The Expert Advisors I build

All trading systems hold a single position at a time. No grid and no martingale.

Feel free to drop me a message about any of them. I’m very happy to explain them or answer any questions you may have. After testing and trading hundreds of expert advisors from other developers over the years I’ve learnt that talking to the developer both builds trust and gives you the best chance of success when you’re using it.

Get in touch

I'm building this portfolio in the open. Every version that didn't work gets posted alongside the ones that did. If you want to follow it as it happens, I'm on X: @HighTideTrading

Expert Advisors

Expert Advisors

Filters

High Tide Ultimate Bundle — All EAs and Tools

Everything, built to run together

Every EA and every tool, plus the set files that run the four together on one account. Bought separately it comes to £945.

The four trade different markets for different reasons, so their monthly results barely move together. Correlations run from −0.1 to +0.3. Low Tide also refuses to open against a live High Tide position, which cuts out the days where both would have lost at once.

The four engines

High Tide v3.20: BTC and ETH trend breakout, volatility filter on [0.15% Risk]
Low Tide v4.09: Bitcoin intraday breakout, weekdays, never overnight [0.75% Risk]
Gold Breakout v2.04: Gold swing-level breakout, C1 and C [20.5% Risk]
Rip Tide v1.07: Index rebound after a stressed close [0.225% Risk]

What the combined backtest shows

Combined from the four separate MT5 £5,000 backtests, April 2018 to August 2026, with today's minimum lot sizes applied.

Growth: £5,000 → £131,000
Return: 47.4% a year
Max drawdown: 9.8% closed trades · around 11.5% equity
Sharpe ratio: 2.70
Months positive: 77% · worst month −6.3%
Worst quarter: +0.3% · no losing quarter
Longest spell below a high: 170 days

Hypothetical results combined from four separate MetaTrader 5 Strategy Tester backtests on historical data, not from one live account. Not live trading. Past performance, simulated or real, is not a reliable indicator of future results.

Please read this before those numbers go to your head

47.4% a year is an upper bound. It comes from stitching four separate backtests together on perfect historical data with today's minimum lot sizes. No live account produced it.

For planning, assume something closer to 25% to 35% a year with drawdowns of up to 15% or 16% on a £5,000 account. On a larger account the position sizing lands more precisely and the drawdowns come down.

I would rather tell you that here than have you find out in month three. The whole reason I built these was that I spent years losing money to grid and martingale systems sold on numbers exactly like the ones above.

What to expect

Quieter years happen. 2021 came in at +25%. 2022 at +17%. 2026 to August at +23%. There was no losing quarter in the backtest, though there was a 170-day stretch below a previous high.

Every trade carries a stop loss and a take profit. No grid, no martingale, no averaging down. That is true of all four.

What you need

  • MetaTrader 5 on a hedging account. Server clock GMT+2 or GMT+3.

  • Four charts: BTCUSD H1 for High Tide, BTCUSD H1 for Low Tide, XAUUSD daily for Gold Breakout, US500 H1 for Rip Tide.

  • From £5,000. Below that the broker minimum lot starts to distort the risk sizing.

  • A VPS, so all four keep running while your machine is off.

What you get

  • All four EAs: High Tide, Low Tide, Rip Tide, Gold Breakout (£896 on their own)

  • The Copier Bundle, sender and receiver, for copying between terminals (£49 on its own)

  • Regime Radar and Health Monitor, both free tools, included here too

  • Portfolio set files for running the four together, plus standalone set files for each one

  • All five summary sheets and the full manual

  • Free updates, plus email support from me

Trading carries risk and you can lose money. Test on a demo account first. Only trade with money you can afford to lose.

£699.00 £945.00
Sale
Gold Breakout EA

Swing-level breakouts on gold

Gold Breakout watches swing highs and lows on the gold daily chart, then rests an order just inside the most recent level that has not broken yet. Each level gets traded once. Every distance scales with the daily ATR, so the EA adapts on its own as gold gets calmer or wilder.

