The max drawdown in 10 years is around 1700
The size of the positions is around 1200
Seems to be an averaging down strategy, at first look of the equity curve? Or orders are not close until they reach a new equity curve high?
That was my guess too.
I think the equity curve would look very different with 10 months (and much more) of overnight charges taken off the profit especially for such large position sizes. As a long term buy and hold for a bit and then sell it might work as long as you don’t mind watching your equity go into a massive draw down and you don’t use CFD or spread betting to trade it. I doubt that it would actually beat just buy and hold forever unless money management is involved. I suspect that if it uses averaging down then it is just another averaging down strategy that just got lucky and didn’t buy the first position in a series of positions at the high just before a crash.
I have overshadowed everything so as not to openly share strategy.
If you unshadowed / showed everything we still couldn’t / wouldn’t know how the Strategy worked as we would not see the code.
Re the 99% winners … open trades must sit there for ever and a day until they get back into profit.
I have overshadowed everything so as not to openly share strategy.
If you unshadowed / showed everything we still couldn’t / wouldn’t know how the Strategy worked as we would not see the code.
Re the 99% winners … open trades must sit there for ever and a day until they get back into profit.
You’re right, but it wouldn’t have changed much anyway.
Now I am busy with exams but later I will work to try to solve the problem of open positions for too long
How does this look in a markjet where not everything have gone straight up?
i see that you hold some positions for a long time, im wondering what would ahppen if market dropped -70% during one of those long holding periods?
Can u share a screenshot of backtest in a tough market like IBEX or france or something that can show a -50+% drop in market?
im wondering what would ahppen if market dropped -70% during one of those long holding periods?
Which is when a fuse emergency stop or quit would help. If we find that we have an extraordinary event then we give back some of the profit and quit. An unexpected out of the ordinary 25% loss is far easier to swallow than a 70% loss.
Hmm … sorry to say Francesco but I find this form of withholding a wee bit precious. Obviously not everyone shares everything they’ve done, they just keep quiet about it.
But to announce your big discovery, post results, admit that you need help with it … and then say you are only willing to share with some select group? to me, that’s weird.
Frankly, it strikes me as juvenile and contrary to the spirit of free help that is offered here. But hey, that’s just me and it’s not my forum. Do whatever. I’m sure that others will have no problem with it at all.
Very best of luck.
Everyone is free to share or not their codes here. However if Francesco needs our help, I think that the answers we could give him would also help many other people who read the forum topics without ever commenting on them.
Since this is a “fork” of one of the strategies present on the site, it is useless to tell us about the inputs, what interests us are the outputs which generate such a high positive trades ratio. Many of us have raised the issue of not cutting losses, which often results in this type of impressive result.
If Francesco wants to tell us about exits, then he will quickly be informed about the robustness of the strategy and he will be able to move on to something else if necessary. Rest assured, everyone is going through the same stages of discovery and learning in the world of algorithmic trading! 🙂
Hmm … sorry to say Francesco but I find this form of withholding a wee bit precious. Obviously not everyone shares everything they’ve done, they just keep quiet about it.
But to announce your big discovery, post results, admit that you need help with it … and then say you are only willing to share with some select group? to me, that’s weird.
Frankly, it strikes me as juvenile and contrary to the spirit of free help that is offered here. But hey, that’s just me and it’s not my forum. Do whatever. I’m sure that others will have no problem with it at all.
Very best of luck.
My intention was not to brag about my discovery. The problem is that i’m a newbie and I still don’t have the skills to refine the code and remain silent about it.
Although I madly love the mindset of this forum based on the spirit of sharing, to which I have tried to collaborate since the first day despite my ignorance in coding.
And just because I have enormous respect for this mindset, as I mentioned in another topic, I find that when a seemingly incredible strategy is found, it makes me angry to know that any jackal who visits this forum can “steal” it and draw from it profits.
I have never restricted the circle to selected people, I have made some examples naming some of the people I have “known” since I attend the forum, but the invitation was open to all those who had a real interest.
For all the others, as I said, I am now stopped for the exams and as soon as I have time I will work on it and post the results.
Aware of the fact that, being a newbie, it could easily prove to be a not working strategy.
No worries Francesco, it’s perfectly possibly that I just woke up the wrong way this morning … that’s happened once or twice before.
