Choppiness Index Settings for MT4: Period 14 Below 45 Made +$3.16 an Ounce on Gold M15, and 0 of 56 Settings Made Money on M5

The Choppiness Index is the filter most traders reach for when a trend rule keeps getting chopped up. Guides give the same two numbers, 38.2 and 61.8, and rarely say where they came from. This guide shows how to put the Choppiness Index and GRI on an MT4 chart, then what each period and cut-off did to the same breakout rule on 12 weeks of EURUSD and gold.

Whether these filters work at all, compared with ADX, is in Choppiness Index vs ADX vs GRI. Why GRI is closer to a volatility meter is in the Gopalakrishnan Range Index test.

How do you add the Choppiness Index to MT4?

1. Get the file: MT4 does not ship with the Choppiness Index. Free versions are on the MQL5 Code Base; search “Choppiness Index” and download the .mq4 file.
2. Install it: in MT4, File, Open Data Folder, then MQL4, Indicators. Copy the file there, go back to MT4 and right-click Indicators in the Navigator, Refresh.
3. Add it to the chart: drag Choppiness Index from Custom onto the chart. Set the period (14 is the usual default; the test used 10, 14 and 20).
4. Set the levels: on the Levels tab add the cut-off you trade. The test found 45 and 50 more useful than 38.2 on gold M15; keep 61.8 if you also want to see the textbook “choppy” line.
5. Add the rule: a 20-bar breakout needs no indicator, only the last 20 highs and lows; Insert, Indicators, Oscillators, Average True Range 14 for the stop. Save everything as a template.

The rule on the chart: a bar closes above the highest high of the 20 bars before it, and the Choppiness Index on that bar is below your cut-off: buy. A close below the 20-bar low with the same condition: sell. Stop 1.5 ATR, target 3 ATR, close after 48 bars if neither is hit. One trade at a time. The same steps work in MT5 with the MQL5 version of the file.

GRI is also a custom indicator. It is the log of the n-bar high-low range divided by the log of n. Because its raw value depends on price, I used its rank against the last 288 bars. If you cannot find a GRI file, ATR 14 gives you nearly the same reading: the two correlated at 0.81 on three of four charts.

Which settings made money?

Cost 1 pip per EURUSD trade, $0.30 an ounce per gold trade, from my spread measurement. Every setting had at least 45 trades. Data: EURUSD June 3 to August 26, 2026, 16,926 M5 bars; gold over the same weeks.

On M5 the filter cannot fix the rule. The breakout lost 1.39 pips a trade on EURUSD M5 with no filter, and the best of 28 filters only cut that to 0.75. On EURUSD M15, 21 of 28 settings were positive before the spread and 3 after it: on small bars, one pip of cost is a large share of the move, as in the ATR channel vs Bollinger Bands breakout test. Gold M15 is the only chart where the setting changed the answer.

What were the best Choppiness Index settings on gold M15?

Dollars an ounce per trade after cost, trades in brackets. With no filter: +$0.21 over 177 trades.

Every cell from 45 to 55 made money with every period, and all 9 of them were positive in both halves, as was period 14 at 38.2. The textbook 38.2 cut-off was the weakest row: it keeps only the quietest-looking bars and leaves 57 to 88 trades, and two of its three periods lost. Period 14 was the best column at every cut-off.

Random: 300 runs of random entries on gold M15 with the same stop, target and number of trades (200 runs for the no-filter line). Halves split at July 15, 2026.

Two of the four cleared their random line, one of them by $0.08. Read the halves too: every setting made more before July 15 than after, so expect less than the total going forward. And remember the best of 28 is always a little lucky. The rule vs random entry test explains why the random line is the one to beat.

Which GRI period and cut-off worked?

None reliably. On gold M15, 5 of 16 GRI settings made money in total, but all 16 lost in the first half. The best, GRI 30 with rank above 80%, made +$2.76 an ounce over 45 trades: -$3.64 before July 15 and +$5.95 after. On gold M5 the higher the cut-off, the worse GRI did: rank above 50% lost $0.77 to $1.10 a trade, above 80% lost $1.93 to $2.09. Waiting for a wider range meant buying later.

On EURUSD M15 the one setting positive in both halves was GRI 10 with rank above 60%: +0.34 pips a trade over 171 trades, +0.06 and +0.63 in the halves. The +0.06 is too close to zero to build on.

What settings should you start with?

• On gold M15: Choppiness Index 14, take breakouts only below 45. +$3.16 an ounce a trade in this test, positive in both halves. Test it on your own broker’s data first.
• Use 45 or 50, not 38.2. The textbook level kept too few trades and two of its three periods lost on gold M15.
• Skip the filter on M5. 0 of 56 settings were positive after cost on EURUSD M5 and gold M5.
• Use ATR instead of GRI. GRI tracked ATR at 0.81 and none of its settings was positive in both halves on gold M15. The ATR settings guide gives you the values for each chart.
• Do not stack it with ADX. ADX was the weakest of the three filters, and the ADX settings guide shows what its own settings did on the same bars.

Frequently asked questions

What is the best Choppiness Index setting for MT4?

On gold M15 in this test, period 14 with breakouts taken only below 45: +$3.16 an ounce a trade over 115 trades after cost, +$4.07 in the first half and +$2.79 in the second. On EURUSD no Choppiness setting made money.

Is the Choppiness Index built into MT4?

No. It is a free custom indicator. Copy the .mq4 file into MQL4, Indicators via File, Open Data Folder, then refresh the Navigator.

Should I use 38.2 and 61.8 on the Choppiness Index?

On gold M15, 38.2 was the weakest cut-off: two of three periods lost. 45 and 50 made money with every period tested.

What is the best period for the Choppiness Index?

14 was the best period at every cut-off on gold M15, ahead of 10 and 20.

Does the Choppiness Index work on the 5-minute chart?

Not with this breakout rule. On EURUSD M5 and gold M5, 0 of 56 filter settings made money after cost.