Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
I've been trading similar strategies with real money for ~8 months, the results are comparable.
Note though that this backtest only used scalping orders, the real strategy has runners which generate much more profit but bring the win% down some as they have looser stop criteria. I would not trade the strategy as pictured IRL.
Long term backtesting is pretty pointless BTW for a scalping type strategy where the primary factors that matter cannot be backtested (ie, timing factors). If doing longer term trades it is somewhat useful as long as you don't fall into the optimization overfitting trap..
First guess would be you are using minute charts, and if you examine the Trades tab on the backtester it will show bars in trade to be 1. So since backtester doesn't know the order of OHLC, the win % is off the charts, while in real time it's terrible.
If this isn't the case I'd be curious for more details if for no other reason than to educate more people about problems like this.
So why haven't you retired to Barbados with a pile of cash already?? With 80% winners (lower than your ~90%), you only need the profit factor (average win/average loss) to be >0.25 (in other words you ave loss can be 4x than your ave win) to have a positive ROI (albeit with a low expectation of ~0.01)...With 80% wins and a profit factor of just 1 (i.e. ave win = ave loss), the expectation is ~0.6!
It would be intersting to know what your %win and profit factor is...and of course the best way to silence any doubters is to post the equity curve from your 8 months of trading with real money Can't wait to see it
I agree, but it has the same limitations if using minute data and can still be completely inaccurate. It's best to code strategies to use tick data or to write strategies so they cannot take profits until after the first bar, or to just dump the trades tab into csv and turn every winner on bar1 into a loser, and see if it's still profitable.
Perfectly agree with that ....... I will try .... I can PM you the strategy so we can take a look together .... I am working on completely different strategies anyway.....
I beleive that one big problem affecting it is also curve-fitting .....
Now related to tick strategies , I agree itis much more accurate ..... but I have never been able to test on tick data ....each time it returned 0 $ ( with both IB and TD) when I tried... any advice ?
I disagree. I've written lots of strategies, ran probably 6-8 with real money, probably 4 for several months or more. They ALL fell apart. The market is constantly changing and it's not easy for a strategy to hold up. I have only one that has held up and it uses daily data. I don't trade it cause it has large drawdowns.
Mike - From what I understand Ninja cannot reliably backtest when one enters & exits on the same bar. You can enter or you can exit but not both on the same bar. I believe this limitation is for all bar types including tick. When you backtest a tick chart it doesn't go tick by tick, it uses the OHLC of the tick bar. Please correct me if I'm wrong.
Any more I have given up on automated strategies. Even to test an idea. Now I just write a little indicator that will mark the entry bar with a dot and I draw a line for the target & stop (if it's fixed). Then I eyeball it to see if it will work. The dot can help me trade it real time. I write out the MFE & MAE for each trade in excel and then I compute the average and refine my target & stop. I've had far more success with this method than with automation. Staying out of chop and news is not easy to do in automation. Also I use the daily range to anticipate for the day.
It's just too complex for me to program it and i'm a professional software developer by profession. My current project is rewriting from scratch the system that analyzes tv audience (like the Nielson ratings). It's really complex but in my opinion not as complex as an automated strategy!