How it trades

  • Tracks two sets of swing points at once: a 20/3 set and a 10/2 set.

  • Rests a buy stop just inside the most recent unbroken swing high, a sell stop just inside the most recent unbroken swing low.

  • Stop loss 0.5 × daily ATR. Take profit 3 × the stop.

  • Once a trade reaches +0.5R the stop moves up to lock in +0.1R.

  • When both swing sets break at the same level they each take their own trade. These confluence breakouts have been the strongest in testing.

  • Closes positions and pulls resting orders around US Non-Farm Payrolls, from 100 minutes before to 60 minutes after.

  • Retries on broker errors. New orders pause if median slippage goes past $0.40 an ounce.

  • No grid. No martingale. A stop loss and a take profit on every trade.

What the backtest shows

MT5 Strategy Tester on IC Markets data, 1 April 2018 to 30 September 2026, £5,000 start, 1.0% risk per trade running on its own, C1 and C2 sets, 99% history quality. Commission and swap included.

Net profit: +£14,353 (£5,000 → £19,353)
Return: 17.6% a year
Max drawdown: 12.1% equity · 10.3% balance
Profit factor: 2.89
Trades: 383 · around one a week
Win rate: 72.1%
Positive years: 9 out of 9

Hypothetical results produced by the MetaTrader 5 Strategy Tester on historical data. Not live trading. Past performance, simulated or real, is not a reliable indicator of future results.

What to expect

About one trade a week, held around 16 hours on average. Seven in ten finish in profit, though most of those winners are small break-even locks rather than full targets. The regular 3R winners are where the money comes from.

Every calendar year was positive in testing, from +3% in 2022 to +37% in 2025. The worst month was −3.9% and 63% of months were positive.

Flat spells run long. The longest gap between new equity highs was about 16 months across 2024 and 2025. That is the part people find hardest.

Longs have done better than shorts through gold's bull run, with a profit factor of 4.22 against 1.44. Both sides make money.

At 1% risk the backtest drawdown was 12% on equity. Plan for a worst case nearer 15%.

What you need

  • MetaTrader 5, attached to an XAUUSD daily chart.

  • A hedging or netting account. Server clock GMT+2 or GMT+3.

  • A gold spread you can live with. Slippage protection is built in, though a decent broker still helps.

  • A VPS if you want it running unattended.

What you get

  • HighTide_GoldBreakout_v2.04.ex5

  • Two set files: one for running it on its own, one for the portfolio risk setting

  • The summary sheet and the full manual

  • Free updates, plus email support from me

Trading carries risk and you can lose money. Test on a demo account first. Only trade with money you can afford to lose.

£249.00
High Tide EA

Trend breakout for Bitcoin and Ethereum

High Tide trades the big moves in Bitcoin and Ethereum on the hourly chart. When price closes above its highest level of the last four or eight days it buys. When it closes below the lowest it sells short. Then it holds on, trailing a stop behind the best close, so a strong trend can run for weeks.

That is the whole idea. Trend following loses small most of the time while it waits for the few moves that pay for everything else.

How it trades

  • Buys hourly closes above the 96-bar or 192-bar high. Sells short below the low. Two independent systems per coin.

  • Initial stop 4 × ATR, target 20 × ATR. The stop then trails behind the best close.

  • One position per system, six open at most. Risk is a fixed percentage of equity measured to the initial stop.

  • A volatility filter, new in v3.20, stands aside while Bitcoin's daily volatility sits in the top 20% of the past year. Blow-off tops and crashes are where breakouts whip back and forth. Open trades carry on being managed as normal.

  • No grid. No martingale. No averaging down. A stop loss and a take profit on every trade.

What the backtest shows

MT5 Strategy Tester on IC Markets data, 1 April 2018 to 30 September 2026, £5,000 start, 0.30% risk per trade, volatility filter on.

