Indicator Combinations Tested: Ichimoku Cross + Cloud, Stochastic + 200 MA, ADX + Supertrend on 16,926 Bars (2026)

An indicator combination is supposed to add information. Most add a second copy of the same information and remove trades. Below are the pairings I measured on EURUSD M5, June 3 to August 26, 2026, with the same hold and rules for each, sorted by what the second indicator actually did to the first. A pairing that is not here has not been tested; none of these numbers are estimates.

Which combination improved the base signal?

Two did, and both pair an entry with a filter on market state rather than with another entry.

The Ichimoku case is the cleanest: the cross alone was the second worst rule I have measured, at the 8.7th percentile of random entries, and the cloud filter took it to the 96.4th percentile. Nothing else changed. The breakdown by hold and setting is in Ichimoku vs trend following and the cloud breakout test. The Stochastic case is in the Stochastic overbought test, where the oscillator itself never cleared 50% at 30 minutes and the filter carried the whole edge.

Which combination made things worse?

Two lessons. First, stacking two filters produced the worst row on the site, because ADX above 25 already threw out four signals in five and the Supertrend condition then removed the few good ones left; ADX vs Supertrend has the full comparison and the ADX 25 test shows that 20 was the only ADX threshold that helped, at +0.11. Second, “confirming” an Ichimoku cloud breakout with the Tenkan/Kijun cross turned +0.29 pips into -0.12, because both conditions describe the same event and the second only delays entry. The same delay is why an EMA 200 filter on a fast crossover cut the win rate from 44.6% to 41.2% in the moving average crossover test.

Does confirming one oscillator with another help?

It changes the trade count and nothing else. Across 1,997 signals run under identical rules, Stochastic and RSI finished within half a point of each other on win rate and agreed on the same bar only 149 times; the rest of the time one was silent while the other traded. An arrow that requires both to agree fired 372 times and scored 43.8% and -0.30 pips, indistinguishable from either alone. The numbers are in Stochastic vs RSI. MACD is the same family again: a 2-line MACD is a fast EMA pair drawn differently, so pairing it with a fast crossover is one signal counted twice. If you want a second opinion on a momentum entry, take it from a trend gate, not from a second oscillator.

What about bands plus a channel?

Bollinger and Keltner squeezes measure the same compression with different maths. On the same bars the Keltner version continued 42.7% of the time against 39.3% for Bollinger, with far fewer signals, and neither crossed 50%. Running both and trading only when both squeeze is a valid way to cut noise, but the combination still lost before spread on this pair and timeframe; session filtering did more than either band, per Bollinger vs Keltner.

What is the working recipe?

One entry, one filter on market state, and the filter should be slow. Cloud or EMA 200 for state, one oscillator or one cross for timing, a fixed stop measured on your data. Supertrend, per the Supertrend for MT4 guide, works as the state filter on its own but not on top of ADX. For the entry line, my DeMARK trend line is the one I combine with an EMA 200, because it is drawn from price pivots rather than from a smoothed average and does not move after the candle closes, so the two are not the same information twice.

FAQ

What is the best indicator combination for forex?

On this data, Ichimoku cross plus cloud filter at +1.10 pips per trade over 281 trades, then Stochastic plus a 200-period moving average at +0.92. Both pair a timing signal with a slow state filter.

Can I combine RSI and Stochastic?

You can, but they agreed on only 149 of 1,997 signals and the combined arrow scored 43.8%. They are the same information with different smoothing, not two opinions.

Is ADX plus Supertrend a good trend combination?

It was the worst result measured: 38.1% and -1.48 pips. Supertrend alone helped; ADX above 25 hurt; both together kept 63 of 324 trades and the wrong ones.

What data are these from?

EURUSD M5, 16,926 bars, June 3 to August 26, 2026, from a live MetaTrader terminal, two-hour hold unless stated, no stop or target, gross of spread.

{“@context”: “https://schema.org”, “@type”: “FAQPage”, “mainEntity”: [{“@type”: “Question”, “name”: “What is the best indicator combination for forex?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “On this data, Ichimoku cross plus cloud filter at +1.10 pips per trade over 281 trades, then Stochastic plus a 200-period moving average at +0.92. Both pair a timing signal with a slow state filter.”}}, {“@type”: “Question”, “name”: “Can I combine RSI and Stochastic?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “You can, but they agreed on only 149 of 1,997 signals and the combined arrow scored 43.8%. They are the same information with different smoothing, not two opinions.”}}, {“@type”: “Question”, “name”: “Is ADX plus Supertrend a good trend combination?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “It was the worst result measured: 38.1% and -1.48 pips. Supertrend alone helped; ADX above 25 hurt; both together kept 63 of 324 trades and the wrong ones.”}}, {“@type”: “Question”, “name”: “What data are these from?”, “acceptedAnswer”: {“@type”: “Answer”, “text”: “EURUSD M5, 16,926 bars, June 3 to August 26, 2026, from a live MetaTrader terminal, two-hour hold unless stated, no stop or target, gross of spread.”}}]}