SPB detail testing

Post Reply
User avatar
Gertje
Trader
Posts: 1936
Joined: Mon Dec 12, 2011 1:17 pm
Location: Middle of the Netherlands

SPB detail testing

Post by Gertje »

Hi Thomas,

I can't stop looking at the charts at the moment, you triggered something in me... :mrgreen:

Below is the current GBPUSD chart, buys closed@TP and sells still going strong.
GBPUSD.png
I was wondering, during the year you had this pony running, how often did price action go to the 'trade-line' [lime circle] only to go back and close at the opposite signal later [red circle]?

D1 CSS takes a lot of time to change over, sometimes even so much that PA is already in the 'wrong' peaky half to generate a complete opposite signal and close the trades.

I'm thinking over to have the TP set at the opposite trade line instead of closing at an opposite signal.

On the other hand, I see lots of opportunities where PA makes higher highs/ lower lows before turning around, gaining double the amount of pips.

Food for thought.
At least for me it is... :smile:
You do not have the required permissions to view the files attached to this post.
billv
Trader
Posts: 283
Joined: Fri Apr 13, 2012 1:57 pm

SPB detail testing

Post by billv »

Gertje » Fri Feb 15, 2019 2:20 am wrote:
I'm thinking over to have the TP set at the opposite trade line instead of closing at an opposite signal.

On the other hand, I see lots of opportunities where PA makes higher highs/ lower lows before turning around, gaining double the amount of pips.

Food for thought.
At least for me it is... :smile:
Hello Gertje :hi:

I have been thinking about the exact same thing but for risk minimization purposes.

So what do you think of this scenario?

1. When there are no open market orders in the same currency pair I would set the TP of each new pending order at the previous pending order. This would stop situations where price goes hugely in our favor but it then reverses and we are loaded with many -ve trades which now could keep going in the wrong direction for longer than our account can cope.

2. When any pending order in the same currency pair becomes a market order, then the EA stops placing a TP for each new pending order. This step is needed for trade recovery purposes.

You would think that by letting some trades close in profit we will not take advantage of the long runs,
but as you know the market doesn't move often in a straight line so the EA will replace closed orders with new pending orders during retracements, and we be compensated by collecting profit multiple times.

I think there is no perfect scenario but we should not lose our focus which in my opinion should be to protect our capital at all times.

cheers
Bill
User avatar
tomele
Administrator
Posts: 1166
Joined: Tue May 17, 2016 3:40 pm
Location: Germany, Forest of Odes, Defending the Limes

SPB detail testing

Post by tomele »

billv » 14 Feb 2019, 21:02 wrote:
Hello Gertje :hi:

I have been thinking about the exact same thing but for risk minimization purposes.

So what do you think of this scenario?

1. When there are no open market orders in the same currency pair I would set the TP of each new pending order at the previous pending order. This would stop situations where price goes hugely in our favor but it then reverses and we are loaded with many -ve trades which now could keep going in the wrong direction for longer than our account can cope.

2. When any pending order in the same currency pair becomes a market order, then the EA stops placing a TP for each new pending order. This step is needed for trade recovery purposes.

You would think that by letting some trades close in profit we will not take advantage of the long runs,
but as you know the market doesn't move often in a straight line so the EA will replace closed orders with new pending orders during retracements, and we be compensated by collecting profit multiple times.

I think there is no perfect scenario but we should not lose our focus which in my opinion should be to protect our capital at all times.

cheers
Bill
Such wise words. Prove them !
Happy pippin, Thomas :-BD

It ain't what you don't know that gets you into trouble.
It's what you know for sure that just ain't so.
(Mark Twain)

Keep the coder going: Donate
billv
Trader
Posts: 283
Joined: Fri Apr 13, 2012 1:57 pm

SPB detail testing

Post by billv »

tomele » Fri Feb 15, 2019 7:06 am wrote:
Such wise words. Prove them !
Hello Thomas
I'm working on it. :hi:
Pips are important but IMHO our account survival is even more important.
As you know it takes time to prove that a strategy works and an EA can work well for a very long time until the day when it all goes belly up.

