My Path To Success

Status
Not open for further replies.
Toward Cyborg

I've finally managed to build my "clone" system. What this allows me to do is to import all of my trades into Tradestation and run hypothetical scenarios over my trading history, in essence it allows me to backtest myself. I can run optimization or hypothetical conditions that I can factor into my previous decision making process while, also, gathering useful stats on my trading.

If I had traded my discretionary approach (primary instrument only) then I could have traded using 1 contract only and would have returned around 12k and would have needed about a 10k account.

One of my primary research areas is my stop and target placement. I haven't interpreted all of the results. But, one trend is pretty evident and has been something I've started to learn myself, and that's basically that I do much better at hitting consistent winners then big hitters.

I can, also, take the entries and then run hypothetical exit scenarios on the trades. With one stop/target combination, I would have returned over 14k with a better then 82% win ratio. I will be running extensive analysis. After running the optimizer, I found combinations where I won up to 92% of the time. My optimal target is slightly less then I thought it would be.

But, what it shows me is that I am usually right. Yet, I'm not "right" for very long. It tells me something about the nature of how I'm rightm and how I can lose even when I'm right. It suggest that I may want to shorten my holding period.

The simulation invalidates some of my beliefs about stop placement. The biggest stop isn't always the best when trading off of time-sensitive information. There is a high cost of not placing good trades when one is stuck in a bad trade. Even though the biggest stop isn't always the best, the optimal stop was nearly 2x as large as I believed it needed to be. Also, what a lot of people believe about stops and targets is the fast way to the poor house -- at least for my trading style.

I'll be writing more up about this at my blog.
 
Here we can see my actual equity curve imported into Tradestation without my use of confidence based betting, excluding crude contracts, at the reduced 1 minute accuracy, and assuming I trade the ES as a proxy for all similar contracts over the past year based. Average profit per trade: around $30 after commissions of $8/round. Win rate around 65%

actual_equity_predictor.png


And here you can see the equity curve when I don't use my exits but instead chose an optimal exit/stop target based on the Tradestation optimizer. Average profit per trade: around $80+ after commissions of $8/round. Win rate around 80%+

optimized_equity.png


It obviously can't turn a bad/losing trade into a winning trade. But, what it shows me and makes clear is that I really do a better job of predicting direction. Indeed, my trading skills may not be very good.

This is just one of my more exciting projects. I have some other projects that I'll be starting today... I really feel that I'll hit my stride again soon. Can feel it.
 
Lucias,

When will you be going live? instead of paper trading? Just curious.
If you really want to sell your system, you should at least be posting live info trades. I just think that anyone can do well on paper trading. The real thing is a lot tougher.
 
Quote from Lucias:

Toward Cyborg

I've finally managed to build my "clone" system. What this allows me to do is to import all of my trades into Tradestation and run hypothetical scenarios over my trading history, in essence it allows me to backtest myself. I can run optimization or hypothetical conditions that I can factor into my previous decision making process while, also, gathering useful stats on my trading.

If I had traded my discretionary approach (primary instrument only) then I could have traded using 1 contract only and would have returned around 12k and would have needed about a 10k account.

One of my primary research areas is my stop and target placement. I haven't interpreted all of the results. But, one trend is pretty evident and has been something I've started to learn myself, and that's basically that I do much better at hitting consistent winners then big hitters.

I can, also, take the entries and then run hypothetical exit scenarios on the trades. With one stop/target combination, I would have returned over 14k with a better then 82% win ratio. I will be running extensive analysis. After running the optimizer, I found combinations where I won up to 92% of the time. My optimal target is slightly less then I thought it would be.

But, what it shows me is that I am usually right. Yet, I'm not "right" for very long. It tells me something about the nature of how I'm rightm and how I can lose even when I'm right. It suggest that I may want to shorten my holding period.

The simulation invalidates some of my beliefs about stop placement. The biggest stop isn't always the best when trading off of time-sensitive information. There is a high cost of not placing good trades when one is stuck in a bad trade. Even though the biggest stop isn't always the best, the optimal stop was nearly 2x as large as I believed it needed to be. Also, what a lot of people believe about stops and targets is the fast way to the poor house -- at least for my trading style.

I'll be writing more up about this at my blog.
More...

Blog promotes his trading systems. This guy is a cheesy vendor just like the ones he bases.
 
Why its important to place the trades....

Lately, it reminded me why it is important to place the trades. The reason is that there is always some performance lost due to the complexity of reality. I've lost a lot of performance the last few days even though I'm on break and only taking a few trades.

Some examples of the problems that can happen:

* My system expired temporarily meaning that I lost a great trade when I tried to place the order. I didn't have my credit card on hand and so I was unable to renew. The trade would have worked.

* The computer generated a trade that I had missed which would have been excellent.

* I lost a trade due to 1 tick missed on fill.

* In one case, I closed some contracts when market hit my limit and added size due to the closed contract. The subscriber never reported the fill resulting in possibly a larger loss for the subscriber then would be reported in my system, possibly.

* I lost another great short because I cut it too soon, with a small loss.

* I've had entry order mistakes where I've lost performance due to entering wrong contract sizes.

* C2 showed the titles of my 2 systems wrong resulting in my entering 3 contracts on a system that should have only traded 1 and only 1 contract on the system that should have had 3 contracts.

* I thought the markets were closed today and lost some other opportunities.

