Bad Robots - The Swing n' Miss approach to learning Forex

Before you read anything else: I'm a complete novice to Forex. It’s hard to understate how little I know, so please bear that in my mind before posting.
(But I'm a bright guy and I think my approach is a good way to learn what’s what)

I'm tired of not having money, so I've decided to learn how to make money. I'm not looking to get rich, but I want my savings to make me more than the 2% annually that my bank offers.
My approach to learning is pretty straight forward: Do it and see what sticks.

This thread is gonna try and document my stupid attempts at making profitable robots, so hopefully it’ll be interesting
I’ve got no experience in coding, so please don’t laugh at how I approach “creating” robots.
I’ve got a feeling that it’ll be “easier” to start manually trading, but my RL circumstances don’t allow me to do that (two kids, wife and a job without a computer) and, more importantly, I know myself well enough to know that it would take me a long time (years) to learn the discipline of following a strategy.
I've listened to some podcasts and they didn't offer a lot except for a recommendation for John J. Murphy, so I'm reading him and he makes sense (so if you're an experienced trader, know that he's my only point of reference).

I’ve mashed together an EA in order to try things out and I made a couple of posts about it here and here. I’m quite surprised that it’s working (at the time of writing it’s only been running for 14 days), but it makes me hopeful that maybe this Forex business isn’t that hard after all

It’s ridiculously simple: 2 MA, open trade at a cross. I then close all trades when I have a combined 30 Pips. Each individual pair has a SL of 30 pips, so my account don’t go bananas.
I’m trading five pairs: AUDCHF, AUDUSD, EURCHF, EURUSD and NZDJPY. They were picked pretty much at random and I did some backtesting, so some of the settings are a little different, but not much: FMA= 8ish, SMA= 20ish. I started out with a basket TP of 50 pips but I changed it after the first week to 30 pips.

You can see the results in my trade explorer, and I’m trying to figure out what I’m doing wrong. I think that I’m risking quite a lot and being a bit lucky at the same time and that’s what’s giving me good results. I’m doing 0.1 lots per trade, so that’s ~1% risk on each open position, with a potential of 5 positions failing at once (or an infinite number of positions failing in a row while I’m sleeping), but so far it seems to be doing good. Am I being reckless?
The R:R ratio is a little tricky to figure out since I’m having a TP of [all trades=30pips] vs. a SL of [one trade =-30pips]. If anyone has any input on this, please comment

The trades are very random, so I think the only redeeming feature is the basket system, but I’m not sure how reliable it is (because the math is too hard for me).

Since demo accounts are free, I’ve decided to test the same concept, but with more pairs: My other account is trading on 23 pairs with a 30 SL on each individual pair and a TP of 30 pips on the basket.
My initial thoughts are that it’ll do fine, but I’m really worried that I’m not seeing the Worst Case Scenario. If anyone experienced can enlighten me, it’d be much appreciated
Also, I’m really not sure how I should change my lot size when I’m increasing my number of pairs. Should I keep the 1% per pair? Is it insane, since we’re talking 23 pairs? Should I increase the basket size?

Anyways, I hope this’ll be helpful to some people. If you have advice, please feel free to share it. If you don’t, feel free to ask questions.

/Simon

PS: so far, I’ve only used freely available robots, so there’s no magic there. Feel free to ask for the set files, they’re rubbish.

PPS: Everything here is a losing strategy.

Fairly strange week. The original newbie Ea continues to chuck along and I'm actually wondering whether it'll hit 10% growth in 4 weeks. I don't know if that's good or bad, but to me it's pretty impressive.

The newbie on steroids was really weird, starting out losing money big time. It reached a drawdown of 6.5% withing the first day, so I figured I'd run it till the end of the week and then disconnect it. The markets opened up triggering almost all of the pairs and then going against the positions, so most of them slowly worked towards their SL without my EA having an opportunity to close them, because there was never an overall profit. But once it got back on track it actually ended up with a profit high of 7.6% today! It then proceeded to produce some bug which opened and closed a little under 100 positions withing two minutes, ending in 2.8% loss. I've got no idea what happened (I even considered whether I had pocket-traded, since I have an app on my phone), but I'll definitely have to be superconfident in the software from now on.
The worst part is, it actually closed another basket, so at the time of writing it's only at 1.1% loss! It would've been a supergood week if it hadn't been for that freak bug!
I've asked both the authors of the EA's whether they have any idea what happened. I won't be disappointed if I don't get an answer, they're free after all.
Might have to learn coding myself in order to feel safe about ever using EAs

Nice beginning Simon.

