Gold indicator settings, tested on real data
Twelve popular indicators, the settings gold traders actually use, and what happened when we tested them on two years of hourly XAUUSD data and daily data since 2015, with costs on every trade. Small edges and losses are shown as they are.
Results at a glance
Each row shows the best or most telling hourly result for that indicator (net dollars per ounce over two years, after costs) and the one-line finding. Open any page for the full table, the daily results since 2015, the split by year and how to use the indicator by timeframe.
| Indicator | Key hourly result | What we found | Verdict |
|---|---|---|---|
| RSI Momentum oscillator | RSI 14, 20/80: −$163 | All five standard settings lost money as a buy-oversold, sell-overbought rule on H1 after costs; a 200 EMA filter only brought it to break-even. | Lost |
| Stochastic Momentum oscillator | 21,5,5: −$585 | Every hourly setting lost in both years, the default 14,3,3 most of all; one daily setting (8,3,3) was positive, which may be chance. | Lost |
| CCI Momentum oscillator | CCI 20 momentum: +$2,611 | Read as momentum (buy above +100) every setting was positive in both years; read as overbought/oversold, the same CCI lost. | Depends on use |
| EMA Trend | 9/21 cross: +$2,287 | All six EMA setups were positive on H1; the 9/21 cross was the most even across both years, but win rates were only 16% to 35%. | Positive |
| Ichimoku Trend | 9/26/52: +$2,148 | All four settings positive on H1 and D1 with some of the smallest drawdowns; the cloud filter kept it out of many choppy trades. | Positive |
| ADX Trend strength | ADX 7 > 25: +$1,641 | The classic ADX 14 > 25 filter turned a profitable DI-cross rule into a loss; only the fast ADX 7 filter helped. | Mixed |
| Parabolic SAR Trend / trailing stop | 0.01/0.1: +$4,762 | Every setting positive on H1 in both years, but the slow settings that won on H1 lost heavily on daily charts. | Mixed |
| Supertrend Trend / trailing stop | 20,4: +$665 | Small hourly profits in year 1 turned into losses in year 2 for all five settings; on daily charts a multiplier of 3 or 4 beat 2. | Weak |
| Heikin Ashi Smoothed candles | 2 candles + 200 EMA: +$1,872 | Thin edges (profit factor 1.05 to 1.23) that costs can erase; waiting for two candles and a trend filter helped most. | Thin |
| Keltner Channels Volatility breakout | 20 EMA, 1.5 ATR: +$1,582 | All four breakout settings positive on H1; the tighter 1.5 ATR band made more than three times as much as the common 2 ATR default. | Positive |
| Donchian Channels Breakout | 20/10: +$2,588 | Every hourly setting positive in both years with low drawdowns, but every daily setting lost from 2015 to 2020. | Positive |
| OBV (volume) Volume | OBV vs 20 EMA: +$3,731 | Positive on COMEX futures volume, but a control with volume switched off also made money; spot XAUUSD has no central volume. | Mixed |
For reference, simply holding gold gained about $1,194 an ounce in the first test year and about $14 in the second, which included a run to about $5,627 in January 2026 and a fall back to around $4,000.
Pick an indicator
Every page has the settings traders use, the test table, a plain-English reading, and how to use the indicator for scalping, intraday and swing trading.
RSI
Best RSI Settings for Gold (XAUUSD): Tested on 2 Years of Data
Read the testStochastic
Best Stochastic Settings for Gold: 5,3,3 vs 14,3,3 Tested on XAUUSD
Read the testCCI
CCI Indicator for Gold: Momentum Beats Overbought/Oversold on XAUUSD
Read the testEMA
Best EMA for XAUUSD: 9/21, 20/50 and 200 EMA Tested on Gold
Read the testIchimoku
Ichimoku Cloud on Gold: Which Settings Work on XAUUSD? A 2-Year Test
Read the testADX
ADX Indicator on Gold: Does the ADX 25 Filter Work on XAUUSD?
Read the testParabolic SAR
Parabolic SAR for Gold Trading: 4 Settings Backtested on XAUUSD
Read the testSupertrend
Supertrend on Gold: Best Settings for XAUUSD and Why They Faded
Read the testHeikin Ashi
Heikin Ashi on Gold: Does It Work on XAUUSD? Tested on 2 Years of Data
Read the testKeltner Channels
Keltner Channels on Gold: Best Settings Tested on XAUUSD Breakouts
Read the testDonchian Channels
Donchian Channels on Gold: Turtle-Style Breakouts Tested on XAUUSD
Read the testOBV (volume)
OBV and Volume on Gold: Does On-Balance Volume Work on XAUUSD?
