If you enter a trader, long or short, why are you entering and what is the edge you have identified that
this entry is a good entry ?
Subsequent there is the discussion of the exit, if you don't know there are a couple ways to approach it
Kills the trade can be based on :
- risk management
- the "thing" that is your, says you don't need to be in this trade any more
- any other thing that you have identified , that justifies to kill the trade, base on risk or edge
For me, starting a trade based on risk, seems the opposite world
For me, risk management can only tell me, not to take a certain trade, not inverse
Attached is my edge. My edge is I know how to identify daily support and resistance, and I identify with my eyes the direction of price. The trend of the moment, I call it. I gained this edge from 2 years of just look in that the chart alone each morning. No indicators, just watch whichever way price goes each day.
I would like to enter the trade once price had broken the next level of support or resistance. That is my confirmation that price may continue to in that direction, and I may be profitable.
See my attachment example: I would like to go short below the green line somewhere.
My problem: I don't know where to enter once price has broken and MOST importantly, I do not know what data to collect when manual back testing that will allow me to analyze optimal entry, stop loss, profit taking.
My Possible solution for stop loss: Decide max amount of money i want to lose per trade, one contract. This is my stop loss. Example, -$150. 15 ticks for /CL.
My Possible solution for entry: I don't know, 5 ticks below the green line with a stop market order.
My possible solution for break-even: Common sense, if I am up $150 bring stop to break even. Just my thinking.
My possible solution for profit taking: 1. Trail by 15 ticks 2. Exit for 15 ticks 3. Exit at next support 4. Exit at 30 ticks. I have no idea.
Here is my road block and current weakness? I have no idea how to collect data to test the optima entry, exit, stop loss, nor breakeven besides risk management I explained above. I would hate to do months of backtesting this idea to find, I forgot to record something that is needed during the data analysis.
1. How do I build a spreadsheet so I can collect data to determine the above "I don't knows"?
2. Do traders build programs to test alll the different scenarios above?
So here is where I am with my idea/strategy. I want to determine if my idea is profitable, but I want to be efficient as possible.
Thanks for any help
Last edited by goodoboy; December 30th, 2016 at 01:46 PM.
1. false break-out, sometimes the price will break the level and when you are in the trade,
the price will rotate and just come back, it is known as a 'false break-out', some people are
by purpose moving the price to simulate a break-out, to trigger all the stops and and then
two things you can do :
- try to find a means of detecting, is it a break-out legitimate or not ?
- deal with it and measure statistically if it is still profitable in the long run
2. if the market is trending, trading is easy, compared to chop market and just sideways
everybody knows that, the question is how do you detect the market is trending and you
will not be caught in chop ?
You can build a strategy that will trade those levels when broken
you can then just replay the historic prices and check what is the result MAE/MFE, expectancy, % win/loss, etc...
Most of the methods I use are based on finding suitable places for long entries during uptrends and for short entries during downtrends.
I try to identify them from bar patterns according to what I've learnt from the various price action authors whose work has proved most reliable to me. They're all ultimately based on parameters of support and resistance, however localised.
Personally, I would test all of those, and more - apart from the first one. (I've done enough backtesting and forward testing of different exit methods to believe that the chances of an automated trailing stop ever working out "best" for me are simply too small for monitoring/testing that one to be worthwhile.)
But my "numbers of ticks" to test will always be volatility-related and S/R-related rather than just fixed numbers.
Exits are a big and important part of trade management (more important than entires, in my view), and like all trade management parameters, one ultimately derives/proves them by meticulous testing and the collation of results. It's really labour-intensive and time-consuming. All true but not very helpful, arguably, in that one still has to decide which parameters to test in the first place.
It all depends on the relative extents to which people are motivated by "fear of missing out" and "fear of loss", which in turn depend on their own overall risk management parameters.
I try to combine "not losing too much profit already made" with "not missing out on too much future profit potentially available" by dividing my trade, closing some proportion of it at a fixed target according to the volatility, and seeing if the remaining part will go further, depending on anticipated support or resistance. It's just what I was taught to do.
The following user says Thank You to Tymbeline for this post:
I have to admire someone who puts thought into how they plan to trade.
What I am learning is that a trader really does not know how profitable his system can be if traders does not record the right data. The trader initial idea of the system after backtesting may results in profitability some other way. For me, I would not want to backtest manually some data for a month or so, to find out I missed some critical information.
Out of curiosity, for a system like I am discussing here, can someone program all the parameters of interesting and the automatically backtest and obtain the results rather then manual backtesting. In the future, after I do some backtesting, this will be my goal.
I appreciate your thoughts. I just bought a book by Kevin Davey, Building Winning Algorithmic Trading Systems, although I don't know how to program, not yet anyway, I look forward to his method on building and testing systems.
My system (which is a series of subsystems) initially, set out to estimate what the HOD and LOD would be and which one was first. I then added estimating what the first leg of the day would be - where and when would it start and end.
I have set a rule that I want trades of 6pts or more (ES). From this other things flow.
For example, if I want 6PPP (6 pt profit potential) then it is not worth trading days of less than 12 pt range.
You asked about risk management / stop loss for setting your rules.
I think of it the other way around. Where is the profit potential?
"where do I place my entries/exits?"
If there is an uptrend for the first leg
the distance to the HOD is 6pt +
I would enter long at the estimated first point with a target of the HOD
and a stop of 5pts
If there is an uptrend for the first leg
the distance to the HOD < 6pt
I would enter short after the HOD with a target of the LOD
and a stop of 5pts
Exit = price exhaustion
Stop loss adjustment = BE once price has moved 5pts in my favour.
... so it continues for other situations.
If my signals are unclear prior to the opening I would wait until the market had shown:
choppy sideways mess.
I don't think any of the above is useful to your situation as it appears you are looking for indicator based system.
From your comments on S/R levels I would suggest (for the ES) that a 50% retracement point be considered.
Good trading to everyone.
Last edited by aquarian1; December 30th, 2016 at 04:04 PM.