ES Trend Following

[Bombardier]

Quote from Daal:

Ok but do you believe your rules are programmable or you think there are elements of intuition and discretion that can't be coded?

The way I trade involves a lot of visual interpretation which might not be easily coded. My understanding is that you need rigid and systematic rules to automate it. If you put my method of detecting support and resistance aside, I believe there is a high chance that you can code a trend following system that detects trend patterns based on the Dow Theory in the 4H ES time frame (higher highs, lower highs for uptrend; lower lows, lower highs for downtrend). The ES trends in 30 point steps, meaning that it ranges 30 points back and forth before it advances to the next 30 points range. You will have to detect the relevant peaks and troughs it creates on its way higher or lower. Price must break AND hold a new high/low to mark a valid peak/trough. Those will be the channel lows and highs until one of them is broken to reverse a position. I've illustrated it in the recent price action (attached).

If someone can code this, I deserve to see the result and rights to use it for myself

UPDATE: I'd like to add that I never enter somewhere in the middle of an ongoing trend. My entries have always been at reversal points at important junctures. Perhaps this should be considered as one of the primary rules for successful trend following.

 

[Bombardier]

Quote from gibdi0n:

everything is very bullish. Are U waiting for braking of 1300 resistance line AND close above the line?

I don't see 1300 as any relevant resistance to enter a long position for. But you are generally correct that the current trend continues to be bullish.

 

[shopster]

The way I trade involves a lot of visual interpretation which might not be easily coded........

................................

B,

you have answered your own question.

the EA is between your ears.

the highway is strewn with dead traders with the EA / back testing needle stuck in their arm.

great thread.
nice to see another pro at work.

cheers,

s

"trade what you see, not what you think".

 

[masterm1ne]

Quote from Bombardier:

My understanding is that you need rigid and systematic rules to automate it.... If someone can code this, I deserve to see the result and rights to use it for myself...

Maybe another geek can correct me, but I have done EasyLanguage, C++, PHP, ASM, HTML, etc... the core of programming is loops or control structures. They are all the same at the core.

All you need are control stuctures to detect the break of support and resistance (for entry), and a sequence of higher/lower highs and higher/lower lows (for stops).

The problem comes in where most systems fail. It's because they don't use an infinite array of integers to determine the patterns they are looking for... This is because in order for this to maintain subjectivity in decision making, you'd need to be able to account for changes in all the values between the high and low points and stops, which any student of the markets vary widely over time. Look at the variance in the timeframe of your last screenshot, there are lots of differences in the size/shape of the highs and lows support/resistance breaks.

How do you select from an infinite array which value you should use to determine the stops and support/resistance breaks when they change all the time? If you don't use an infinite array, you would end up with something like 1-100. So what happens when there are 101 values between the first support point and the support break (for example)? On top of that, when the S/R is broken, how will the program know what value in the array to select as the last point, or last few points of S/R (again, for example)?

If someone could code this, they wouldn't give it to you for free Matt lol... It's obviously worth a lot of money! This method isn't new. If you can recognize turning points and trends in a market, obviously you will never have to worry about money. Anyone whos read anything will know this!

 

[shopster]

M1,

it is support and resistance

add an indi. with a pop up and email alert.

check the chart and pull the trigger.

set the stop and go to lunch.

one hour chart.


s

 

[ammo]

Quote from shopster:

M1,

it is support and resistance

add an indi. with a pop up and email alert.

check the chart and pull the trigger.

set the stop and go to lunch.

one hour chart.


s

shop,can you show us what you are looking at

 

[vinc]

Quote from shopster:

M1,

it is support and resistance

add an indi. with a pop up and email alert.

check the chart and pull the trigger.

set the stop and go to lunch.

one hour chart.


s

that would be a great journal ! go, create one .. how about we call it 'Piece of Cake' ?

 

[Bombardier]

Quote from masterm1ne:

How do you select from an infinite array which value you should use to determine the stops and support/resistance breaks when they change all the time? If you don't use an infinite array, you would end up with something like 1-100. So what happens when there are 101 values between the first support point and the support break (for example)? On top of that, when the S/R is broken, how will the program know what value in the array to select as the last point, or last few points of S/R (again, for example)?

You point out a valid challenge in automation, masterm1ne. I've also programmed in PHP using OOP, but have yet to learn the dominant languages used for trading systems. Programming is about describing objects or scenarios. The intellectually thrilling part is putting your perception into very specific instructions. We would need to describe the right thing: I suggest to describe what a valid swing is, then the peak/trough which results from this swing. Hence you wouldn't need to cherry pick integer arrays (or an X number of candles) because this would be precisely known once the system knows what a swing is.

I envision a system that grasps the core of market movements, such like an autopilot that understands the law of physics in aviation and makes automated decisions upon it (from take-off to landing). Such standardization helped to decrease human errors in commercial aviation, and I do hope that we get some standard in the financial industry, too. Point me to a quasi-standard system if you know of one. Hence my utmost interest is first of all in understanding the ongoing dynamics by philosophizing about the battle between bulls and bears. If we could define its basic principles with lines of code, we would be one significant step further in the standardization process.

The decisions a trader and the automated system would do should match most of the time. Hence, a trader can only be an educated individual who fully understands the dynamics in price action, nothing less. I compare the retail trader with an amateur pilot, with a significant difference that traders generally receive no education or mentorship. This free-at–will and try-my-luck-in-stock-market mindset needs to be changed. As retail traders get more educated in financial markets, the gurus scamming innocent people would dissolve gradually.

 

[Bombardier]

Quote from shopster:

add an indi. with a pop up and email alert.

check the chart and pull the trigger.

This is still the most sensible thing to do, once the trader comprehends price action. He wouldn't need to sit in front of the screen around the clock, but set alerts or orders to trigger automatically.

 

[shopster]

Quote from Bombardier:

This is still the most sensible thing to do, once the trader comprehends price action. He wouldn't need to sit in front of the screen around the clock, but set alerts or orders to trigger automatically.

.........................................

B,

the market is simple and stupid at the same time.

always has been, always will be.....

s