The SPB seems very safe when traded on very small trade size 0.01/$10000, and it probably is safe at that trade ratio but my new broker only offers me 200:1 leverage so I'm naturally concerned about market crashes and margin calls. Therefore, I have to find a way to limit the number of open pending orders (or to allow some of them to close) but at the same time, I don't want to make the grid smaller.

Another way to limit this risk would be to start deleting old pending orders (but only in un-triggered groups of Stop orders) and to start deletions after let's say the 8th pending order which will give the EA a 300-400pip work range. I will do this manually for now.

Take care
Bill
tgm0
Trader
Posts: 22
Joined: Fri Jul 28, 2017 4:18 pm

SPB detail testing

Post by tgm0 »

What I have noticed so far is that SPB needs to be able to open as many trades as it wants to, or else the overall profitability will decrease, and drawdown will increase.
I take as an example this 3 “extreme” demos of mine; DT-N02, N01, N02.

I started testing DT-N02 on Mar 16, and my idea was to see how long it took to blow it up.

SS D1(1.5/7/50), Peaky H4, 0.5ATR grid, [c]IndividualBasketTargetPips=270[/c], [c]BasketTargetPips=550[/c]
A starting balance of 1000 EUR on a 200:1 leverage and a very aggressive position sizing of 0.01/500 and no limits to how many positions could be opened.
To my surprise the account started rocketing and by May the 3rd had already doubled with little to no DD. Margin though was indeed going down to scary levels, most often than not even lower the 200%; as of today, it still hasn’t got margin called yet.
DT-N02.PNG
DT-N02_Empty4.PNG
By the beginning of June this beast was up 400% so, thinking this was pure luck (and in part I’m sure it was), I decided to open 2 identical demos, but this time with a healthier balance of 10.000EUR (not that it made any difference as I was still using position sizing of 0.01/500) and limiting the number of open trades via the ForexKiwi option, respectively set to 800 and 1500. I also didn’t want to blow the initial account anymore, so I set ForexKiwi to 800 here as well to try and avoid margin calls.
N02 Overview.PNG
You can immediately notice a sudden slope change; looking at the 2 arrows on the chart, you can clearly see what I mean.
The first arrow point at June, when I switched on ForexKiwi with a value of 800
From here on, the slope moves along more or less the same as the new accounts.
End of December came, and I disabled ForexKiwi again on the first demo; the slope started gaining momentum again.
SPB likes to breath…

Limiting the EA to trade less does not seems to have a good effect on both balance and DD neither.
N01.PNG
N01_Empty4.PNG
N01 (ForexKiwi 1500) up 157% but DD is out of control; it did not margin call yet, but its close.
N02.PNG
N02_Empty4.PNG
N02 (ForexKiwi 800) is now up 224% with a DD of almost 28%.

All this said, it is no discussion that 0.01/500 is taking the accounts to their limits. 0.01/10000 on the other hand, seems a bit too conservative.
Maybe we can consider 0.01/3000 or even 0.01/2000 as being safe trade sizes already, what do you guys think?

For whoever may be interested, both N01 and N02 are available on FXTestify; D01 under tester account “Tgm0 3” and N02 under “Tgm0 4”
You do not have the required permissions to view the files attached to this post.
User avatar
MurphyMan
Trader
Posts: 345
Joined: Tue Nov 15, 2011 2:35 pm
Location: Southern Illinois, USA

SPB detail testing

Post by MurphyMan »

All this said, it is no discussion that 0.01/500 is taking the accounts to their limits. 0.01/10000 on the other hand, seems a bit too conservative.
Maybe we can consider 0.01/3000 or even 0.01/2000 as being safe trade sizes already, what do you guys think?
For the past 11 months I have been using. 01/1,500 and it has been working for me.
billv
Trader
Posts: 283
Joined: Fri Apr 13, 2012 1:57 pm

SPB detail testing

Post by billv »

tgm0 » Sun Feb 17, 2019 3:15 am wrote: All this said, it is no discussion that 0.01/500 is taking the accounts to their limits. 0.01/10000 on the other hand, seems a bit too conservative.
Maybe we can consider 0.01/3000 or even 0.01/2000 as being safe trade sizes already, what do you guys think?
Hello Tgmo :hi:
Many thanks for showing us this,
food for thought :)
You would have noticed that the SPB version which clocked the 100,000 pips was trading 0.01/$10000 and from memory in July/August last year it got close to 30% DD.

