Is it worth it, when to draw the line?

[numbertea]

I developed my system trading on opens of 5 minute bars. I noticed that if I trade on the minute bars instead I can almost make up the spread both ways which would add around 10% profit per trade on average. There are always other types of improvements I could try to make but I think this is a sure thing. Problem is that I would be incorporating a lot more code into my system to be able to accomplish this. It is a pretty major framework rework and I would have to modify every code section. I estimate it could take 500 hours to completely implement.

When should I just say...good enough. I am already putting all my free time non trading hours into system development but this can of worms will make me a hermit in my office.

Should I just tuck this finding away for someone else to code for me later or just pretend I didn't ever even notice it?

Cheers

 

[timsk]

Should I just tuck this finding away for someone else to code for me later or just pretend I didn't ever even notice it?

Hi numbertea,
To be able to answer your question, I think you have to answer two other questions first. Do that - and you'll have a good idea of what to do.
1. How confident are you in your abilities to gain the spread? It sounds like an extraordinary claim to me, such that if you really can do it, institutions will be very keen to put you on their payroll or buy your code! Many profitable traders operate within tiny margins and the spread is often the difference between profit and loss. Being able to effectively remove it would provide a massive boost to many traders bottom line - especially day traders or swing traders who trade in size.
2. Why are you thinking about doing this? By that I mean, are you hoping to market the code or is it strictly for your own trading purposes? If it's the latter, then run the numbers based on your last year's records to work out how much better off you'd be. Then project forward, add in any increased size you can reasonably expect to take and you'll get a pretty good idea of your hourly wage. There will be a fulcrum point below which you'll conclude it's not worth your while and above which it is. If your profits last year were £100 then, based on your estimate, they'd be £110.00 with the code written. Probably not worth your while. On the other hand, if your profits were £100k, then spending 500 hours for an extra £10k p/a looks like time well spent.
Hope that helps!
Tim.

 

[numbertea]

Hi numbertea,
To be able to answer your question, I think you have to answer two other questions first. Do that - and you'll have a good idea of what to do.
1. How confident are you in your abilities to gain the spread? It sounds like an extraordinary claim to me, such that if you really can do it, institutions will be very keen to put you on their payroll or buy your code! Many profitable traders operate within tiny margins and the spread is often the difference between profit and loss. Being able to effectively remove it would provide a massive boost to many traders bottom line - especially day traders or swing traders who trade in size.
2. Why are you thinking about doing this? By that I mean, are you hoping to market the code or is it strictly for your own trading purposes? If it's the latter, then run the numbers based on your last year's records to work out how much better off you'd be. Then project forward, add in any increased size you can reasonably expect to take and you'll get a pretty good idea of your hourly wage. There will be a fulcrum point below which you'll conclude it's not worth your while and above which it is. If your profits last year were £100 then, based on your estimate, they'd be £110.00 with the code written. Probably not worth your while. On the other hand, if your profits were £100k, then spending 500 hours for an extra £10k p/a looks like time well spent.
Hope that helps!
Tim.

Thanks for the advice. I guess it just comes down to the hourly wage then. It is the type of realization I was trying to avoid I guess. As it turns out my system is quite profitable without spread losses and though I am in the black I have been spending so much time trying to improve things. I guess my feeling a little burnt out printed my question.

I think I can achieve the same effect by incorporating my scalping program into my current system for better limit entry. That will be a lot faster to code. This forum is a great sounding board for me.

IMHO there is no boundary that cannot be pushed and no goal that cannot be achieved through imagination and dedication.

Cheers

 

[timsk]

IMHO there is no boundary that cannot be pushed and no goal that cannot be achieved through imagination and dedication.

There is a former forum member called the hare who used to say exactly this about T2W all the time!


Fighting talk numbertea - that's the spirit!
Tim.