SchKithZo - GridBender EA (EURUSD) | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

MT4: how to change "EURUSD" to "#EURUSD"? 3 replies

Re: EurUsd short term 15 replies

did oanda just drop its spread for eurusd to 1 pip? 11 replies

EA for multiple lot limit order for EURUSD 0 replies

NFP nice bump up on EURUSD 2 replies

Attachments: SchKithZo - GridBender EA (EURUSD)
Exit Attachments

SchKithZo - GridBender EA (EURUSD)

  • #61
  • Aug 4, 2020 3:59am Aug 4, 2020 3:59am
Quoting goodexp
i backtest it on other pairs. use eurusd name,but replace other pairs history price to eurusd. it can open and close orhers,but results not as good as eurusd. so this ea not cheat in backtest. looks it do many backtest use strategy quant(eurusd),and chose the best setting to make the ea. so in the next months,it maybe have good performance. and it need change setting or codes every months "My plan is to develop at least 1 good trading system every 2-3 weeks for EURUSD trading pair and forward test them for some months to determine if they are worth...
Despite the suspicious EA size (1.4 MB!), it is obvious that this EA has been overfitted to EURUSD history data. Unfortunately, an overfitted model will not generalize and hence will fail on unseen future data.
Trading is the hardest way to make easy money...
  • #62
  • Aug 4, 2020 4:21am Aug 4, 2020 4:21am
Quoting leleitaly
I really have to congratulate you because a similar product, if real experimentation confirms the backtest at 99.9 would really be worth a lot of money, I can ask you for curiosities ... of course only if you think you want to answer. Are the settings those present at the time of download? How many indicators are present in this EA, because at the end of the backtest with visualization, I have seen a lot. Thanks so much
he used so many default indicator, that why in backTest EA works slowly even with out visual.....
  • #63
  • Aug 4, 2020 4:28am Aug 4, 2020 4:28am
Quoting MathTrader7
{quote} Despite the suspicious EA size (1.4 MB!), it is obvious that this EA has been overfitted to EURUSD history data. Unfortunately, an overfitted model will not generalize and hence will fail on unseen future data.
Nice point stated and noted..I dont know for others but my aim in developing systems is not to make holygrails or build winning systems.... What i aim at is just build something tradable and then have a data history of its performance on live forward test and then build money management around it.. Example i have a curve fitted system that trades with reward to risk ratio 1:1 and after backtesting the WinRatio is lets say 80% and highly profitable but only 40% Winrate when traded live even tho the whole dynamic has changed from a winning system to losing system all i need to know is the average,maximum consecutive losses and other parameters to build money management either by using cost averaging or even martingale in a very wise way to make money.. Not that its important but have turned mere $10k deposits into $250k and above trading with just money management,math and logic....you dont need a profitable system to make money trading
  • #64
  • Aug 4, 2020 4:34am Aug 4, 2020 4:34am
And for the curve fitting aspect i can run simulations on up to 10pairs and find systems which work on probably all pairs its just takes 20x more processing power and all that and since my aim is not in actually developing something profitable i dont see the need to spend 20x more money to get what i want done... Expensive servers and other paid subscriptional products are involved in my building process so its more about what i want done and whats available.. i could just work on my systems and not share them no matter the flaws they have..
  • #65
  • Aug 4, 2020 6:24am Aug 4, 2020 6:24am
Quoting MathTrader7
{quote} Despite the suspicious EA size (1.4 MB!), it is obvious that this EA has been overfitted to EURUSD history data. Unfortunately, an overfitted model will not generalize and hence will fail on unseen future data.
Maybe, maybe not. I’ve tested the EA on individual months, one at a time, and during several different yearly time frames, with different starting points. It always did well, and DD was always very low, even with MM.

You very well could be right. I’m willing to put some money at risk and will report back with real-world, live account results. Wish me luck.
  • #66
  • Aug 4, 2020 6:28am Aug 4, 2020 6:28am
Quoting ChicagoRob
{quote} Maybe, maybe not. I’ve tested the EA on individual months, one at a time, and during several different yearly time frames, with different starting points. It always did well, and DD was always very low, even with MM. You very well could be right. I’m willing to put some money at risk and will report back with real-world, live account results. Wish me luck.
Good luck!
Trading is the hardest way to make easy money...
  • #67
  • Aug 4, 2020 6:54am Aug 4, 2020 6:54am
Quoting ChicagoRob
{quote} Maybe, maybe not. I’ve tested the EA on individual months, one at a time, and during several different yearly time frames, with different starting points. It always did well, and DD was always very low, even with MM. You very well could be right. I’m willing to put some money at risk and will report back with real-world, live account results. Wish me luck.
Good Luck
  • #68
  • Aug 4, 2020 8:41am Aug 4, 2020 8:41am
The EA doesn't place trades on my mt4, despite the smiley being positive.

