Advisors that work - page 10

 
Vladimir Baskakov #:
You're the missionary.

I would be a missionary if I claimed there was gold in the bucket and some rubbish. Unfortunately, it isn't. Very unfortunately. In the League, the vast majority of systems are unconditional rubbish. And there are only precious few, and those are not too permanent.

 
Georgiy Merts #:

"Stable" apparently means "for several years".

My answer is that there are NO such things. At least in forex. The difference between the two types of investments is not significant. In scalpers, this usually happens after three or four weeks.

There are legends on the forum about EAs that have been showing stable profit for decades, but no one has ever demonstrated a real account of such an EA... I am not even talking about the Expert Advisor itself

Stable means stable operation with no drawdowns for more than half a year, and constant balance growth.

For example, I've been working with my forex robot for 20-15 years and try to work without optimization.

 
Georgiy Merts #:

I would be a missionary if I claimed there was gold in the bucket and some rubbish. Unfortunately, it isn't. Very unfortunately. In the League, the vast majority of systems are unconditional rubbish. And there are only precious few, and those are not too permanent.

i don't know what kind of robots you have in your league, but i don't know what kind of robots they have.

 
Georgiy Merts #:

If you're talking about Walk-Forward-Optimisation, that's what I started out with, and I'm convinced that even the highest scores with this method are completely unrelated to system stability.

So you either don't understand anything, or you've done something wrong, or you're just riding on your ears now, probably all at once.

 
Georgiy Merts #:

I would be a missionary if I claimed there was gold in the bucket and some rubbish. Unfortunately, it isn't. Very unfortunately. In the League, the vast majority of systems are unconditional rubbish. And there are only precious few, and those are not too permanent.

Is that why you're allowed everything?
 
Georgiy Merts #:

"Stable" apparently means "for several years".

My answer is that there are NO such things. At least in forex. The difference between the two types of investments is not significant. In scalpers, this usually happens after three or four weeks.

There are legends on the forum about EAs that have been showing stable profit for decades, but no one has ever demonstrated a real account of such an EA... I am not even talking about the Expert Advisor itself.

This is a 10-year backtest without optimisation on default settings of MT5. If I look at the analytical results I would say that the backtest is not cool, but it proves that 10 years of stable operation is not a fairy tale, it is clear that there is a drawdown and everything goes on, but you can optimize it and it will give good results. The accounts are easy to do, I have 5-6 options, half of them will be gone in 5 years.

 
Evgeniy Ilin #:

10 years backtest, without optimisation, on default MT5 settings. Not that it is cool, but it proves that 10 years of stable operation is not a fairy tale, it is clear that there is a drawdown and everything, but you can optimize it and it will give normal results. I have easily created such accounts, I have added 5-6 variants and half of them will be left in 5 years.

interesting, but an EA showing profit in 5-3 years in the Tester should go to a well-deserved rest :-)

Our currency markets (their volatility and let's say leader-ship), due to objective reasons, have been changing drastically and it somehow cannot be predicted from the quotes.

the same proverbial corona or mega brouhaha about american-presidents, or the gas mess

the backtests in this case show nothing. nothing at all. and quite good strategies go to the trash. because the backtest.

TESTER, OPTIMIZER IS EVIL

or rather in applications it is Evil, it is misused and falsely interpreted. Trying to pull a pike out of the ice when chopping an ice hole is enough.

A backtest and optimizer is not a universal grail detector for fuck's sake...

 
Evgeniy Ilin #:

10 year backtest, no optimisation, on default MT5 settings

ugh, monkey, or arbitrage, even worse
 
Evgeniy Ilin #:

10 years backtest, without optimisation, on default MT5 settings. Not that it is cool, but it proves that 10 years of stable operation is not a fairy tale, it is clear that there is a drawdown and everything, but you can optimize it and it will give normal results. I have made such accounts easily, I have added 5 or 6 variants and half of them will still be there in 5 years.

Just to clarify: real drawdown of the balance, for example, during testing was not below 14% ?

did it outpace the growth of the trading tool itself ?

and of course it's not a netizen or a martin ?
 
Maxim Kuznetsov #:

It's funny, but an advisor who shows a profit for 5-3 years in the Tester should go to a well-deserved rest :-)

Our currency markets (their volatility and let's say leader-ship), due to objective reasons, have been changing drastically and this somehow cannot be foreseen from the quotes.

the same proverbial corona or mega brouhaha about american-presidents, or the gas mess

the backtests in this case show nothing. nothing at all. and quite good strategies go to the trash. because the backtest.

TESTER, OPTIMIZER IS EVIL

or rather in applications it is Evil, it is misused and falsely interpreted. Trying to pull a pike out of the ice when chopping an ice hole is enough.

The back test and optimizer is not a universal grail detector for fuck's sake...

well the back test is 15 years old, isn't that enough ?

Reason: