair to dclayw.

Just want to take a trade on colour change of daily CSS thats all, and reverse trade if needed etc


I also have another concept at play, not related to CSS but may prove invaluable with CSS being the trigger. Wont detract from things here with that. PM if you wish.

I intend to setup my VPS and forward test for some time. So keep the correspondence coming.

Subscribed.
dclayw
Trader
Posts: 139
Joined: Wed Nov 13, 2013 6:53 am
Location: Australia

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by dclayw »

andwen » Thu May 18, 2017 9:34 pm wrote:Hi Guys.

I am on board with this , not a lurker, just directed here recently by TJF...Thanks TJF. :hi:
Welcome aboard. This thread has been pretty quiet but I've been quietly running tests in the background. D1 testing is soooo slow. If you read my other posts you'll see I'm taking a very methodical approach just to see what the SS EA does in it's purest state over a reasonable period, i.e. no filters, no hedging, no HTF control, no indicator contamination etc etc.

I've had a D1 test and a H1 test running for almost 1 month now on 20 pairs. Testing runs for the entire trading week starting at Monday open and all trades closed 1 hour before Friday close. Not holding any trades over the weekend but I have been re-entering trades on the D1 on Monday mornings if both the Friday SS colour and the SS Monday colour are in sync with the trade that was closed last Friday night.

Both these tests simply take a trade on the SS colour change and stop and reverse when the SS colour changes again. Trying to get as much out of runs as I can so the only "filter" I have is I ignore SS yellow colour on both entry and exit, (SS settings are 14,50,0.8 as previously posted). I'm not testing this with the view to trading it, I just want to get an idea of what stop/TP to use for each pair and ofcourse which pairs not to trade. To get stop/TP I will optimise MAE/MFE and then normalise the results so stop/TP can be expressed in terms of ATR. Once I have that I'll start looking at HTF control and HGI etc.

This "vanilla" D1 test actually performed reasonably OK on it's own for the first 2 1/2 weeks, then fell off a cliff this past week. Check the equity curve for D1 below, it's a great example of "up by the stairs and down by the elevator". The H1 equity curve is pretty close to a perfect sine wave :lol: . But these results are meaningless, I will only use them to extract stop/TP values.

Happy testing.
D1 14,50,0.8.PNG
You do not have the required permissions to view the files attached to this post.
dclayw
Trader
Posts: 139
Joined: Wed Nov 13, 2013 6:53 am
Location: Australia

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by dclayw »

I was going to use myfxbook for my MAE/MFE optimisation, but I'm afraid it is not up the the task at the moment. There are just too many inconsistencies in the data coming out of myfxbook (they are not trivial), and missing data too.

Is anyone aware of a a script that analyses trade history to produce MAE/MFE data for each trade. I just want the data in pips, not $.

Thanks
Dave
dclayw
Trader
Posts: 139
Joined: Wed Nov 13, 2013 6:53 am
Location: Australia

Holy Graily Bob 'n Dotty 'n SuperSlope

Post by dclayw »

dclayw » Wed May 24, 2017 2:12 pm wrote:I was going to use myfxbook for my MAE/MFE optimisation, but I'm afraid it is not up the the task at the moment. There are just too many inconsistencies in the data coming out of myfxbook (they are not trivial), and missing data too.
These threads seem to be pretty quiet but I'm still merrily testing away in the background.

In case anyone is interested I contacted FXBlue a few weeks back about MAE/MFE data (because that data coming out of Myfxbook was pretty inaccurate). FXBlue did have this data but it was only shown graphically, it wasn't included in their export file. I asked them to include it in the export and they had that up and running pretty quickly. The data is quite accurate as well.
Post Reply

Return to “Thingy Bob EA's”