Read the testWhat the tests taught us about gold
How you read an indicator mattered more than its settings. The clearest example is CCI: the same CCI 20 made money when a move above +100 was treated as the start of momentum and lost money when it was treated as overbought. RSI and the stochastic, used as overbought and oversold signals, lost at every setting we tried.
Trend and breakout rules did well, in a trending period. Donchian channels, Ichimoku, EMA crosses, Keltner breakouts and Parabolic SAR were positive on the hourly chart. But gold's average daily range on COMEX futures went from about $34 in 2024 to about $64 in 2025 and around $123 so far in 2026, with powerful trends in both directions. That is close to ideal for trend-following. On daily data, most of the same rules lost money from 2015 to 2020, when gold was quieter and often range-bound.
Popular filters did not always help. The classic "ADX above 25" filter made results worse. A 200 EMA direction filter helped Heikin Ashi and RSI a little and did nothing for the stochastic on the hourly chart.
Edges were small and win rates low. Most positive rules won only 25% to 45% of their trades and had drawdowns of $400 to $1,100 an ounce. Several, such as Heikin Ashi and the fast Parabolic SAR, traded so often that doubling the cost would remove a large part of the profit.
Limitations you should know
- One market, one period. Two years of hourly data is around 11,000 bars but only one market regime. Daily data since 2015 adds a quieter period, and many results changed sign there.
- Futures, not spot. The continuous futures series includes contract rolls, where the price can jump by tens of dollars as the feed switches contract. We left the data unedited; it adds some noise to every rule.
- Simple rules. Each indicator got one plain rule with a fixed exit. Real traders add discretion, structure and session filters, which can make results better or worse.
- Costs. We used $0.30 per ounce per round trip. Your spread, commission and slippage may be higher, especially on news and at the Sunday open.
- Small samples. Some daily rows have fewer than 30 trades. We point these out on each page.
Related lessons
The lessons below teach indicator-based strategies in depth. The pages above are about settings and tested behaviour, and link back to them.
- RSI divergence on gold
- MACD on gold
- Moving averages on gold
- ATR and Bollinger Bands on gold
- VWAP on gold
- Best indicators for gold trading
- How to backtest a gold strategy
Looking for complete named strategies rather than settings? See 14 gold trading strategies tested on real data, from 1-minute scalps to inside bars, with the same method.
Size every test trade properly with the position size calculator, and check what a move is worth with the gold pip calculator. Today's volatility is on the gold volatility tool.
Test these settings with other traders
Our numbers are a starting point. The useful part is testing a setting on your own data, in your own sessions, and having other traders challenge the results. Inside the XAUUSD Global community traders do exactly that every day. The first month is free, then $19.99 a month.
Frequently asked questions
What are the best indicator settings for gold?
No single setting won everywhere. On two years of hourly gold data, trend and breakout rules such as Donchian 20/10, Ichimoku 9/26/52, the 9/21 EMA cross and CCI used as momentum were positive in both years after costs, while RSI and stochastic used as overbought/oversold reversal signals lost money at every setting we tried. How you use an indicator mattered more than its numbers.
How were the indicators tested?
On COMEX gold futures (GC=F) 1-hour bars from 30 October 2024 to 30 October 2026, and daily bars from January 2015. Each indicator got one simple, standard rule, signals were read on the bar close and filled at the next open, and every trade was 1 ounce and paid $0.30 per ounce for spread and slippage. We report trades, win rate, average result, profit factor, net result, maximum drawdown and the split between periods.
Why did trend indicators do better than oscillators?
Gold trended unusually strongly in the test period, rising about $1,194 an ounce in the first year and swinging between roughly $3,900 and $5,600 in the second. Trend and breakout rules ride those moves; reversal rules keep betting against them. In a long sideways market the ranking could flip, and on daily data from 2015 to 2020 many trend rules lost.
Can I trade these results?
No. They are simple mechanical rules tested on one data feed and one historical period, shown for education. They are not signals or recommendations. Use them as a starting point for your own testing, with your broker’s prices and costs, in a demo account first.
Do the results apply to my broker's XAUUSD chart?
Approximately. Futures prices sit a few dollars away from spot and include contract rolls, broker spreads differ, and MT4 or MT5 volume is tick volume rather than exchange volume. Price-based indicators should behave similarly on spot charts, but re-run any rule on your own data before relying on it.