I'm going to be very busy for a while. Pardon me if I don't post till have have some time to catch up. - Wayne
No, I would never trade anything that had average losing trade 100 times larger than average winner. Furthermore, any trading method must be targeting some non-random relationship. As for me, I'm trying to look for strong, persistent dependencies that are tested objectively and have high confidence level. I cannot imagine any other way to create a model that has an edge. Otherwise you're just placing random trades and their expectation depends on current market conditions. And trying to solve this through money management/position sizing is futile.
I highly doubt that posting general concepts about system development money management, position sizing, weighting between systems, pyramiding, etc. will give away ones edge. The only thing that may possibly get watered down by sharing it would be some obscure entry signal. There are so many different variables that go into a strategy that it seems like someone would need to say 'Buy ABC @ this price & sell @ this price'. Even then most people would change something in the process that would yield different results. I have found collaboration on the system development process & idea generation to be invaluable. I have learned a lot from this thread and others on the development process. The biggest thing I learned is that it doesn't really matter what systems we use, as long as they throw off enough expectancy per time period in a controlled risk manner that the trader is happy with. Regards, Eric
This thread has been a fascinating experience. It has become clearer the reason the successful traders divulge minimal information. That understanding came both from the experience here and a lucid PM. Just to clarify: This strategy I'm using works. It's not the holy grail yet. A more successful and experienced trader proved he's using the same fundamental concept in that he never takes a losing trade for many years. HOWEVER, he uses a variety of ingenious techniques to avoid, reduce, and mitigate the outliers. Interestingly, those who trade this way apparently coined insider terms to describe different aspects of the mathematical and psychological nature of this model without any terminology to describe it. It's as though an entire new world opens up to manage risk. Gone are the thoughts of stop loss, trailing stop, % trailing, parabolic, etc. etc. for risk management in my research There's an equal or greater variety of options for managing risk with 100% win rate that are a true paradigm shift to grasp. So I'm also forced to learn the lingo to more easily grasp the advanced concepts. It's very humbling indeed. I have a lot of homework to do. Thanks for all the posts and encouragement. Later, Wayne
30 pips profit today against the real time demo server USD/JPY. 3 trades. Not too shabby. Have a couple technical issues before running it on live server--probably next week. But I'll do that asap with micro lots since I hate demo servers, they use AI to "similated" trades. They're close but still not enough like real order queues and live trades. Wayne
Now it is only question of the outliers or should I say question of controlling the exposure, right? ZB
Yes. I already know a manual way to avoid the vast majority of those due to a fascinating correlation I found to events external to the markets. Also, a long time trader using this same technique gave me a tip via PM on how to completely avoid all the outliers using automation. But that will require additional hardware/software and analysis to work out the specifics. I don't feel qualified to discuss it or share it until I can verify and test that concept myself. That will take months at least. He has a team of developers and equipment at his disposal that I do not--yet. Sincerely, Wayne
To summarize this thread for those who have showed up late, "I knowwwwww a secret youuuuuu don't. Na na na na na na!"
I know what you mean, Corey. It would be nice if Acrary and all the other advanced traders would just spell out in detail with examples and video instruction of all their techniques here on the forums. The reality is that's unreasonable for us to expect. I'm simply grateful for any amount of hints or tidbits that we can use for our research. Acrary's tips and hints have taken tons of work for me to figure out and put into practice, for example. But well worth it. The point is that people need to do their own research. For example, simply knowing there is a correlation of outliers that can prevent most of them is a huge secret revealed. Personally, noone ever told ME those events correlated to external events that are predictable. I found it out just by analyzing all those outliers individually and as set. It took many hours. But the pattern became obvious. So the answer is there for anyone who puts the effort to find it. Why would anyone expect experts to share the specifics without doing a little work themselves? In reality, just KNOWING they're avoidable is a HUGE tip and can make a person very rich if they put some analysis into it. To emphasize the value of that secret, most people assume the fat tails are a statistical anomaly and unavoidable. Once you avoid most or all of the fat tails, outliers or whatever you wish to call them, the randomness of the markets probably fits a normal Gaussian curve. ( I should filter them and create a curve to see for sure--but it follows logically since they're the only element creating the fat tails). In that case you can profit immensely due to the predictable nature of random time series. (I know that sounds contradictory--see other threads on that line of thought.) That's an enormous secret revealed. So to whoever is reading this. Go figure out the rest of the details, and prosper! I personally would love to hear about your progress. Sincerely, Wayne