I don't see the website link?

Is it a forward or reverse grid.... ie add to winners or add to losers?  What timeframe does it run on and do you manually handle it during market transitions or does it figure it out on it's own?  I've tinkered with some grids myself in calgo but as I'm sure you know it's hard to get one that doesn't blow up at some point.  I had a pretty neat one that was driven off the daily/4h heiken ashi (with some overbought/sold and trend filters) and could get some pretty massive scores in demo but it was difficult to get stable enough to run off less than a 10k account and I didn't trust it enough to put it on one lol.

Ran it through the backtester over 2 years with 1k lot size and 10k account... It works beautifully until it doesn't (and then it consumes the entire content of the account in very short order.)  Know a guy who has a pretty wild money management strategy with grids that works for him.  Instead of keeping it small trying to keep it alive, he pushes big lots knowing that they'll blow up at some point and then harvests weekly so he can restart after a blowup (as well as running baskets which it looks like you're already doing.) 

I've always been more of a fan of the snowball style of grid where you add to winners and only add positions when the stops are at breakeven, going the other way just seems like a fractionalized martingale.

Could you explain why you view backtesting of limited use?

thank you akoorts, I backtested on MT4 with FxPro demo account, timeframe H1 EurUsd in 2016 only. It's very risky, it keeps opening the same position while increasing the lot size but account balance can go down fast, and it went down from 1.000$ to less then 300$ in May.

It's better for accounts with big balance, I seen you have a 30.000$ real account doubled in one year starting with 0.01 lot size, but what why did you stop the EA at July 2016?