Pathway To The Black Box

[Timmyb]

Well I do belive that when we moved, most threads were deleted, If i am wrong I apologize MC for starting yet another thread, It is 4 am here, and i havent slept in 26 hours, Had a crap load of cars to work on, and well the developing of this "black box" is keeping me awake at night, Which in general when this occurs I get my best ideas, I finished a waladoch conversion paint job on a van once in 37 hours so hopefully things will start to flow, I figured i would start by showing some indicators i have developed, Nothing fancy but allot for me since a month ago I had no C sharp programming knowledge.

[Timmyb]

Also have TEMA, SMA version, hoping to combine all in a easy drop down menu in ninja instead of different ones, I really like TEMA adn EMA

[Timmyb]

Calculates the rise to run, or the angle of the EMA with thresh holds




Also have it for SMA and TEMA

[Timmyb]

ADX in Linear format, totally adjustable thresholds to show gaining adn falling ADX in any period or past bar calculations, IE more bars back the stronger and longer the angle calculation




In process of doing a DMI divergance and also MACD divergance

[MC]

I'm impressed Timmy...that is awesome stuff!
You're thread is still here, nothing was lost in our conversions. BUT this deserves it's own thread just for your programming stuff.

[Timmyb]

Well here I sit in the middle of a old quiet mechanic shop. Pondering once again the complexities of life, Trying to come up with that great idea that will make me get an income from the market without being glued to the screen for hours daily. I have been developing indicator, mostly inidicators that exist and converting to what i consider a more user freindly format. I read a bit online last night that as much as 20% of the futures in the ES could be automated hedge funds. This has been bothering me lately. If the hedge funds can muster up an auto trading algorithm for thier use, This should not be far away from a consumer level trader who is not detterered by hours of developement. My standars have been considerebly lowered lately to what is expected from daily return.

The new target which i hope to not be out of line is as follows:

2 point profit daily

The loss for the day is killing me, In bactesting it appears that by limiting gains i lose on some killer days, and sometimes turn a good day into a loser by letting the pc trade more, The loss limit for the day has been also problematic. What works for one month doesnt always work for antoher month:

For these reasons I have been trying to develope a system with less signals but with around 80% chances of being right, But I also need to have gains be almost double of the losses. Keeping a max days loss at a 2 day recover would be nice.

When Ninja trader allows multiple indicators and timefreame in the next release i feel i can nail it, Because then i can reference bigger timeframes in order to trade the shorter ones. I have also realized in the last 2 weeks that my daily charting skills on the ES far outweight my personal intraday abilities.

Trying to add daily info to the intraday trade i feel is key, Perhaps by analyzing the previous days indicators in order to change the way the pc trades intraday.

Onother big thing i have been combating is chop

To me chop should be easier to identify using volume and range bars interpretation of price action, In theory i want entrance in trends and not chops.

Ok will quit rattling for now

Tim

[trustdnb]

Originally Posted by Timmyb

For these reasons I have been trying to develope a system with less signals but with around 80% chances of being right, But I also need to have gains be almost double of the losses. Keeping a max days loss at a 2 day recover would be nice.

Tim

Hey Tim, kudos on your hard work. I'm wondering if you've read Van Tharp? I only ask because he does a great job of explaining why you dont need 80% reliability to make money if you focus on improving the other parts of your system. For example, trend following systems that lose on 3 out 4 trades but whose winners return 5x or 6x the initial risk is still has a positive expectancy. We often get that "need to be right bias" that draws us to trying to maximize the reliability of entries while ignoring the effect that stop placement, position sizing, and exit strategy.

I am sure there are 80% reliability system out there, but I doubt there are very many that provide gains that are twice their loses. If you winning 4/5 trades, then you average gain is most like less than your average loss. Conversely there are systems that will give you gains that are 2x or 3x your risk, but the win/loss ratio is probably shy of 50%.

That said if you can build a system with the parameters you have described than you will have a bona fide holy grail in your hands. Good luck in your search.

[MC]

Originally Posted by trustdnb

Hey Tim, kudos on your hard work. I'm wondering if you've read Van Tharp? I only ask because he does a great job of explaining why you dont need 80% reliability to make money if you focus on improving the other parts of your system. For example, trend following systems that lose on 3 out 4 trades but whose winners return 5x or 6x the initial risk is still has a positive expectancy. We often get that "need to be right bias" that draws us to trying to maximize the reliability of entries while ignoring the effect that stop placement, position sizing, and exit strategy.

I am sure there are 80% reliability system out there, but I doubt there are very many that provide gains that are twice their loses. If you winning 4/5 trades, then you average gain is most like less than your average loss. Conversely there are systems that will give you gains that are 2x or 3x your risk, but the win/loss ratio is probably shy of 50%.

That said if you can build a system with the parameters you have described than you will have a bona fide holy grail in your hands. Good luck in your search.

This was what I wanted to say but couldn't find the wording for some reason.
I couldn't have said it better myself, otherwise I would have.

The need to be right is a hard habit to kick and I still struggle with it myself. Probably all an 80% system would fuel would be the mind. In order to be right that often, odds are you're not letting winners run and are probably scalping where a few losses will chew up profits. Scalping is hard because you often need to give wide stops and tight gains, exactly the opposite of how most succeed in the market. There are scalpers that make mad money though, so it can be done.

I wish you success no matter what type of trader you are Timmy.

[Timmyb]

Hi,

Here i am again, been busy with real life lately, trying to get things back in order, I need to buy a new pc for the office, because the one that i use now wont run Ninja and do the calculations nessecary with enough stability, So that means i have to do most of my research etc at home, which i am not there very often due to staying a girlfreinds house, Which in turn makes everything grind to a halt. I switched venues today and traded the russell, and i must say i really liked it, i think i am getting burned out on the ES and will be concentrating more efforst on the YM or ER2. My system seems to work better on the YM and russell also.

Decided on a few new rules to try to establish

Minumum of 3-1 risk/reward ratio on traded on bactesting

Tryin to keep stop around 100 dollars with a goal of 300 on average trades, otherwise 200

I worry allot about risking 5 points to get 1 point and i just feel that that kind of trading is far to risky, so even if it takes me a year longer out of the market to perfect i want to try to establish a 3-1 or better risk to reward

I have begun programming candle pattern analyzing for Ninja to try to get away from total indicator trading, I feel this will help to get me into a more price action type of trading.


I have built about 5 more indicators but they are just variations of some that are available so i havent posted them, if i find one i really like i will post it.

Tim

[MC]

Sounds like your busy Timmy. Don't forget to take some time away from coding and the market to do some fun stuff. I got stuck in the grind doing work, then the site and on top of it all trying to learn more about the markets. I got burned out, took a break and boy did I need it.

Good luck on the YM and keep up the good work on the system coding.