Quite reasonable. If you are into slow holding and sitting out it should be profitable.
The reason for failure regarding retracements might be laziness to explore non-linear systems theory.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
POLL - Futures: How do you determine the trend?
- Thread starter billpritjr
- Start date
Quite reasonable. If you are into slow holding and sitting out it should be profitable.
The reason for failure regarding retracements might be laziness to explore non-linear systems theory.
Quote from hirsch.im.wald:
Quite reasonable. If you are into slow holding and sitting out it should be profitable.
The reason for failure regarding retracements might be laziness to explore non-linear systems theory.
No need to call me lazy, I have some very profitable systems and work very hard. I don't understand what you mean by "non-linear systems theory," could you please elaborate.
Could you give me an example of a non-linear system that you know of that is profitable trading retracements?
Thanks
5yr
That the term laziness offended you is a very good sign. So I assume I have to apologize.
To be more specific:
Markets are defined by the actions of their agents.
Agents don't buy or sell in a linear sense, they are using strategies, i.e. complex modes of recursive actions.
Agents tend to watch and interpret actions of other agents. In instances where they try to avoid the assumed action of other agents we have an unstable, non-equilibrium environment, described best by evolutionary game theory.
In instances where they indeed try to mirror the actions of other agents we have a reinforcing, temporarily stable situation. An environment best described by the term "Eigenbehaviour", which would be cybernetics.
If you were to measure timely and perceivable actions around self- reinforcing phenomena such as oscillations and areas of high volume and were to especially look for "failure of self-reinforcement" or at least divergence from the path of "linear self- reinforcement" you could reasonably foretell with a better than random precision when the temporarily stable environment will collapse to the unstable environment again.
That gives you a non-linear, systemic continuation pattern and a non-linear, systemic stop-loss.
The main point is the recursive function of cause and effect. As long as a pattern of actions is perceived to be ESS (evolutionary stable) it will be prefered. If agents' preference make it more successful (i.e. it's self-reinforcing) it will be ESS in realiter. Agents with other patterns of action will finally switch over.
Sadly the number of agents in every timeframe is rather stable and definitely not infinite. Failure of further reinforcement will crack the underlying perception rather quickly.
The real work starts of course with finding patterns of actions that are self-reinforcing and meaningful for your timeframe. Watching floor-traders' confidence around pivot points won't help you at all, if your time frame is rather long.
I'd rather not give a specific example. I'm one of those guys who thinks that indeed an edge can vanish into thin air in seconds. And I wouldn't post this if I thought it's helpful to a lot of folks.
It's basically just general courtesy and if you're hard working and successful you might not even need to care about other people's approach. Just continue what you do and good luck!
To be more specific:
Markets are defined by the actions of their agents.
Agents don't buy or sell in a linear sense, they are using strategies, i.e. complex modes of recursive actions.
Agents tend to watch and interpret actions of other agents. In instances where they try to avoid the assumed action of other agents we have an unstable, non-equilibrium environment, described best by evolutionary game theory.
In instances where they indeed try to mirror the actions of other agents we have a reinforcing, temporarily stable situation. An environment best described by the term "Eigenbehaviour", which would be cybernetics.
If you were to measure timely and perceivable actions around self- reinforcing phenomena such as oscillations and areas of high volume and were to especially look for "failure of self-reinforcement" or at least divergence from the path of "linear self- reinforcement" you could reasonably foretell with a better than random precision when the temporarily stable environment will collapse to the unstable environment again.
That gives you a non-linear, systemic continuation pattern and a non-linear, systemic stop-loss.
The main point is the recursive function of cause and effect. As long as a pattern of actions is perceived to be ESS (evolutionary stable) it will be prefered. If agents' preference make it more successful (i.e. it's self-reinforcing) it will be ESS in realiter. Agents with other patterns of action will finally switch over.
Sadly the number of agents in every timeframe is rather stable and definitely not infinite. Failure of further reinforcement will crack the underlying perception rather quickly.
The real work starts of course with finding patterns of actions that are self-reinforcing and meaningful for your timeframe. Watching floor-traders' confidence around pivot points won't help you at all, if your time frame is rather long.
I'd rather not give a specific example. I'm one of those guys who thinks that indeed an edge can vanish into thin air in seconds. And I wouldn't post this if I thought it's helpful to a lot of folks.
It's basically just general courtesy and if you're hard working and successful you might not even need to care about other people's approach. Just continue what you do and good luck!