Net profit: +£9,968 (£5,000 → £14,968)
Return: 13.8% a year
Max drawdown: 11.0% equity · 10.1% balance
Profit factor: 1.49
Trades: 1,774 · around four a week
Win rate: 37%
Positive years: 9 out of 9

Hypothetical results produced by the MetaTrader 5 Strategy Tester on historical data. Not live trading. Past performance, simulated or real, is not a reliable indicator of future results.

What to expect

Six trades in ten lose. The average win is about 2.7 times the average loss, so a handful of big trends each year carry the result. 64% of months were positive in testing. The worst rolling twelve months came in at −0.4%.

Every calendar year was positive in the test, from +3% in 2021 to +37% in 2020. Both coins make money in both directions.

It does well when crypto trends. It struggles through long sideways markets. You will sit through flat spells, so size it as something you can leave alone.

What you need

  • MetaTrader 5, attached to a BTCUSD H1 chart.

  • A broker carrying BTCUSD and ETHUSD, on a hedging or netting account.

  • Server clock GMT+2 or GMT+3, which is the New York close.

  • A low-spread, low-commission account. Costs come straight out of the result.

  • A VPS if you want it running while your own machine is off.

What you get

  • HighTide_v3.20.ex5

  • Two set files: one for running it on its own, one for the portfolio risk setting

  • The summary sheet and the full manual

  • Free updates, plus email support from me

Trading carries risk and you can lose money. Test on a demo account first. Only trade with money you can afford to lose.

£249.00
Low Tide EA

Bitcoin day trading that goes home flat

Low Tide day-trades Bitcoin from Monday to Friday. At 01:00 server time it rests a buy stop above the day's opening price and a sell stop below it, each one 0.4 × yesterday's range away. Whichever side breaks first is the trade. Everything is closed out by 23:00, so it never pays overnight swap.

It is a simple idea run strictly. One position at a time, flat every night.

How it trades

  • A buy stop and a sell stop placed at 01:00, each 0.4 × yesterday's range either side of the open.

  • Stop loss 1 × yesterday's range. Take profit 2 ×.

  • At most one long and one short a day. One position open at a time.

  • Flat by 23:00 every day, so no overnight swap is ever paid.

  • No weekend trades, new in v4.09. Weekend breakouts lost money in every year they traded.

  • No grid. No martingale. A stop loss and a take profit on every order.

What the backtest shows

MT5 Strategy Tester on IC Markets data, 1 April 2018 to 30 September 2026, £5,000 start, 1.0% risk per trade, weekends skipped, 87% history quality, commission included.

Net profit: +£20,538 (£5,000 → £25,538)
Return: 21.2% a year
Max drawdown: 10.8% equity · 10.2% balance
Profit factor: 1.33
Trades: 2,087 · four or five a week
Win rate: 50.9%
Positive years: 9 out of 9
At the portfolio setting: 0.75% risk · 15.2% a year · 7.4% drawdown

Hypothetical results produced by the MetaTrader 5 Strategy Tester on historical data. Not live trading. Past performance, simulated or real, is not a reliable indicator of future results.

What to expect

About half the trades win. The average win is 1.3 times the average loss. Plenty of trades close at 23:00 rather than at the stop or the target, which is the system doing its job.

69% of months were positive in testing. The worst month was −7.1%. Every calendar year was positive, from +6% in 2025 to +46% in 2024. The longest losing run was 11 trades.

Because it enters with resting stop orders it fills at the breakout level. Costs still bite at this trade frequency, so a raw-spread account makes a real difference to the result.

What you need

  • MetaTrader 5, attached to a BTCUSD H1 chart.

  • A hedging or netting account. Server clock GMT+2 or GMT+3.

  • A raw-spread, low-commission account. This one is cost-sensitive.

  • A VPS is strongly advised, since the entries are time-based.

What you get

  • LowTide_v4.09.ex5

  • Two set files: one for running it on its own, one for the portfolio risk setting

  • The summary sheet and the full manual

  • Free updates, plus email support from me

Trading carries risk and you can lose money. Test on a demo account first. Only trade with money you can afford to lose.

