A new addition. Worst closed trade in every pair. Its plotted on chart too. Mainly for fast check rules bad mistakes and try fixed them
Try don't lose pants never...
1
MagicGirl dashboard 221 replies
Help with dashboard style indicator 2 replies
Dashboard indicators 4 replies
Dashboard Indicator Required 1 reply
Dashboard/Grid in lower indictor window 6 replies
Dislikedhumble and honest advice when/if you overfit in history then it might most probably fail in the futureIgnored
Disliked{quote} I know past not guarantee future. Its main reason i have some extra report info for try guess if it will be reliable in long term. Anyway this kind of reports are more headed for help to improve rules in coding stage. But still it can give an idea how can performs system ( if its red then its clear it will be a disaster, if its green its a hope ) If you compare reports i uploaded last weeks you can see DD decreased a lot. That means system risk is lower now. Still not don't guarantee anything but improvement and higher success chances are...Ignored
Dislikedhumble and honest advice when/if you overfit in history then it might most probably fail in the futureIgnored
DislikedNow i have TE slightly messed due to builds changes and broker symbols names changed. This is "clean" run with last stable build and new symbols names ( 21k performance build ) While I dont change build overall is always green ( as backtest reports ). When i change build things start to be ugly with old trades since new build cycle is very different and cant manage old trades as expected Max DD it was less -200p. Not so high for a cold start {image}Ignored
Disliked{quote} A couple things I forgot comment about history backtest. 1) From system perspective, backtest is future since system runs as should do in real time. System dont know next candle and if make a performance in a date range, then it should be same if it was running in real market ( ignoring swap/spread/slippage differences ). 2) At least in my case, backtest AI for get optimal parameters is not tricky as search parameters giving the best profit. Now its simple but it uses some collected data for decide what parameters it will be more consistent...Ignored
Disliked{quote} backtest is essential. no dubt. just sayin, overfitted curves fails on the forward testsIgnored
Disliked{quote} My question is why you think mine is overfitted. All opinions are welcomeIgnored
Disliked{quote} I am not. however while you fix every error, its getting overfitted. I have coded too many lines, probably close to million. while you are making your system you are overfitting due you are working with history data. how much you fix, that much you overfit. there is a line that should not be passed. here is the metric; how much you make profit with history data that much it fails with forward testIgnored
DislikedNew build finished ( v0.17 +38K pips ). Prev build was around 21K pips with higher DD I have been add a parameters cache because as MT4 is single thread it needs 30m for process all parameters. I added parameters info in every pair in reports too I will use tomorrow in TE account {image}Ignored
Disliked{quote} I was just curious if you planned on sharing your ea and preset after you are finished with optimization?Ignored
Disliked{quote} yes, i agree Brain.. brilliant strategy not necessarily get a good profit without good money management. that i also test for some money management with consider how much balance need. {image} look have good build for u, brain. on there can test many pair in mt4. and can see the result like that.Ignored