Why automation is a must

[babyjake1961]

I used to scorn fully automated Forex EAs when I thought I had gained enough trading experience but even more trading experience has brought me to believe they are just about the only way to stay in the game.

The reasons for that is strategy stagnation. Just about every trading strategy (or a combination of strategies) which is profitable in the long run has fairly lengthy stagnation periods of up to half a year. Surely, someone who is not a professional trader working for a big corporate name cannot afford to stay flat for that long.

Therefore, they either quit for good and go back to their day jobs or lose patience, start experimenting, and stop following the strategy which leads to an imminent loss.

With an EA, after making sure it's coded right, you can set it up on a VPS and run it from there while doing your usual daily stuff and not missing on any trades. Needless to say, you also don't feel any stress or emotions which oftentimes leads you to things like reluctance to cut your losses, premature profit taking, or revenge trading.

 

[babyjake1961]

Just to provide some examples, here are the equity curves of one algo I'm currently running. The first curve is for February 2007 - August 2007 and the second curve is for January 2007 - April 2016 (to fully include the first time span in it too).

While the second backtest shows a mind-boggling net profit of over 21 million on a 1K initial deposit, the first curve shows a depressive stagnation period of as long as full six months and an even more depressive 75% drawdown.

There's no way a typical financially struggling punter could endure such a drawdown: over those lengthy, depressive six months he/she must earn and consume their daily bread for which they weekly thank their Lord at a local chapel or else life will brutally consummate them in no time.

As such, they will not be able to sit tight and follow the strategy manually. Yet if they toss the bot onto a cheap VPS and only check it periodically for platform crashes, they can confidently wait out any amount of time to come from the rags to riches, provided their trading strategy is sound.

 

[Pat494]

Automation is getting increasingly sophisticated and I am sure it is only a matter of time before the EAs start to produce regular profits from whatever market condition. What will happen then no one knows.

:clap:

 

[babyjake1961]

Automation is getting increasingly sophisticated and I am sure it is only a matter of time before the EAs start to produce regular profits from whatever market condition. What will happen then no one knows.

:clap:

Only if an EA combines several strategies for different market conditions, trending markets and ranging markets being the main ones. I've completed my trending market strategy, now looking for a good ranging one. Are you aware of any?

 

[Pat494]

Only if an EA combines several strategies for different market conditions, trending markets and ranging markets being the main ones. I've completed my trending market strategy, now looking for a good ranging one. Are you aware of any?

Try an oscillator

 

[babyjake1961]

Try an oscillator

That's too little info to code into an algo.

Care to elaborate?

 

[Pat494]

RSI is the usual one

 

[babyjake1961]

RSI is the usual one

"Just code some RSI EA" won't do the job.

BTW, I don't use any indicators in my trending strategy. They are always lagging.

 

[kalott]

maybe some kind of martingale like milestone or blessing but it will probably blow your account sooner or later if you dont get the MM to work?

 

[babyjake1961]

maybe some kind of martingale like milestone or blessing but it will probably blow your account sooner or later if you dont get the MM to work?

Martingale will blow your account no matter what.

 

[kalott]

Martingale will blow your account no matter what.

"Only if an EA combines several strategies for different market conditions, trending markets and ranging markets being the main ones. I've completed my trending market strategy, now looking for a good ranging one. Are you aware of any?"

it depends on how the trending strategy is working and lot size

 

[babyjake1961]

"Only if an EA combines several strategies for different market conditions, trending markets and ranging markets being the main ones. I've completed my trending market strategy, now looking for a good ranging one. Are you aware of any?"

it depends on how the trending strategy is working and lot size

It's working really well And the lot size can be decreased to any value

The ranging strategy must be self-sufficient, profiting in the long run on its own just like my trending strategy does. Martingale is never one.

 

[NetTecture]

I used to scorn fully automated Forex EAs when I thought I had gained enough trading experience but even more trading experience has brought me to believe they are just about the only way to stay in the game.

The reasons for that is strategy stagnation. Just about every trading strategy (or a combination of strategies) which is profitable in the long run has fairly lengthy stagnation periods of up to half a year. Surely, someone who is not a professional trader working for a big corporate name cannot afford to stay flat for that long.

It is not only stagnation - you also significantly reduce the drawdown chance because while one strategy may be in a drawdown, others may offset. Every week, some strategies may be in minus - but there is a big chance you STILL take some money home because other strategies may make money. With a good selection of different types of strategies and various markets - "something should work all the time".

And it is trivial to trade 50 strategies each in 10 markets. Yes, you have to come up with them, but a decent computer and an infrastructure written for handling many strategies (sadly most are not - ridiculous if you ask me) have no problem running 500 strategies 24/7.

