Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
I realize, after going through the thread, that I have omitted position sizing details that probably would have made things clearer to the in-depth reader. I did this to keep things simple (e.g. "next trade will be 2 contracts"), and unfortunately that has confused things and muddied the waters. I apologize for that.
I understand you want to stay under the DLL. But was all this part of your original research and testing? Does your strategy performance that you've analyzed and compare yourself to include this type of rule or decision making?
I guess I am just saying that in my experience, when I develop strategies --- whenever I've tried to apply rules like stop after a losing trade, or even stop after some daily loss limit that I define (an arbitrary number), the results of the backtest suffer. I would look to control risk from position sizing and not from these other methods, and just rely on a "emergency stop" for the strategy as a whole whenever it has a sizeable loss that is a fair bit outside of the historical figures.
I feel like some of your sizing decisions are not backtested to be improvements and that is where I would get into trouble myself. I would want to design the strategy so that sizing is part of the strategy itself, so that it could be properly tested including a scenario where you dig a hole out of the gate.
Have you run a monte carlo analysis on this new strategy? I imagine the answer is yes and I've just missed it
Yes, I incorporated the Daily Loss Limit into initial development and all subsequent studies. Kind of tricky, though. It would have been much simpler without DLL. But, rules are rules, and I must follow them.
You do bring up an interesting point: should trading strategies incorporate position sizing during development, or is it better to apply position sizing after the strategy is developed?
Personally, I develop the strategy, then later apply position sizing (where I also factor in the correlations of other systems I am trading). Many traders do it this way. But, I know traders much better than me that do it the opposite way (they include positions sizing right from the start). Both methods are valid, in my mind.
I have run simple Monte Carlos on what I am trading for this Combine, and those results can be seen in some of the daily equity curves.
Here is my thinking on this topic as I have been trying to figure out what to do for my combine. My thinking breaks sizing into two categories.
Initial sizing
Maximum Sizing.
For me, Initial sizing would be based on the Daily Loss limit. My plan is to allow for three consecutive losing trades to have the best chance to catch a move for the day. I know this severely limits my contract size and definitely restricts me from being able to trade the maximum allowed contracts in the combine. But I'm ok with this.
My mind set would not allow the first losing trade to "knock me out" for the remainder of the day. On the other side, if your system is good at catching an initial winning trade more often than not, then I can see how maximum size from the start would yield the largest profits. I see this calculation being based on: Average Stop size, Number of Trade Signals per day, and Winning trade percentage.
Once profits are booked for the day, I can see how the Combine rules enable you to increase the size by effectively risking those banked profits. I'm struggling with how to determine the increase in contract size (if at all). If I jump to Maximum sizing after one winning trade, I risk that if the second trade is a loser I could end up with the same net result as two standard size losing trades. If this happens, then one more losing trade would knock me out for the day. So my thinking is that one would step up the size as additional profits are banked.
I struggle with the step up in size contracts too because as soon as I catch a losing trade, I could have double or triple the loss which will kill daily profits. For me, I wont be able to answer these questions until I can get automated back testing figured out.
Currently, I'm leaning towards fixed size and trying to scale into a position over increasing size.
Thanks for sharing. I think adding to winning trades is a good approach, in general. Many times it makes a good system even better, and I use the idea in a few of my live strategies.
I think you are on the right path - figuring out how to best trade to the Combine rules, and realizing it might be different than other "live" trading. I'm guessing most who try a Combine never go to the level of depth you are. Of course, most fail Combines, so maybe it is cause and effect...
I did not read all the new posts before before my last post.
If there are few trades per day I understand why you would swing for the fence on the first trade and have the risk of a loss keeping you from trading that day. Especially as you might not have another signal. I know I either missed this detail or simply did not remember this fact about your system.
My only other thought is have you considered a system that would trade more times per day? Or revising this system for more trades?
They are real big on maximum loss. I would doubt you could get much changed related to daily loss. Or if they allowed it in the combine, you would have a harder time getting to the funding stage.