Kevin,

I am very appreciative of you keeping all of us abreast of the progress for this system. As you are likely aware, from a readers perspective, this thread has slowed down a bit from its inception. Part of that is due to the nature of the life cycle of where this system currently resides (you spent quite a bit of time explaining the system testing and modeling early in the thread, and now we are in the system maintenance/management mode). And part of it is that the system is not very active.

I am wondering if you might consider interjecting some of the activities you perform from some of your other systems on a regular (or at least somewhat regular) basis or consider the topic of looking at a portfolio of systems. I appreciate your ability to not focus on the entry triggers, as I believe that allows this forum thread to focus on the system itself. However, if you are so inclined, I would like to touch upon the idea of considering furthering the topic by discussing some other aspects as well (maybe in a new thread?) like system portfolio management, or the like. Or, possibly creating a thread discussing your thoughts on topics like system exits, which I think many folks (including myself) could benefit from.

For example - how does one determine when and where to exit to get a profitable system. I know you can run hundreds of thousands of tests using software to do this on historical data, but there is a lot more to this aspect that meets the eye. We always hear the axiom - 'Cut your losers short and let your winners run', and 'don't change your stops once set'. So far in your system, your average losing trade is quite a bit larger than your average winning trade - which many would consider non-conformist and would allow many to comment - 'that will be a losing system because it breaks a 'rule' of trading'. And, even though I am suspecting you do not change your stops after entry, I am curious as to if you have analyzed (on this or other systems) if a system is more profitable/ less profitable when doing so.

As an example, I was incubating a system for a time where I was changing my stop on a regular basis (always in my favor - never against), and I had a few folks state that this was a sure way to lose money. But what I did not see was them providing analysis of the details as to why this was true (other than this price was the stop you analyzed to be a level you knew you were wrong when you entered, then why would this change after you got into the trade). For example on my system my trades always started with a -1R stop - after entering, if the trade went in my direction +.25R, I would move the stop from -1R to -.75R, a +.5R move had me moving my stop to -.5R, a +.75R move had me moving my stop to -.25R, and a +1R gain had me moving my stop to BE - .01. When I conducted my analysis (in real time, not historical), it showed that moving my stops in this manner were more profitable than not moving them, even though this flies in the face of what many/most people call for (and many people state here and other forums).

This got me wondering, am I wrong?, do these other folks run a thorough analysis of their trading to determine if they are more profitable? does this only apply to certain trading styles? etc. etc. It also got me thinking trading is like a religion (defined as: a specific fundamental set of beliefs and practices generally agreed upon by a number of persons or the body of persons adhering to a particular set of beliefs and practices), once someone gets their mind set in a particular direction, they are extremely reticent to change it and consider other ideas. I would be curious from a systems development perspective, what you and other people (who lend an analytical eye to systems) are thinking. Again, this topic may likely get this thread off track, or it may stimulate more conversation, I will let you be the judge.

Just a thought, and thank you again for your continued support in educating us on this topic.

kayaktri 

Kevin, I am very appreciative of you keeping all of us abreast of the progress for this system. As you are likely aware, from a readers perspective, this thread has slowed down a bit from its inception. Part of that is due to the nature of the life cycle of where this system currently resides (you spent quite a bit of time explaining the system testing and modeling early in the thread, and now we are in the system maintenance/management mode). And part of it is that the system is not very active. I am wondering if you might consider interjecting some of the activities you perform from some of your other systems on a regular (or at least somewhat regular) basis or consider the topic of looking at a portfolio of systems. I appreciate your ability to not focus on the entry triggers, as I believe that allows this forum thread to focus on the system itself. However, if you are so inclined, I would like to touch upon the idea of considering furthering the topic by discussing some other aspects as well (maybe in a new thread?) like system portfolio management, or the like. Or, possibly creating a thread discussing your thoughts on topics like system exits, which I think many folks (including myself) could benefit from. For example - how does one determine when and where to exit to get a profitable system. I know you can run hundreds of thousands of tests using software to do this on historical data, but there is a lot more to this aspect that meets the eye. We always hear the axiom - 'Cut your losers short and let your winners run', and 'don't change your stops once set'. So far in your system, your average losing trade is quite a bit larger than your average winning trade - which many would consider non-conformist and would allow many to comment - 'that will be a losing system because it breaks a 'rule' of trading'. And, even though I am suspecting you do not change your stops after entry, I am curious as to if you have analyzed (on this or other systems) if a system is more profitable/ less profitable when doing so. As an example, I was incubating a system for a time where I was changing my stop on a regular basis (always in my favor - never against), and I had a few folks state that this was a sure way to lose money. But what I did not see was them providing analysis of the details as to why this was true (other than this price was the stop you analyzed to be a level you knew you were wrong when you entered, then why would this change after you got into the trade). For example on my system my trades always started with a -1R stop - after entering, if the trade went in my direction +.25R, I would move the stop from -1R to -.75R, a +.5R move had me moving my stop to -.5R, a +.75R move had me moving my stop to -.25R, and a +1R gain had me moving my stop to BE - .01. When I conducted my analysis (in real time, not historical), it showed that moving my stops in this manner were more profitable than not moving them, even though this flies in the face of what many/most people call for (and many people state here and other forums). This got me wondering, am I wrong?, do these other folks run a thorough analysis of their trading to determine if they are more profitable? does this only apply to certain trading styles? etc. etc. It also got me thinking trading is like a religion (defined as: a specific fundamental set of beliefs and practices generally agreed upon by a number of persons or the body of persons adhering to a particular set of beliefs and practices), once someone gets their mind set in a particular direction, they are extremely reticent to change it and consider other ideas. I would be curious from a systems development perspective, what you and other people (who lend an analytical eye to systems) are thinking. Again, this topic may likely get this thread off track, or it may stimulate more conversation, I will let you be the judge. Just a thought, and thank you again for your continued support in educating us on this topic.