This week for example I lost big on crude - need really to work some spike protection into my base code so that trades that are not working at all are redirected to simulator (which my framework can do)). But the week still is good - thanks to other strategies winning enough to make it into the green.

For a human, handling a significant number totally separate strategies is simply not possible.

 

[IFeelFree]

Only if an EA combines several strategies for different market conditions, trending markets and ranging markets being the main ones. I've completed my trending market strategy, now looking for a good ranging one. Are you aware of any?

Try Ernie Chan's "Algorithmic Trading: Winning Strategies and Their Rationale". The first half of the book is mean-reversion trading algorithms. He gives you the code (in MatLab) for implementing the strategies, and shows examples of their performance, as well as their rationale.

 

[IFeelFree]

It is not only stagnation - you also significantly reduce the drawdown chance because while one strategy may be in a drawdown, others may offset. Every week, some strategies may be in minus - but there is a big chance you STILL take some money home because other strategies may make money. With a good selection of different types of strategies and various markets - "something should work all the time".

And it is trivial to trade 50 strategies each in 10 markets. Yes, you have to come up with them, but a decent computer and an infrastructure written for handling many strategies (sadly most are not - ridiculous if you ask me) have no problem running 500 strategies 24/7.

This week for example I lost big on crude - need really to work some spike protection into my base code so that trades that are not working at all are redirected to simulator (which my framework can do)). But the week still is good - thanks to other strategies winning enough to make it into the green.

For a human, handling a significant number totally separate strategies is simply not possible.

Instead of increasing the number of strategies and markets, which is a brute-force approach, one may also investigate more sophisticated strategies that have the potential for higher return/drawdown ratios. This would include strategies based on trading a cointegrated portfolio of assets (stocks, ETFs, etc.), as opposed to trading a single asset within each trading strategy. Cointegrated portfolios of 3 or more assets have better statistical properties for forecasting price changes, and so can be implemented in an automated system using a trading algorithm. Chan describes some of these algorithms in detail in the book I mentioned earlier. You may still want to implement multiple strategies, but if you're drawdowns are lower, you don't need to run hundreds of them just to stay in the green.

 

[TickCOM]

This is a classic problem - strategies stop working cause the market behavior change. Since market behavior change at an every year shorter interval, imply strategies need to be adjusted with shorter intervals, or even new strategies need to be developed.

Some compensation can be done by using portfolios of strategies. As mentioned, automation becomes increasingly sophisticated, and its about to make your research on the net for new development platforms.

 

[NVP]

I used to scorn fully automated Forex EAs when I thought I had gained enough trading experience but even more trading experience has brought me to believe they are just about the only way to stay in the game.

The reasons for that is strategy stagnation. Just about every trading strategy (or a combination of strategies) which is profitable in the long run has fairly lengthy stagnation periods of up to half a year. Surely, someone who is not a professional trader working for a big corporate name cannot afford to stay flat for that long.

Therefore, they either quit for good and go back to their day jobs or lose patience, start experimenting, and stop following the strategy which leads to an imminent loss.

With an EA, after making sure it's coded right, you can set it up on a VPS and run it from there while doing your usual daily stuff and not missing on any trades. Needless to say, you also don't feel any stress or emotions which oftentimes leads you to things like reluctance to cut your losses, premature profit taking, or revenge trading.

office politics my friend ..........happens in any large organisation ..............people change things for the sake of it to be noticed......doesn't mean it is for the best ........just change for changes sake mainly

N

 

[NVP]

I can only mainly talk forex experience in recent years but what most traders miss when benchmarking their performance is what was POSSIBLE in the market vs their performance ...

eg a trader shows me that they made on average 30-40 pips a day on the London morning session this week............. (avg say 4hours a day trading)

do I praise them or kick them in the butt ?

depends on the day !!.....and what the market offered vs what they took............

some days that was pretty ok as market was choppy.........other days when the dynamics were in full motion I would have expected 30-40 pips PER PAIR traded for that period of time scalping

so they get a reprimand

N

 

[NVP]

With a good selection of different types of strategies and various markets - "something should work all the time".

And it is trivial to trade 50 strategies each in 10 markets. Yes, you have to come up with them, but a decent computer and an infrastructure written for handling many strategies (sadly most are not - ridiculous if you ask me) have no problem running 500 strategies 24/7.


For a human, handling a significant number totally separate strategies is simply not possible.

100% agree here ............so I have a lot of algo support that set me up to then trade manually"..................sorry I havnt got the major money to invest in the algo systems that give 100% assurances on relaiability ....that's the big bank/hedgefund domains.........if anyone 100% trusts their (or someone elses) homegrown EA in this retail sector then you are being very naïve .............they can and will fail eventually ...........and excuses don't pay the rent when they fail

N

 

[NVP]

Martingale will blow your account no matter what.

remember BJ...........its not martingale that will fail you .......its the system the martingale is being used on