------------

The reason I share this is because these things happen. It makes me angry. It makes really angry when I lose performance due to factors. But, it is part of the trading process. I'm not good with dates but I'm starting to work on that. I mean that's part of the "reality" which isn't captured if I'm not actually placing the trades. One professional told me that he lost over $50,000 in order entry mistakes on average per year. Anyway, that's some of the reality. A solid plan makes allowances for mistakes, lost trades, etc.
 
Human Nature

You might think that the trader with the best risk/adjusted and strongest statistics offering exceptional/abnormal returns would be the most subscribed too trader. You'd, also, think that a trader/system developer with dozens of failed systems wouldn't be subscribed too. You might think that a trader who took huge swings and was down huge would have few followers.

The reality is that most traders/subscribers just look at the net return. They look for the biggest returning system. I've monitored multiple subscription/signal sites and reviewed them and found this to be the case on every one of them. The systems with the best risk/adjusted returns and strong performance were under subscribed too compared to systems with high but volatile net returns. In one case, I found a vendor who was hugely successful who had dozens of failed systems.

It goes back to human nature. The human nature is to want everything right now. Instant gratification. Why bother with trying to get rich in the markets over several years when one might get rich in a few months?

I can't say the system developers are doing anything wrong provided they aren't trying to deceive anyone about their record. I mean, I can't know that my systems won't suffer draw downs or even breakdown completely. Everyone will push the limits to an extent. Who's to say how much is too much?

I can talk about goals all I want. My goal is to produce around 100% per year with 15% or less drawdown or equivalent to making 1k when I feel like I trading. That's a goal. That's a want, a desire but is certainly no indication of what I will actually do.

What it takes to survive....

The more I think about it then the more I'm thinking about this idea that if I really want to make a living at this game then I really don't want to take a large draw down. I mean when I think what would be comfortable for me then I'm thinking a 7x return to risk taken, i.e 70% return per 10% drawdown. This is outside the range of the statistics of good systems/traders I've tracked. Objectively, the ratio is more like 1x return to risk taken, i.e 50% return and 50% drawdown. Some great traders can hit 4x. There is little evidence for anything much outside the 4x range. So, you see I can see okay I want to do 7x and realistically I might have to settle with something in the range of 1x to 4x. If we work backwards from the 15% and the 4x then we reach a goal of 60% return per year as a high water mark and 15% as a low mark.

I'm willing to revise or adjust my beliefs. Certainly, Neke has had large swings but then again he's not really trading for a living. If anyone here trades for a living then I'd be curious to know what your risk/reward is acceptable to you.. i.e net return/max dd.
 
I've ran analysis on many successful ES traders/systems and on average the successful traders, myself included, average around $30 per unit trade. My optimized clone made up to $80 per trade. From this, I get anywhere from $25 to $90 per trade for a great ES trader with the rate being lower for the more frequent traders.

From this it derives a few things: the trader who manages to capture 1 tick more can boost profits by 42% per trade. I try to use markets and limits smartly. I prefer to enter on market and exit on limit. I know some traders who prefer to enter on limit and exit on market. In some cases, I have to enter on market and exit on market. I try to be smart about it. But, I think it is an area that I can improve.

One can also do the calculations and find number of trades/frequency of trades based on that ratio for given profit goals.
 
Building From Strengths: Key Concept

One of my key concepts and one lesson I learned from playing poker simulations that I need to stick to my strengths. My strength is calling direction. My ability to call direction has been unprecedented. It is worth sharing that I do not use a single methodology but rather have various methodologies that are rather defined and distinct. I'm like an artist trying to put the pieces together or trying to make something that I like. Unlike my mentor Gary Smith, I did not struggle with knowing what the market might do. What his book did was to help me to accept what I already had.

I was also strongly influenced by Dr. Steenbarger. His work of purposeful practice and really the idea of using past performance. My major problem has always been psychological, just accepting what I already have because I must already have the components for success. I mean I already have them if my basis is sound, i.e that I'm doing what I'm best at. If I'm wrong on that then I'm wrong on everything. This doesn't mean that I can't improve. One problem I had with the idea of purposeful practice was that I won most of the time. How do you practice something that you usually win at and how do I know if I'm getting better or worse? Of course, those were predictions with huge risk tolerances.

From this I decided, I needed to take my work in a new direction: I needed knowledge to improve and technology. My work can be summarized as combining the best of my natural abilities with the best of technology to create a new superior synthesis. This can be seen in my creation of the clone system which provided me knowledge.

What is important for me is that as I look toward technology that I don't forget where I came from. I don't forget what my strengths are. It is important that I focus in on what I do really well and continue to improve that with technology versus going off on a tangent. I've tried this in different ways. For example, I can predict the market in all time frames -- even months in advance. But, my personality is such that I don't like to wait for more then a few days to know if I was right or wrong. So, when I build systems then I look to build systems that hold for at most a few days because that's my personality. I don't like to risk a lot. A "fair" game is the one that I always win: that's fun. Losing is not fun. Losing big is not fun at all. I like limited risk/reward. I don't like the idea of unlimited risk.

So, for me it is all about combining my existing skills with technology. Summarized, it is human performance improvement through technology. It can take many forms. Systematic trading strategies and systems, automated systems, expert knowledge gained from systematic analysis of my trades, or developing/refining the methods I already use. For example, I will program into the computer my discretionary methods and see if they have value. If they don't then I will consider to discard them.

What's important for me though is I really focus in on where I'm using my energy the best and where I'm wasting my energy.. what types of things I could improve through technology... Its really important that I'm very clear on what I'm trying to do.
 
Status
Not open for further replies.
Back
Top