Free MQL5 Indicators vs MetaTrader Built-Ins: QQE Matched MACD on 86% of Bars, 0 of 16 Free Runs Passed and 2 of 16 Built-In Runs Did

MetaTrader 5 comes with 38 technical indicators in the Navigator. The MQL5 Code Base adds thousands more for free. The reason to download one is the idea that it sees something the standard set does not.

That can be measured two ways. First, how often does the download point the same way as something already installed? Second, when both are traded on the same bars with the same stop, which one keeps more money? I counted both.

Which indicators were compared?

All eight at the default inputs. The table is two lists side by side, not four pairs: the pairing comes from the data further down. The four downloads and their individual results are in free MQL5 indicators tested. The method is the same here:

• Data: EURUSD, June 3 to August 26, 2026, 16,926 M5 bars and 5,642 M15 bars; gold, June 16 to August 26, 13,736 M5 bars and 4,577 M15 bars.
• Exit: stop 1.5 ATR, target 3 ATR, 48 bars at most, one trade at a time.
• Cost: a standard 1 pip on EURUSD and $0.30 an ounce on gold, charged on every trade. What a 1-pip spread does to a 5-minute chart is in my spread and swap measurement.
• Three checks: profit after cost, profit in both halves (split on July 15), and a result above the best 5% of 200 runs of random entries with the same exit.
• Noise test for a gap: I resampled the trades of both sides 2,000 times. A gap counts only if 90% of the resampled gaps sit on one side of zero.

How often does a free indicator agree with a built-in one?

Every one of these indicators, except RSI, is always in one of two states: up or down. I compared the states bar by bar.

Two indicators with nothing in common would agree on 50% of bars. For Keltner the state is the side of the last close outside the channel.

QQE is the clearest case. It is a smoothed RSI with a trailing line, and it has its own window, its own colours and its own name. On 86% to 88% of bars it said the same thing as the MACD line against its signal line. The spread between the four charts was under two points, so this is how the formulas relate, not an accident of one market.

SuperTrend against the EMA 9/21 cross was almost as tight, and just as stable: 83.4% to 84.2%. If you already have two moving averages on the chart, a SuperTrend line tells you something new on about one bar in six.

Among the four downloads, the pair that agreed least was Keltner and QQE, at 43% to 53%. A channel breakout next to an RSI-based line is the one combination here that behaves like two separate opinions.

What did the built-in indicators make?

Per trade after cost: pips on EURUSD, dollars an ounce on gold.

And the same table for the downloads:

The scoreboard for the two groups:

The built-ins come out slightly ahead on every line but the last. I would not read much into the two passes. Both are the EMA 9/21 cross. On EURUSD M15 it made +0.33 pips a trade with +0.03 in the second half, and the random line it had to beat was +0.01. On gold M5 it was sturdier: +$1.16 a trade over 366 trades, +0.66 and +1.45 in the two halves. The same cross on EURUSD M5 was the worst of all eight indicators at -1.68 pips, which is the result I got in moving average scalping on M5 as well. Out of 32 runs, one or two will clear a 5% bar by luck.

The last line matters more. On EURUSD M5, six of the eight indicators did worse than the worst 5% of coin-flip entries: three downloads and three built-ins. Being free or built-in made no difference to that. Being a trend-confirmation signal on a 5-minute EURUSD chart did.

Does the download beat its closest built-in?

Each download against the built-in it agrees with most. The number is the download’s result minus the built-in’s, per trade after cost.

Bold marks a gap bigger than noise.

The download was ahead in 7 of 16. On EURUSD M5 it was ahead in all four, by 0.10 to 0.19 pips, which is under a fifth of the spread and well inside noise. On gold M5 it was behind in all four.

Two gaps of the 16 were bigger than noise, and both say the same thing: entering on a close outside the Keltner Channel was worse than entering on the EMA cross, by about 2 pips on EURUSD M15 and $2 on gold M5. The Keltner signal is the latest of the eight. On EURUSD M5 it printed a median 5.85 ATR from the recent extreme, against 3.63 for the EMA cross.

QQE against MACD is the match-up I would point to. Two indicators that agree on 86% of bars produced results within 0.19 pips of each other on EURUSD, and neither gap on gold was outside noise. I tested the MACD cross on its own in MACD crossover rules backtested. If you have read that, you already know roughly what QQE will do.

What if you take the best of each group?

The best rule on the chart was a built-in on three charts and SuperTrend on one. Picking the winner after the fact flatters both sides, so this table is the most generous view either group gets. Even here, only one gap was outside noise, and it favoured the plain RSI.

Pooling all trades of each group gives the same answer. The free group was behind by 0.13 pips on EURUSD M5, 0.43 on EURUSD M15 and $0.33 on gold M5, and ahead by $0.47 on gold M15. None of the four gaps was bigger than noise.

So should you download or use what is installed?

• Do not expect a download to see what the built-ins miss. The four I tested agreed with a built-in on 77% to 87% of bars, and no download beat its closest built-in by more than noise on any chart.
• Do not stack a download on top of its twin. QQE under a chart that already has MACD, or SuperTrend over two EMAs, is the same vote counted twice. What stacking really does to results is in indicator combo vs a single indicator.
• A download is still worth it for the display. SuperTrend gives a trailing level you can read as a stop. Two EMAs do not. That is a real difference, even if the entry is not.
• If you want a second opinion, pick one that disagrees. The only pairs near 50% agreement were the channel breakout next to QQE or MACD.
• The chart moved two results more than any download did. Going from EURUSD M5 to M15 changed the EMA cross by 2.00 pips a trade and SuperTrend by 1.49, both outside noise. The other six indicators did not change by more than noise, and nothing on gold did.

If you do install one, the folder, the steps and a grid of 159 tested settings are in the MQL5 Code Base indicators guide for MT5.

Frequently asked questions

Are free MQL5 indicators better than the built-in MetaTrader indicators?

Not in this test. 0 of 16 runs of the free downloads passed all three checks against 2 of 16 for the built-ins, and no gap between a download and its closest built-in favoured the download by more than noise.

Is QQE the same as MACD?

The formulas are different, but the direction they showed was the same on 86% to 88% of bars. Their results per trade were within 0.19 pips on EURUSD.

Is SuperTrend better than a moving average crossover?

SuperTrend 10/3 pointed the same way as an EMA 9/21 cross on 83% to 84% of bars. It was ahead on 2 of 4 charts and behind on 2, and none of the four gaps was bigger than noise.

Which built-in indicator did best?

The EMA 9/21 cross passed on gold M5 and EURUSD M15, and it was the worst of the eight on EURUSD M5. No indicator was best on more than two charts.

Why did so many indicators lose to random entries on EURUSD M5?

Trend signals print after the move. On a 5-minute EURUSD chart the pullback that follows is often larger than a 1.5 ATR stop, so six of eight indicators lost more than a coin flip with the same stop.