£249.00
Rip Tide EA

Buys indices under pressure, sells before the close

Rip Tide buys stock indices on bad days. It only looks at an index once it has closed below its 200-day average. Then it buys at 03:00 server time and closes by 22:58 the same day, catching the intraday rebounds that stressed markets tend to throw up.

I will be straight with you about this one. On its own the return is modest. That is not what it is for. Its job is to earn in the sell-offs where trend and breakout systems have their worst months, which is why it sits in the portfolio alongside the other three.

How it trades

  • Long only. It will not touch an index until it has closed below its 200-day average.

  • Buys at 03:00 server time, latest 12:00. Closes by 22:58. No overnight swap.

  • Stop loss 1.5 × daily ATR, take profit 3 × ATR.

  • One trade per index per day across US500, US30, DE40 and UK100. USTEC is added automatically from £15,000.

  • Skips US market holidays and early closes.

  • Skips a trade altogether if the broker's minimum lot would risk more than 1%.

  • No grid. No martingale. A stop loss and a take profit on every trade.

What the backtest shows

MT5 Strategy Tester on IC Markets data, 1 April 2018 to 31 August 2026, £5,000 start, 0.30% risk per trade with 0.75% on US500, 99% history quality.

Net profit: +£1,497 (£5,000 → £6,497)
Return: 3.2% a year
Max drawdown: 8.2% equity · 7.6% balance
Profit factor: 1.21
Trades: 1,591 · long only
Win rate: 53.7%
Best year: +17% in 2019, after the late-2018 sell-off
Correlation with the others: −0.1

Hypothetical results produced by the MetaTrader 5 Strategy Tester on historical data. Not live trading. Past performance, simulated or real, is not a reliable indicator of future results.

What to expect

It can go quiet for months while markets climb steadily. It earns most when they fall over. Just over half the trades win, with the average win and the average loss about equal.

By index across the test: US500 +£498, US30 +£483, DE40 +£389, UK100 +£128.

Buy this one as a diversifier rather than as a standalone earner. If you only want one EA, Rip Tide is not the one.

What you need

  • MetaTrader 5, attached to one US500 H1 chart. It trades all the indices itself.

  • A hedging or netting account. Symbol names can be changed to match your broker.

  • Server clock GMT+2 or GMT+3.

  • Works best as part of a portfolio from £5,000 upwards.

What you get

  • RipTide_v1.07.ex5

  • Two set files: one for running it on its own, one for the portfolio risk setting

  • The summary sheet and the full manual

  • Free updates, plus email support from me

Trading carries risk and you can lose money. Test on a demo account first. Only trade with money you can afford to lose.

£149.00

No results found

No results match your search. Try removing a few filters.

I'm building this portfolio in the open on X. If you want to follow it as it happens: @HighTideTrading

What my EA’s are designed for

Because a system built to do one thing in one kind of market is easier to understand, easier to test, and easier to switch off when its conditions disappear. Trend systems struggle in ranges. Reversion systems struggle in trends. An EA claiming to handle everything usually handles nothing especially well. So my EA’s are designed to work as part of a portfolio where you can run multiple EA’s on the same account. They have a low correlation in an attempt to smoothen the equity curve and reduce potential drawdowns.

Running an EA reliably

Any EA that trades outside the hours you're at your desk needs to keep running when your computer doesn't. Here's what to look for in a VPS, and the one I use: VPS.

Gold

I've written about why automated gold trading is a different problem from FX. That article is here, the Gold Reaper page covers the system I use.

Other developers' EAs

The ones I run alongside my own are on EAs I use, with a full write-up for each.

MetaTrader mobile app showing an open NZDUSD position with stop loss and take profit levels

Software only. Automated trading carries risk, including losing more than your deposit on leveraged products. Nothing here is financial advice or a forecast of returns. High Tide Trading is not authorised or regulated by the FCA

30-Day Money-Back Guarantee

All products include a no-questions-asked 30-day refund policy.
If the EA does not meet your expectations, simply request a refund within 30 days of purchase for a full return.

Feedback and suggestions are always welcome. I am aways looking to improve them so feel free to get in touch.