Best of luck with the exams, and the code, and everything else.
Update: As i thought, the strategy was a failure. I analyzed it with the support of a member of this forum.
Itf and code below to make fun of me 😀
DEFPARAM CumulateOrders = false // Cumulating positions deactivated
n1= 0
n2= 61
// Conditions to enter long positions
buysignal = (close CROSSES over Average[30](close))
IF buysignal THEN
BUY 2 CONTRACT AT MARKET
ENDIF
// Conditions to exit long positions
sellsignal = rsi[n1](close) crosses under n2 //(close CROSSES over BollingerUp[20](close))
IF sellsignal THEN
SELL AT MARKET
ENDIF
//************************************************************************
// MFE trailing stop function
//************************************************************************
///trailing stop
trailingstop = 21.5
StartTrailingStopValue = 4.5
//resetting variables when no trades are on market
if not onmarket then
MAXPRICE = 0
priceexit = 0
endif
//case LONG order
if longonmarket then
MAXPRICE = MAX(MAXPRICE,close) //saving the MFE of the current trade
if close-tradeprice(1) > StartTrailingStopValue*pointsize then
if MAXPRICE-tradeprice(1)>=trailingstop*pointsize then //if the MFE is higher than the trailingstop then
priceexit = MAXPRICE-trailingstop*pointsize //set the exit price at the MFE - trailing stop price level
endif
endif
endif
//exit on trailing stop price levels
if onmarket and priceexit>0 then
EXITSHORT AT priceexit STOP
SELL AT priceexit STOP
endif
Hey, we’ve all been there – and me more than most. Every time I write a code that makes more than a hundred quid I think it’s going to change my life. Then someone (usually Vonasi) will very kindly point out that it’s actually a bit sh*te.
Non importa. The one that really works is just around the corner … I promise.
I think it’s going to change my life. Then someone (usually Vonasi) will very kindly point out that it’s actually a bit sh*te.
I don’t mean to be the harbinger of doom but I do hate to see someone get over enthusiastic about a strategy before it has undergone at least a bit of basic checks, analysis and tests to see whether it is just curve fitted. The reality is everything we code is curve fitted as we only have the past data to work with – so really we are looking for the least curve fitted. The more checks, analysis and tests a strategy undergoes and the more checks, analysis and tests a strategy passes the more likely it is to be a goodish one.
Even after all these checks, analysis and tests it could turn out to be a disaster for a trading account if we suddenly go through something like the way the markets have behaved in the last few days. Things like quit fuses have to be part of a strategy just because the future will never be exactly like the past and we are witnessing exactly that right now. I was in fact playing with various indicator ideas the other day and thinking to myself these all show that we are somewhere very extreme and whenever we have been somewhere very extreme before there has been a crash of some sort. Seems the indicators were right but even they didn’t predict anything this fast and violent – now if only I could remember which indicator ideas they were! The point is that we have never seen a crash exactly like this one so how can we code a strategy to cope with it – the answer is we can’t but we can minimise our losses when the unpredictable happens and then lick our wounds and go away and count what we have left. No one said that getting rich by auto-trading was either easy nor guaranteed.
One thing that we can all learn is that there has never been a coronavirus in our data before so how can we code a strategy to cope with a coronavirus?
Dear Vonasi,
the enthusiastic feeling when you think you’ve found the right way to make money is the classic mistake of the beginner, so please don’t blame me, i think all of us have gone through this moment…
I am a young guy, probably more than you guys, and as i said in another topic i’m working hard to find something that can make me live as i want, being financially free and not being slave of everyone after my master’s degree.
This topic was the starting point to let me be more rational about this world.
In a world like the automatic trading is very simple to fall in the dreaming of the “life saving system”, but i don’t like to be as much pessimistic as you.
We are working here on something concrete, unlike many other businesses… despite the fact that we are working only on the past: we can’t change it, but we can try to make the future better.
I don’t think that here there’s never been a guy that made money with algotrading, this is a reality, and everyone here is working with his capabilities (very small like in my case) to pursuit the dream.
Also, something like the market crash due to the coronavirus could be easily predicted with a minimum of common sense, and probably this is the case in which the human intervention is necessary in the algotrading.
