Does anyone make any money out of auto trading?? - page 3

sent you a short pm

A graph would never show a strategy. Be honest with yourself and with others. I'm not attempting to trap you or anything. The original poster asked for Automated Experts not Semi-Automated Systems. RaptorUK sounded correct about his assessment of what you were trying to describe. I haven't seen anyone provide Mathematical proof of a system which could beat forex on this site or anywhere else. While you seem ecstatic about your math, I'm thinking it's just another Roulette system. You have a chance to prove me wrong but I don't think thats gonna happen.

You say you've tested your system but the part which [you think makes it profitable] you didn't test. How do you teach someone impulses, subjectiveness and blah..blah. If you don't feel comfortable showing a Graph, the Strategy-Report Summary [without the trades would work].

I have backtested, of corse, but i can do so only for 1 yr. Btw I do not see the point in posting equity curves. Anyone can fake 1 in 0 seconds.

Dont make anyone believe that a fully automated (fire and forget) strategy that just works exists. You will never make anything like that in basic mt4 without neural nets or outher machine learning thechnique and it will be always semi manual, because you will need to verify ad refine day by day. The difference is that I suggest to consider the implementation of strategies that make some assumptions (call fundamental) on the market. It could be an snb peg, a weakening yen, even the widening after hours spread..

I am honest to myself and you and I say that i bet with 1000 on my EA but I dont want anyone to use my ea/trading idea for 2 reasons: 1: the less ppl use it, the better 2) i respect ther ppls money better than mine, I don't want others to lose on my advice.

@mpeter: Anyone can fake 1 in 0 seconds. Sure, but that makes them a Scammer.

Dont make anyone believe that a fully automated (fire and forget) strategy that just works exists. It exist, I can assure of that. Its not a question of what, its a question of how much. Semi-Automated, Neural Nets, AI, Human Adjusted can-lose just like the Black-Box. If you have proof to the contrary then show me [I'm always willing to learn].

Well, time will tell :) I still believe that time to time strategies have to be re-optimalized to fit actual market coditions. The only 'proof' i have is that if such a strategy exists and you can find it, sooner or later others will find it, so too much capital will use that strategy -> it will degrade. If you keep tweaking it, it could work, maybe?

Of course that could work, but it could also fail. When you re-optimize the strategy, you'll just be changing the parameters of the strategy slightly. All this optimization is base on yesterdays data not what will happen tomorrow. So if the market changes tomorrow you'll keep repeating this cycle and in the process losing money. Sounds like a coin-toss to me ;) 

hey sir,,

i usually have 50/50 HEAD/TAIL after 100 toss,, but is it possible to have 50/50 win/loss trading the "coin toss"?

If so,,then what will be the effect on win/loss ratio if i try  to modify R/R?

Thanks!

That is exactly what the chart shows . . .  https://www.mql5.com/en/forum/143890

 

If the strategy you are using is no better than a Coin Toss it's WR  (Win Rate)  will be a product of it's R:R ratio. 

R:R is the Risk:Reward ratio,  so 50 risk: 50 reward is 1.0  and this results in a WR of 50%,  the chart took approx 1.3 million trades to compile using the Strategy Tester

 

Looks like ,all roads lead to "break even" thats why"forex is a zero sum game"

So,,How i can  generate some EDGE?(most of the systems out there are nothing better than "coin toss")

I once heard ""if u can double your account (having 1:500 leverage) within 30-40 coin-toss,then u are considered professional enough to go live""

Is this possible by-any-means?

Thanks again!

That is what you need to figure out.