In my opinion 0.01/$10000 is too conservative for the SPB. However, now I have made my trade size based on pip value (using riskpercent=2 and SL=1400). By doing this I have allowed some pairs to trade 0.01 and others to trade 0.02 and 0.03 in the same 10K AUD account.

However, as I found out myself, when large sums of money are at stake, our risk behavior changes :smile: Those of us who have been through very long trending periods and large market spikes would understand why we have to be conservative with our trade size.

Ofcourse there will not be Brexits every year but other sudden market events could happen at any time and they could trigger margin calls so we have to monitor and manage our combined open trade size.

Some people would argue (and rightly so) that if we don't allow the EA to take some risk then our equity will not grow and our account won't be able to handle the large market event or the DD when they happen (and they will, its only a matter of time). Another thing we can do is to let the EA take more risk but to be prepared to top up the account with more funds when it gets to dangerous levels.

At the end of the day, we have to decide by ourselves what trade size we are comfortable with. Brexit is still ongoing so personally I don't want to break my nest egg but one day the dark clouds would be gone.

:!!: :!!: :!!:
User avatar
tomele
Administrator
Posts: 1166
Joined: Tue May 17, 2016 3:40 pm
Location: Germany, Forest of Odes, Defending the Limes

SPB detail testing

Post by tomele »

Bill,

you may be right regarding all the trade size aspects. I don't know.

But I definitely know one thing you got totally wrong: Brexit was a huge gift for SPB. An economy with a death wish (mislead by populistic liars) taking a foreseeable route over a cliff. First, SPB was caught on the wrong foot when the Brits unexpectedly voted "Leave" in June. But then the markets reacted with long trends that SPB did perfectly ride. Take a look at the charts. Doubling an account within one year under these circumstances.

We need more of these events, don't we? :twisted:

Cheers
Thomas
Happy pippin, Thomas :-BD

It ain't what you don't know that gets you into trouble.
It's what you know for sure that just ain't so.
(Mark Twain)

Keep the coder going: Donate
billv
Trader
Posts: 283
Joined: Fri Apr 13, 2012 1:57 pm

SPB detail testing

Post by billv »

tomele » Sun Feb 17, 2019 9:25 am wrote:Bill,

you may be right regarding all the trade size aspects. I don't know.

But I definitely know one thing you got totally wrong: Brexit was a huge gift for SPB.
We need more of these events, don't we? :twisted:

Cheers
Thomas
Hi Thomas,
Are you saying that the SPB can handle spikes and large market moves?
It could, or it couldn't I have no opinion because the outcome would depend on what that particular event is. What we saw last year could have had a totally different outcome if the market reversed and kept going in the wrong direction with all those positions open.

Yes volatility will work in the SPB's favor but large numbers of open trades can become a problem.

I'm only writing my thoughts on trade size regarding risk minimisation.
I have enough scars from past market events to know that I only want to survive
and I can't rely on luck being on my side because it often isn't.
That's all, I'm not here to prove anything.

Have a good night :hi:
User avatar
tomele
Administrator
Posts: 1166
Joined: Tue May 17, 2016 3:40 pm
Location: Germany, Forest of Odes, Defending the Limes

SPB detail testing

Post by tomele »

Hi Bill.
billv wrote:Are you saying that the SPB can handle spikes and large market moves?
Yes, of course.
billv wrote:What we saw last year could have had a totally different outcome if the market reversed and kept going in the wrong direction with all those positions open.
Then SPB would have reversed or even stayed in the right direction if it happened fast enough. Do you think it is static? It isn't. It reacts to price action.
billv wrote:I have enough scars from past market events to know that I only want to survive
and I can't rely on luck being on my side because it often isn't.
My tests are not relying on luck. They are relying on empiricism.
billv wrote:I'm not here to prove anything.
I am.

Sleep well.
Thomas
Happy pippin, Thomas :-BD

It ain't what you don't know that gets you into trouble.
It's what you know for sure that just ain't so.
(Mark Twain)

Keep the coder going: Donate
Post Reply

Return to “Coordinated EA Testing Forum”