Bollinger Bands and RSI System for MT4: 49 of 54 Settings Made Money on Gold M15, and 0 of 54 on EURUSD M5

In my test of five forex trading systems, one run of 20 passed all three checks: positive after cost, positive in both halves and above the best 5% of random entries. It was the Bollinger Bands and RSI reversion system on the 15-minute gold chart. Below are the exact rules, how to put them on an MT4 or MT5 chart, and the settings that held up. Read the limits too: this is 10 weeks of one metal.

What are the exact rules?

• Chart: XAUUSD, M15.
• Buy: a 15-minute bar closes below the lower Bollinger Band (20, 2) and RSI (14) on that bar is below 30. Enter at that close.
• Sell: a bar closes above the upper band and RSI is above 70. Enter at that close.
• Stop: 1 ATR (14) from the entry price, using the ATR value of the signal bar.
• Exit: when price touches the middle band (the 20-period average). If neither stop nor middle band is hit within 48 bars, 12 hours, close the trade.
• One trade at a time: ignore new signals while a trade is open.

How do I set it up in MT4?

1. Open a gold chart (XAUUSD or GOLD, depending on your broker) and switch to M15.
2. Insert, Indicators, Trend, Bollinger Bands. Period 20, Deviations 2, Shift 0, Apply to Close. The middle line is your exit.
3. Insert, Indicators, Oscillators, Relative Strength Index. Period 14, Apply to Close. On the Levels tab keep 30 and 70.
4. Insert, Indicators, Oscillators, Average True Range. Period 14. The value in the ATR window is your stop distance in dollars.
5. Save it: right-click the chart, Template, Save Template, and name it “BB RSI gold M15”.

In MT5 the menus are the same: Insert, Indicators, Trend for Bollinger Bands, Oscillators for RSI and ATR. My Bollinger Bands on MT5 guide shows the band settings window, and the first MT4 chart setup guide covers templates if you have not used them.

How do I place a trade by hand?

1. Wait for a 15-minute bar to close outside a band. A bar that pokes outside and closes back inside is not a signal.
2. Check RSI on that closed bar: under 30 for a buy below the lower band, over 70 for a sell above the upper band.
3. Read ATR on the same bar. Example: ATR 8.50 on a buy at 4,600.00 puts the stop at 4,591.50.
4. For the exit, either watch the middle band or move a take-profit to the middle band at each new bar. The middle band moves, so a fixed take-profit set at entry is not the same rule.
5. Size the trade so the stop is a loss you accept. One standard lot of gold is 100 ounces, so 0.01 lot is 1 ounce and an 8.50 stop risks $8.50. The position sizing table does this for other stops.

What did the system make in the test?

Gold in dollars per ounce after a $0.30 cost, EURUSD in pips after a 1-pip cost. Drawdown is the largest fall from a high in the running total, per ounce or in pips.

On gold M15 the total was +$192 an ounce over 114 trades, about 11 trades a week. A trade lasted 5.4 bars on average, about 80 minutes. Most trades lost: the average winner was $20.02 against an average loser of $9.42. Expect strings of losses. The worst drawdown, $156.70 an ounce, is about 93 trades’ worth of the average result.

The same rules lost on EURUSD M5, where the 1-pip spread was four times the quarter-pip the entry made before cost. Gold M5 was about break-even and changed sign between halves.

Which settings held up on gold M15?

Band periods 15, 20 and 25, widths 2 and 2.5, RSI at 30/70, 25/75 or off, stops of 1, 1.5 and 2 ATR.

Two things stand out. RSI at 30/70 helped: every setting with it was positive. And a wider stop hurt: with 2 ATR, only 2 of 18 settings held in both halves. A tight stop keeps losers small while the middle band takes the reversion.

The single best cell was a 25-period band at width 2.5 with RSI 30/70 and a 1.5 ATR stop: +$5.35 an ounce a trade, but on 66 trades. I would not switch to it. A best cell picked from 54 is the kind of result that fails on the next data, which my optimized vs default settings test shows. The default 20/2 sits in the middle of a field where almost everything worked, which is what you want.

On EURUSD M5 the same 54 settings were all negative after cost. On EURUSD M15, 35 of 54 were positive, but none in both halves.

What are the limits of this result?

• 10 weeks of one market. June 16 to August 26, 2026, while gold rose $323 an ounce. Shorts made money too, which helps, but a long range-bound or crashing period may behave differently.
• 114 trades. Enough to beat random entries here, not enough to promise a number. Treat +$1.69 as a sign of an edge, not as an expected return.
• The exit is the system. The same entry held a flat 12 bars lost $1.48 an ounce a trade, as shown in my trading system vs its own entry signal comparison. Do not swap the middle-band exit for a fixed target.
• Your costs. I used $0.30 an ounce. A spread or commission much above that eats the edge quickly on 80-minute trades.

Before trading it with money, run it on a demo account or in the MT5 Strategy Tester for a few weeks, and compare your fills with the rules above. My Bollinger Bands vs RSI test covers each indicator alone, and gold scalping session by session shows how the hour of day changes gold results.

Frequently asked questions

What are the best Bollinger Bands and RSI settings for gold?

In this test on gold M15, Bollinger Bands 20/2 with RSI 14 at 30/70 and a 1 ATR stop made +$1.69 an ounce a trade after cost on 114 trades, positive in both halves. 49 of 54 nearby settings were also positive.

Does the Bollinger Bands and RSI strategy work on EURUSD?

Not on M5: all 54 settings lost after a 1-pip spread. On M15 the default made +1.00 pip a trade on 112 trades but lost 0.50 a trade in the second half, so I would not trade it there.

Where should the take-profit be?

At the middle Bollinger Band, updated each bar, since the band moves. The same entries held 12 bars without this exit lost money on gold M15.

Where should the stop loss go?

1 to 1.5 ATR (14) from the entry. Every 1 and 1.5 ATR setting was positive on gold M15; with 2 ATR only 13 of 18 were, and 2 held in both halves.

Which timeframe is best for this system?

M15 on gold. On gold M5 the default was about break-even after cost and changed sign between halves.