Thank you hirsch.im.wald for the lengthy and confusing reply. You seem to be answering my question on a theoritical basis, while I am looking for a practical (useful) answer. I am a technical trader, I am used to developing things that work not thinking of reasons why something will/will not work. Best way to test something is to actually do it, not talk about it. From my research I have found no evidence that volume adds anything to any indicators. How do you judge whether a pull back from a trend is just a temporary blip in the stable self reinforcing environment or the beginning of the collapse to the unstable environment?
I believe you were identifying a trending market as a stable environment and a non-trending market as "unstable." One would think that an unstable time would be marked by higher volatility which is exactly the time when trend following systems work the best.
Nothing you said in your post helped me to understand how a system could be created to profitably trade pullbacks, or leads me to believe that you know of one.
Also, I am curious as to where you got the idea that I paid any attention to "floor-traders' confidence around pivot points." I am a systematic trader, I would have a hard time programming that into a computer. Which leads me to what I am looking to you for: I need some sort of quantifiable way to identify your unstable and stable market environments, just like a moving avg. Or maybe you are to lazy to do this, if that is the case please type a long, complicated, indecipherable response to this.
5yr
I believe you were identifying a trending market as a stable environment and a non-trending market as "unstable." One would think that an unstable time would be marked by higher volatility which is exactly the time when trend following systems work the best.
Nothing you said in your post helped me to understand how a system could be created to profitably trade pullbacks, or leads me to believe that you know of one.
Also, I am curious as to where you got the idea that I paid any attention to "floor-traders' confidence around pivot points." I am a systematic trader, I would have a hard time programming that into a computer. Which leads me to what I am looking to you for: I need some sort of quantifiable way to identify your unstable and stable market environments, just like a moving avg. Or maybe you are to lazy to do this, if that is the case please type a long, complicated, indecipherable response to this.
5yr
I have no idea how to code my ideas.
In the moment I'm working at a defined decision process for my discretionary trading.
Pivot points were just an example, because I think it's an example for something self-reinforcing. You could measure retracements from pivots in a Wyckoff style and if the retracement is far too strong or far too weak you have some kind of a signal.
You would have to take into account that the combined buying power of all people looking at pivots, while big enough to be measurable, is limited.
Regarding trends, my concept of trend is more like an invisible underlying force. Price is swinging around the underlying trend, that's a view Bill and I have in common. So I am looking for stable oscillation patterns around an underlying drift. Higher highs and higher lows so to say. But I have criteria when I think the pattern of higher highs and higher lows might be breaking.
Ultimately I think there is a way to make a system out of it, but that's far from where I am right now.
The other thing is, my basic assumption is that the buying power of the locals in a given time frame is relatively stable. If you are trying to backtest a system based on that assumption for twenty years it won't look impressive, because it clearly does change over long time.
PS: I don't know what stockmarket almanac is, so I guess it's no.
In the moment I'm working at a defined decision process for my discretionary trading.
Pivot points were just an example, because I think it's an example for something self-reinforcing. You could measure retracements from pivots in a Wyckoff style and if the retracement is far too strong or far too weak you have some kind of a signal.
You would have to take into account that the combined buying power of all people looking at pivots, while big enough to be measurable, is limited.
Regarding trends, my concept of trend is more like an invisible underlying force. Price is swinging around the underlying trend, that's a view Bill and I have in common. So I am looking for stable oscillation patterns around an underlying drift. Higher highs and higher lows so to say. But I have criteria when I think the pattern of higher highs and higher lows might be breaking.
Ultimately I think there is a way to make a system out of it, but that's far from where I am right now.
The other thing is, my basic assumption is that the buying power of the locals in a given time frame is relatively stable. If you are trying to backtest a system based on that assumption for twenty years it won't look impressive, because it clearly does change over long time.
PS: I don't know what stockmarket almanac is, so I guess it's no.
Quote from hirsch.im.wald:
I
PS: I don't know what stockmarket almanac is, so I guess it's no.
note the company name:
http://www.stocktradersalmanac.com/
surfer
Quote from hirsch.im.wald:
Quite reasonable. If you are into slow holding and sitting out it should be profitable.
The reason for failure regarding retracements might be laziness to explore non-linear systems theory.
So going back to your original response to my post about buying retracements, is that you have also found no way to do this in a systematic manner, correct? My original point is that it is actually a higher risk trade than buying new highs (or selling lows) because it is impossible to tell right now whether the trend is changing or if it is just a retracement.
It is actually quite easy to define trends, any of the ones mentioned above work, and taking trades only in direction of that trend will make any trader a more profitable and stable trader. The trend definition would have to be defined over a time frame appropriate for the given traders average trade time of course.