This is not a cynical reply, just general conversation.

It appears the big picture has been very cleverly engineered to send those who wish to try, to enter down the rabbit hole of trickle feeding brokers our financial wealth, in a never ending search for the holy grail of methods to provide ourselves consistent profit.

From a retail traders perspective and what tools are available to us it takes quite some time and testing to eventually open your eyes and filter out the junk, and miss truths, to what might possibly work. Then eventually when you think you are close to the 'holy grail' it is time to go live, and it starts all over again, at a loss.

It appears the 'potential' profits you 'coulda shoulda' made, has you hooked already, and now along with any readers. Keep trying and don't waste a dime on a broker, yet.

I appreciate the feedback and I'll heed the advice of not throwing away my money just yet

The buggy EA still made progress and ended on an optimistic note, but I'm not gonna waste time trying to figure out how it would work without the bugs. It's out, but I took a screenshot of how it ended.

I found a different EA, which I have much more confidence in and I've set it up with pretty much the same settings. It has the potential to incorporate a lot more signals, so learning the ins and outs of this new EA will benefit me a lot in the future. I started it at a smaller amount, which is probably more in line with the amount I'd be willing to risk some day. Hopefully the new Newbie on Steroids will show the same potential as the old one did.

The original newbie is still running, but at a slight loss so far this week. I think maybe the monday opening is too volatile, so I might have to find a way to skip trades there.

A bad week all around, but I'm feeling kinda happy about it. Bad weeks will happen, and this was awful to my robots, but at least I feel like I'm reading realistic results now.
My original Newbie made -6.5% this week almost wiping out the progress of the last three weeks, but it's still showing 1.7% profit after four weeks. I'm not super impressed, considering the insane risk involved, but profitable after one month (with almost no hands on) is still worth something imo.

The new new Newbie did almost as poorly with 5.9% this week. The new EA has a feature which prevents the robot from opening several positions in the same direction of a single currency (so you're not vulnerable to crazy stuff happening), which seems like a good feature, but I think it prevented quite a lot of positions this week, so I'm gonna turn the feature off. Best thing that can happen is some black swan event that'll teach me patience

Strange week, but a few (embarrasingly obvious) things learned.

My original Newbie continued it's bad streak with a whooping -5.7%. It's now officially a losing strategy, who knew?
Newbie on Steroids was quite the opposite gaining 14.6% in a week. I guess you can get lucky at the casino after all. A few lessons learned here: I had set a TP on the individual trades, thinking it wouldn't matter because I was basket trading as well, but at a point it was in a big drawdown, and the recovery was made more difficult because some winning positions were closed at TP. The entire thing ended with a few positions reaching SL. I'm not sure it would've fared better if the winning positions had stayed open, but it wasn't working as intended. I corrected it over the weekend, so there's no individual TP anymore. The basket trading wasn't working as intended in another way: Since it's based on pips, some positions are worth more than others and some of my baskets closed with a lot less profit than others (because there were losing positions in "stronger" pairs than the winning ones). None of the baskets were closed with a deficit, but again, not working as intended. The EA I'm using has the option of closing baskets based on gains, so that's what I'm doing now. It's actually based on a percentage gain based on your balance, so it scales, which is nice.
Another issue is one I haven't really worked on yet. When baskets closed, some positions immediately reopened. This is because there is a signal to trade "lingering", which makes sense because often there are big moves triggering the baskets to close. This causes an even bigger delay to my already late entry, so it's definitely not working in my favor. I think it's a fairly easy fix, once I dig into it, but the entry itself is shit, so I'm not really sure I'm gonna do much about it now

I've noticed that a lot of the positions I open are so late, that it might be a decent entry for a retracement, except I'm trading in the wrong direction. I'm gonna try and experiment with a pending order in the opposite direction, to see if I can actually catch retracements with this entry.