  • #69
  • Aug 4, 2020 8:47am Aug 4, 2020 8:47am
Quoting Daniel007
The EA doesn't place trades on my mt4, despite the smiley being positive.
You have to be patient. I noticed from my backtests that it can go for a few days, before placing a trade.
  • #70
  • Aug 4, 2020 8:52am Aug 4, 2020 8:52am
Quoting Daniel007
The EA doesn't place trades on my mt4, despite the smiley being positive.
Sometimes it takes a while daniel007.
  • #71
  • Aug 4, 2020 8:52am Aug 4, 2020 8:52am
Hello SchKithZo,
Thanks for sharing your EA. It's very well curve fitting from 2017 to today.
Have you performed a walk forward analysis to determine how much time the EA will be profitable?
  • #72
  • Aug 4, 2020 8:54am Aug 4, 2020 8:54am
@ChicagoRob @Egosu
Thanks guys! Omw to financial independance! :-D
Attached Image (click to enlarge)
Click to Enlarge Name: 845e665757db50af5a0578e20da18ad8.png Size: 1.1 MB
  • #73
  • Aug 4, 2020 8:59am Aug 4, 2020 8:59am
Quoting pedrofx
Hello SchKithZo, Thanks for sharing your EA. It's very well curve fitting from 2017 to today. Have you performed a walk forward analysis to determine how much time the EA will be profitable?
I actually am just really interested in seing how it performs on live market condition..even if it fails i will build money management around it and apply my own risk modules to still place the odds in my favor to gain a winning edge mathematically.
  • #74
  • Aug 4, 2020 9:00am Aug 4, 2020 9:00am
Quoting Daniel007
The EA doesn't place trades on my mt4, despite the smiley being positive.
Lol well thats because you did not read other post of other members.. that topic has already been addressed lol
  • #75
  • Aug 4, 2020 9:04am Aug 4, 2020 9:04am
Quoting SchKithZo
{quote} I actually am just really interested in seing how it performs on live market condition..even if it fails i will build money management around it and apply my own risk modules to still place the odd in my favor.

I am going over this thoroughly and it looks like you have curve fitted. Then when you make quotes like this, it starts to sound a bit strange about what your intentions are. Yes you shared your work (Without the source so that I can go through it and make any changes and share) but its fitted and then you are essentially getting testing data from other users to then "apply your own risk modules to place the odds in your favor". So what are your intentions exactly? I am happy to test and contribute, I will even code if I have to, but given that the backtest data is the "lure" even though its fitted and you have suggested you are hunting others to test it so you can then amend the risk modules to your favor I am wondering if its worth wasting my time on it.

For example, my backtest shows 10k to 2.5 million in 8 months. But I can get that same result for any 8 months I choose almost to the dollar, which means that the backtest is 100% useless.
  • #76
  • Aug 4, 2020 9:09am Aug 4, 2020 9:09am
Quoting Takisd
{quote} I am going over this thoroughly and it looks like you have curve fitted. Then when you make quotes like this, it starts to sound a bit strange about what your intentions are. Yes you shared your work (Without the source so that I can go through it and make any changes and share) but its fitted and then you are essentially getting testing data from other users to then "apply your own risk modules to place the odds in your favor". So what are your intentions exactly? I am happy to test and contribute, I will even code if I have to, but given...
The term curve fitted is about cherry picking results based on trade results so there is an uptrend bias.. My goal in sharing which i made clear was just to target and retain 10% of the winrate and profitability non compound using fixed lot trading
  • #77
  • Aug 4, 2020 9:13am Aug 4, 2020 9:13am
Example to me the backtest is also very useless but at least gives me an overview of whatever happened during this backtest period.. Example i test the system in 6 months and record 100 trades and i only get 30% winrate and the system does not make money but on data i know important and vital information like the maximum consecutive losses over 6months which is maybe a maximum of 5trades in a row i will know how to position the next set of 10trades using and appying money management either by using cost averaging or martingale.. i stated that clearly in previous replies..
  • #78
  • Aug 4, 2020 9:19am Aug 4, 2020 9:19am
Sometimes i dont get the over-skeptical and paranoid mindset in here..its sometimes discouraging and understandable everything is trash should not mean there should not be room for practical thinking.. Anyone wants to play a game of logic and game we can clearly run an experiment with a high frequency trading system that places lots of trades and is curve fitted and has a fix sl and tp and you will me amazed with what data derived from live trading results can still be geared to gain a winning edge and be highly profitable.. Have done a lot of research into systems and trading am all about the math i dont really care that much about the system itself.. with that being said my goal is to share and then gather vital information and also compare forward testing results to build money management and recode the strategy to still retain profitability...
  • #79
  • Aug 4, 2020 9:52am Aug 4, 2020 9:52am
instead of praising the immense work, especially free, envy the desire to criticize regardless, it would make anyone lose patience, explain clearly that you are satisfied with earning 10% of the BT result, I sincerely develop EA every day , and neither here nor in other forums or sites, I have seen such a great job
  • #80
  • Aug 4, 2020 9:54am Aug 4, 2020 9:54am
Quoting SchKithZo
Sometimes i dont get the over-skeptical and paranoid mindset in here..its sometimes discouraging and understandable everything is trash should not mean there should not be room for practical thinking.. Anyone wants to play a game of logic and game we can clearly run an experiment with a high frequency trading system that places lots of trades and is curve fitted and has a fix sl and tp and you will me amazed with what data derived from live trading results can still be geared to gain a winning edge and be highly profitable.. Have done a lot of research...

I have been a member here since 2005 and I've seen threads come and go, I've seen also people hide commercial intent more often than not. If you don't get 'it' read up on the jacko scam for example. Then you will understand why people are skeptical here, especially older members.
The variable parameters in the EA are so limited, I can't actually do anything with it that is worth while other than run it on a demo and report back. But the fact that its blocked to eurusd and requires a "licence" for other pairs as well as the fact you are a "new member" suggests you are planning to become commercial as the plays are already there.

Share the source code and I will believe that your intent is simply to give something for free and help the community.
0 traders viewing now