I can't code a simple system that works

[TriPack]

Quote from travis:

That's correct. I have already tried reversing systems, and they lose both ways.

What that typically tells you is that your particular idea has no edge and probably not much merit. The reason why reversing the signals for a losing system doesn't usually work is because of the bid ask spread and commission (trading costs). Also a true edge is not easy to find. Additionally, losing systems often trade too much and are highly inefficient, so the opposite will also trade too much and be highly inefficient, and also a likely losing system.

Typically the most benefit I get from reversing signals is to clarify which is more appropriate for the particular idea: breakout or countertrend. It can also help you spot the relative strengths or weaknesses of your system.

 

[mogul]

Quote from nitro:

I can tell you flat out that if you try to compete with or against a computer you will lose. The people doing that are in a league ten thousand times more sophisticated than you can imagine.

nitro

another myth that should be dispelled...

 

[nitro]

Quote from mogul:

another myth that should be dispelled...

Ditto.

The myth that should be dispelled is all the system crap that most people espouse on ET, particularly the TradeStation and WealthLab style crap.

nitro

 

[mogul]

Quote from nitro:

Ditto.

The myth that should be dispelled is all the system crap that most people espouse on ET, particularly the TradeStation and WealthLab style crap.

nitro

I take it you don't trade systems?

 

[nitro]

Quote from mogul:

I take it you don't trade systems?

Hahahaha,

You take it quite wrong. It is _because_ I trade systems and I am in contact with some of the most sophisitcated players in this field that I say this with such conviction. 99.9% of the people on ET don't stand a chance IF THEY USE TRADESTATION OR THE LIKE AND TAKE THEMSELVES 100% OUT THE TRADING DECISION.

nitro

 

[martys]

Quote from travis:

One extra problem I have is that as soon as I get on tradestation and into the code I lose the perspective on the market and get lost into the code. Just like someone who wanted to say something in a foreign language but doesn't speak it well enough.

I think I might be better off still trading discretionary until what I do is perfectly clear to me, so I can at least write it down in words. And only later, I will translate my words into easylanguage. I think my problem is that I don't even know 1) if my method works and 2) what that method exactly is, because it's changing quite rapidly.

All I know for now is the two timeframes I look at: 5 minutes and 30 minutes. The importance of what time it is, support/resistance and I also look at the book. I look at about 20-30 things at the same time, I don't know...the only thing that bothers me is that if I can't code it, as soon as I am not in the right mindset, I will lose money simply because I will weigh differently the ingredients. Basically for me the right mindset means that you can wait over 30 minutes before making an intraday trade. But sometimes I don't have that patience, and that's when I lose.

You have to do it piece meal. Don't try to do everything at once.... Code a little bit... Test it out as proof of concept... Debug... Save the code into a backup file.... Add some more element or layer. You can even test individual element as separate simple systems before putting the layers together. Keep the process in small increment and under control. Do not add more stuff until it is rock solid. Of course... add a lot of comments and that would help speak it out.

You will be surprised... The process will even make you a better DISCRETIONARY trader because now you can see how each of the elements plays out over time after doing all the historical testing. You can even decided not to trade the system at the end and it still will benefit you because now you have an intimate knowledge about your edge.

 

[nitro]

Quote from martys:

...You will be surprised... The process will even make you a better DISCRETIONARY trader because now you see how each of the elements plays out over time after doing all the testing. You can even decided not to trade the system at the end and it still will benefit you because now you have an intimate knowledge about your edge.

That is good advice, as long as you don't get stuck in the "IT HAS TO BE 100% AUTOMATED AND I WANT TO TAKE MYSELF OUT OF THE PICTURE" syndrome. A good system should [positively] feed back into a good discretionary trader, as far as 99% of the ETrs go.

nitro

 

[Midas]

A program system should be addressed this way:

1. Find what works

2. develop your program

3. When things change (AND THEY WILL SOONER THAN YOU THINK) take your profits and move on. Don't look back............the ability to move on after things change is what separates the winners from the losers.

This goes for any form of speculation.

" the only constant is change" Buda

 

[illiquid]

Quote from travis:

buy rises in an uptrend, sell declines in a downtrend, buy/sell breakouts in a range - decide the trend/range based on a 30 minutes graph

- reversals tend to happen at 9.35, 10, 3.30, 3.55.

Ask yourself this first of all -- is there an edge in even identifying above conditions in the first place? Do you really believe that assuming the market will turn at certain times of the day is all that is necessary to make money? Is that method of looking at the market even remotely logical? Why not 9:45 or 1:18 or 2:19? etc.

The markets may seem impenetrable at times, but it's not like trying to decipher some alien hieroglyphics that forces us to grasp at straws just to get an initial handle on it. Why not start with the basics? You're trading index futures, they are based on specific groups of equities that can be broken down further into specific industry groups, all of which react differently to incoming data, yet also move together as a whole when seen vs fixed income, commodities, currencies etc. Each day a different angle of market perception becomes the main impetus for price movement. Can market movement be systemized along those lines? Maybe, maybe not, but it would make a heck of alot more sense than building a system based on arbitrary turning times or random averages and levels.

edit: don't mind me, never tried to code anything in my life. But it would seem that someone who would want to develop a system would at least want to look at the basic elements that move the markets in the first place, and use them as their starting point -- or am I completely clueless and "fundamental" issues are by default off-limits to system/program traders?

 

[kiwi_trader]

Quote from nitro:

I can tell you flat out that if you try to compete with or against a computer you will lose. The people doing that are in a league ten thousand times more sophisticated than you can imagine.

Nitro,

Isn't the issue here the same as when you try to compete with anyone bigger/stronger/faster/more skilled than you on their own terms. You have to change the game. My approach to computers is the same as my approach to manipulators and other major player groups --- try to be aware of what will make them act and try to use that to your advantage.

I trade HSI Futures which I think is just about as manipulated as any market. Someone described it as "anti-technical" but by being aware of how it is working there is a good opportunity for a flea like me to extract a good living from this market.

FWIW my style is discretionary but it is 100% driven by rules that are written in clear english on one page of A4. I have plenty of losses, some no doubt resulting from the action of computers but that doesn't matter if the expectancy is positive.

On the original subject:
1. I like the idea of coding bits of the strategy to test if they improve expectancy.
2. I think that if you can't write it all down in 1 or 2 pages (1st with rules and 2nd explaining some things in more detail so that you will still understand them in 6 months) then you shouldn't try to code it.
3. Even though I can trade my page of paper consistently (win 8 days out of 10) I still can't code it because some of the trend change conditions are fuzzy. I could code the core and apply trend rules myself but I would then miss some of the extra points I get by noticing the character of todays market and setting my stoplimits and limits appropriately.