Coding quality is a basic requirement. The number of parameters is most important and tells you if you're just playing around or you've found something worthwhile. You cannot completely remove noise without removing some crucial price data as well. By "backtests are backtests" I meant reality is always worse than a backtest in the long-term. You will have 2 losers in a row at some point. If you truly believe this works that well, $3k is nothing and you'll be making millions with it.
hello, yes, some part of price data removed, this is only solution, to bound price independent-uncontrolled economy variables. long term risks can be removed by different setups, %risk, stop levels etc. future is unknown. to eliminate "future" we need to simulate quanta levels, the probability of price goes a,b,c,d stop levels are different. this is the way how electrons work in mass, that produce "future" i.e "time" . the plan is an investment bank. with more complicated algos, including this one. hope it'll work.
Hello, Good stats and thanks for sharing with me Few questions please 1.Did you backtest with your own purchase tick data or you use the platform historical OHCL price data? I use ninjatrader and only OHCL I'd available for back testing. Thanks
hello, calculations are made at bar openings. here we use M15 bars. in some algo's we use M1 bar openings. other ticks do not produce trade calculation. a healthy OHLC data is enough for this algo. but for some tick scalper algos, we need all ticks.
Thanks for response, Yes, I used OHLC data as well as i do not scalp and like algo with RR of 2 or more with +17 stop loss. And targets above +40 ticks for intra-day trading. Plus using the OHLC is for free and keep the costs down not having to pay for expensive tick data. 1. Are you selling this algo? 2. Did you test it live? If yes, for how long? Thanks
hello, the main problem is data time and economic calendar times matching, this forces to use a suitable dataset. yes, in the first post there is a mql5 market link, FPO. development-testing was in real account. but experimental, during many years. i publish many tips in previous posts. so you can experiment old trades in mt4-mql5, using old calendar data. and can see the idea.
And then after "hello friends" we had... "Trade decision is made by some standart indicators like Stochastics, ATR and CCI. Decisions are made at bar openings." Doesn't take much to know the game is up, just a few seconds of reading does it.