Thanks for the kind words. You've brought up a lot of interesting points and observations. I'll try to address them at a high level here:

THIS THREAD HAS BECOME BORING: Yes, that is what happens when going into live trading. Everything has been done, now the system just has to run. A lot of people I talk to get frustrated by the boredom, and end up in a never ending cycle of "improvements." My experience is that it is better to stick with boring.

MAKE A THREAD ON OTHER ACTIVITIES, OR MAKE A THREAD ON EXITS: I'll look into this. Problem is I only want to do it if enough people benefit, and most people really like "entry" type discussions. But we'll see.

MOVING STOPS, OR NOT? Don't let anyone tell you their way is the only way. The right way is the way you choose, based on hard data and facts. I've seen systems fall apart when employing moving stops. I also won a contest using a system that had a moving stop. Both ways can work.

DOING THOROUGH ANALYSIS In my experience, most people do not do enough analysis. But who am I to say? I have a trader friend in New Zealand, and he read some of my material for my book, and he said "I do about 25% of the analysis you do, Kevin." And guess what, he is much more successful than I am. Maybe I do too much.


Thanks for the comments. I hope I've addressed your big issues. If not, please let me know.

Kevin

kevinkdog 

Yes, volatility plays a big role, and that is probably true for any system. Somewhere in this thread I looked at volatility, and it was down since I went live. That could explain some of the underperformance.

Hi Kevin,

You're not alone - google "bloomberg fx volatility trend" and see the first link on Bloomberg news. (I can't post a link yet...)

I'm running a system on the EUR and its also suffered poor performance since the end of last year.

Cheers, Tim

kevinkdog 

Thanks for the kind words. You've brought up a lot of interesting points and observations. I'll try to address them at a high level here: THIS THREAD HAS BECOME BORING: Yes, that is what happens when going into live trading. Everything has been done, now the system just has to run. A lot of people I talk to get frustrated by the boredom, and end up in a never ending cycle of "improvements." My experience is that it is better to stick with boring. MAKE A THREAD ON OTHER ACTIVITIES, OR MAKE A THREAD ON EXITS: I'll look into this. Problem is I only want to do it if enough people benefit, and most people really like "entry" type discussions. But we'll see. MOVING STOPS, OR NOT? Don't let anyone tell you their way is the only way. The right way is the way you choose, based on hard data and facts. I've seen systems fall apart when employing moving stops. I also won a contest using a system that had a moving stop. Both ways can work. DOING THOROUGH ANALYSIS In my experience, most people do not do enough analysis. But who am I to say? I have a trader friend in New Zealand, and he read some of my material for my book, and he said "I do about 25% of the analysis you do, Kevin." And guess what, he is much more successful than I am. Maybe I do too much. Thanks for the comments. I hope I've addressed your big issues. If not, please let me know. Kevin

I am learning that exits are more important then entries...

kevinkdog 

Some great comments and questions so far... Put yourself in my shoes: continue trading this system, quit trading it, or "it depends?" I think it is a good exercise, because this situation is sort of a gray area...

I vote "It depends"

On how much of your capital you have risk. I'm assuming the risk capital you have is a small percentage of your portfolio.

tturner86 

I am learning that exits are more important then entries...

Agreed. So long as you've got a good entry to start with...

tturner86 

I am learning that exits are more important then entries...

For some people Exit => New Entry.

Thxo 

For some people Exit => New Entry.

Not sure I understand. Can you explain a bit?

Thanks

Slightly up this week, nothing exciting...

kevinkdog 

Some great comments and questions so far... Put yourself in my shoes: continue trading this system, quit trading it, or "it depends?" I think it is a good exercise, because this situation is sort of a gray area...

I suggest to carefully consider volume - look at the moments when the system makes
positive gains and watch especially when the minus had occurred.
Every system has times that has phases of continued gains and others with many losers
in a row. Optimizing these days of taking the trade or to wait for a better setup may
give significantly better results.

If so - the system NEEDS to be continued

GFIs1