Breakout Builder MT5
- Uzmanlar
- Sürüm: 3.14
Why is this free? There's a catch, but it isn't money. (I'll be updating this with a YouTube video and more pics soon)
I'm hoping some of you build something good with this and tell me about it.
That's the whole plan. It's a kit for building breakout systems, and there are more combinations in it than I'll ever get through on my own. A few thousand people messing about with it will find things I'd never think of. If you find something that works and you can show me the backtest, I want to see it.
At some point I'd like to put out a paid EA that's a collection of the best strategies people send in. If yours goes in it, you get a share of what it makes. Agreed in writing, before anything ships. I'm not putting someone else's work in a paid product without paying them for it.
And if that never happens, the tool's still here and it's still free. This market is full of martingale bots with a fake screenshot and a 99% win rate. I'd rather put out something real and see what comes back.
Nothing is locked. No trial, no "message me for the good settings". It all works from the moment you install it.
What it does
Every breakout system is three questions. Where's the range, how do you get in, how do you get out. Most EAs answer those for you and hide the answers. This one puts all three in the inputs.
Ranges: previous day, previous few days, last week, last month, a session window you pick, the opening range of the first N minutes, a Donchian channel, an inside day, an NR4 or NR7, or one specific candle after the open.
Entries: stop orders sitting at the edges, wait for a close beyond, wait for a retest, or fade it instead — sell the top and buy the bottom. There's also a fakeout fade for when price pokes out and gets rejected back inside. Long only, short only or both. Scale in with up to four orders if you want.
Exits: stops as points, ATR, the far side of the range, the middle, or a slice of the range width. Targets as points, ATR, R multiples, range width, or none at all. Then break-even, trailing, delayed trailing, runners, time exits, flat at the close, flat on Friday.
Filters: spread cap, minimum range size, news blackout, day-of-week, and a regime filter so you only trade when ADX says there's a trend, or only when it's quiet, or only in expanding volatility.
Everything's off by default. Turn on one thing at a time or you'll never know which one did what.
The four ready-made strategies
Off by default. Each has its own magic number and symbol and they all run together on one chart.
US 30-minute range on the S&P. Wait for the first half hour of the US session, get in when a 5-minute candle closes past the high or low, stop the other side, target twice the risk. Best one here.
US 15-minute range on the Nasdaq. First fifteen minutes, orders at the high and low, stop the other side, target one range width.
School Run on the DAX. Tom Hougaard's. The first two 15-minute candles of the German open give you a 30-minute range, then you trade the break. The idea is the first half hour is people shuffling positions and what happens after it tells you where the morning's going. The exits are mine, not his he doesn't advise any specific exit, so I've used the far side of the range with some ATR breathing room and flat at the end of the day. Wanted to say that, so it's definetly my fault if it doesn't work as you'd like it to.
DAX first 5 minutes. Mine. Bracket the opening 5-minute candle, stop the other side, no target, hold till the close.
The numbers, and what's wrong with them
2021 to August 2026, real tick data on Darwinex, 0.25% risk on a 10k account, each one tested on its own.
US 30-minute range: 1,447 trades, profit factor 1.15, profitable in 5 years out of 6. US 15-minute range: 1,442 trades, 1.07, 5 out of 6. School Run: 1,372 trades, 1.07, 3 out of 6. DAX first 5 minutes: 1,425 trades, 1.02, 5 out of 6.
All four together: 1.07, worst drawdown 21.6%.
Here's the bit nobody else puts in a listing. Take away each strategy's best year and see what's left. Both US ones survive it. Neither German one does. School Run made more in 2022 than it made in total, and the DAX 5-minute lost 3,120 in 2023 on its own.
A profit factor of 1.07 means you make 107 for every 100 you lose. That's thin. It's real in this data, on this broker, over this stretch of time. It's not a wage.
Both German strategies had their bad year at the same time, in 2023. Run only those two and the rough patches arrive together. The US ones were fine that year, which is the whole argument for running more than one thing.
And there's no out-of-sample period. One broker, one five-and-a-half year window. I'd have liked a proper holdout. Haven't got one.
What didn't make it
Nineteen strategies built, four shipped. The other fifteen are in the manual with their profit factors and why they died. London breakout of the Asian range lost money in all six years across 1,407 trades. Tokyo session was break-even before costs so there was nothing to save. The 5-minute Nasdaq range gave away 3% of gross in commission. Advanced School Run I tried four different ways and the best was 0.94.
I'm listing the failures because a results table with only winners in it tells you nothing. The reasons are the useful part.
The pieces all stay in the builder though. A preset losing money doesn't make fade entries or Donchian channels useless, it means my particular combination was.
Bits that took the longest
It won't over-risk you. If your account's too small for the minimum lot to fit inside the risk you set, it skips the trade and says why. Most EAs quietly risk ten times what you asked and you find out from the equity curve.
It tells you why it isn't trading. Not just "waiting" — armed, done for today, news blackout, spread too wide, range too narrow, minimum lot over budget. Saves you staring at a chart wondering if it's broken.
The clock sorts itself out. Preset times are stored in the exchange's own time and converted to your broker's server clock, including the weeks where Europe and America change clocks on different dates. Doesn't matter where you are. Nothing to set.
And it puts your chart colours back when you remove it.
Limitations
The presets are index strategies — DAX, S&P, Nasdaq. The builder works on anything.
It's intraday and swing, not scalping. If you want fifty trades a day this isn't it.
Backtest two strategies on the same symbol at once and they'll block each other's entries. Test one at a time for a clean read.
Broker symbol names vary. It handles the usual ones automatically but check the journal on the first run.
The news filter doesn't work in the Strategy Tester — MetaTrader doesn't hand the calendar to backtests. It says so in the log rather than pretending.
The panel doesn't draw in a non-visual backtest. That's MetaTrader, not me.
If you find something
Send me the settings and the backtest. Real ticks, couple of years minimum, and tell me what you tried that didn't work as well. That part matters as much as the winner.
If it ends up in a paid product we sort the profit share first, in writing.
If you just use it and never say a word, that's fine too.
Test it on demo first. Your broker isn't my broker — different spreads, different fills, different clock. The numbers above are what happened on my data. Yours will be different and might be worse.
Manual's included and it's a proper one.
MLB Honest Trader
