What do you think of this strategy?
- Thread starter chaykapwr
- Start date
What do you expect people to say? You post these hypothetical resuits that are beyond amazing. So either you are have found the holy grail, and are on your way to creating the next Rennisance, or you have made some sort of error in your backtest. If the former is true, we are all truly blessed to witness the humble beginnings of such greatness sprouting on the ET forums.
Based on your own analytical abilities though, which scenario appears more likely, especially to the third party observer?
You seem to understand some of the possible mistakes with backtesting. However, no one is going to go through your model line by line and determine for sure whether or not it is flawed. If your backtests are truly fair and accurate, I encourage you to get off this board, and begin the road to making your billions
Based on your own analytical abilities though, which scenario appears more likely, especially to the third party observer?
You seem to understand some of the possible mistakes with backtesting. However, no one is going to go through your model line by line and determine for sure whether or not it is flawed. If your backtests are truly fair and accurate, I encourage you to get off this board, and begin the road to making your billions
You seem to be pretty confident you've got it figured out -- why waste your time posting/responding here? Why not just trade your strategy? If it's really as good as you seem to think it is, you can build a hedge fund around it, and you can spend your free time talking to your clients about how much money you're making them rather than defending your claimed backtest results anonymously on ET.
Please guys, the OP was very polite. Maybe he has made some mistakes but we all have at some point in time, haven't we?
If you have a 10-minute bar where all these events can occur and you go into it flat, then you cannot know which event happened first unless you use tick data.
The only legitimate approach is to assume that you hit your stop. You cannot then initiate another position until the close of the bar but it can only be a MOC, not a stop or limit order.
This is the reason your equity curve is like you bought a T-Bill.
If you have a 10-minute bar where all these events can occur and you go into it flat, then you cannot know which event happened first unless you use tick data.
The only legitimate approach is to assume that you hit your stop. You cannot then initiate another position until the close of the bar but it can only be a MOC, not a stop or limit order.
This is the reason your equity curve is like you bought a T-Bill.
I looked at what you are trying to do on ES journal.
It looks as though you are trying to catch tops and bottoms with a small stop (1.25) for a 4+ point profit. More power to you.
But, to be true to yourself, you need to be backtesting/walk forward with a 4 tick price bar. This will eliminate any illusions as to whether you got flipped and will provide true price action.
Do you set your own PASR or do you use pivots for S/R?
Good luck with your "find".
It looks as though you are trying to catch tops and bottoms with a small stop (1.25) for a 4+ point profit. More power to you.
But, to be true to yourself, you need to be backtesting/walk forward with a 4 tick price bar. This will eliminate any illusions as to whether you got flipped and will provide true price action.
Do you set your own PASR or do you use pivots for S/R?
Good luck with your "find".
That's a problem. You're making decisions at a faster rate than you are sampling/bucketing the price data. Backtesting analogue of violating Nyquist.
Who are your competitors?