Coppock Curve: The Indicator That MetaTrader Does Not Ship
Four published explanations of this indicator were available to read for this page, and all four were written from an equity desk. That shows in what none of them mentions: on the platform most currency traders open, this indicator does not exist until somebody else writes it.
Key takeaways
- The MQL5 technical indicator reference enumerates the standard MetaTrader functions and there is no Coppock among them; the MetaTrader 5 help index carries no page for it either.
- On MetaTrader it is therefore a third-party build, which means the formula and the default numbers belong to whoever wrote the copy you installed rather than to the platform.
- The construction was defined around monthly data, and the builds that distribute it expose the lookbacks as plain integers with no unit attached anywhere in the interface.
- The performance claims in circulation are uncited, and the encyclopedia entry that carries the best known of them tags them as uncited on its own page.
- The four explanations do not agree on the year the indicator was published, and one of them also gives the author a different first name.
Table of contents
It Is Not One of the Indicators MetaTrader Ships
The MQL5 reference lists every technical indicator function the platform provides, from the accelerator oscillator through to the volumes indicator. The list runs to thirty-nine entries and none of them is the Coppock Curve. The MetaTrader 5 help index behaves the same way: no page, no entry, no formula.
That absence changes the question a reader should be asking. On a charting service that ships the indicator, the arithmetic is the service’s and can be checked against its own documentation. On MetaTrader there is no such document, because there is nothing for it to describe.
What exists instead is a custom build. One of the four pages read for this page is not an explanation at all but a download, offering a version for MetaTrader 4, MetaTrader 5 and cTrader, carrying its own version number and its own release date. Whoever wrote that build decided the formula and chose the defaults.
The practical consequence is that two readers can both say they are running the Coppock Curve on MetaTrader and be running two different calculations, with no vendor page either of them can consult to find out. An indicator the platform does ship, such as the Relative Vigor Index, carries exactly the vendor page this one lacks. Installing one is the same operation as installing any other custom indicator on MetaTrader, with the same absence of a guarantee about what the code does.
The Lookbacks Were Chosen in Months and the Interface Does Not Say So
The design is a long-horizon one. The two lookbacks it subtracts were specified in months, and the smoothing applied over them was specified in the same unit, which is why the tool is discussed with reference to market cycles rather than to sessions.
Nothing in a charting interface carries that unit. The build distributed for MetaTrader exposes three inputs named for a period and a moving-average length, and every one of them takes a plain integer. The integer means bars of whatever timeframe the chart is set to, and the chart does not know what the designer had in mind.
So a reader who loads it on a one-hour currency chart is running a fourteen-bar lookback where a fourteen-month one was intended. The output still draws, still crosses zero and still looks like the pictures in every explanation of it. It is measuring fourteen hours where fourteen months were intended.
One of the four explanations does raise adjusting the parameters for other timeframes, and treats that as flexibility. It is flexibility in the sense that the code accepts the number.
Whether the resulting series still carries the meaning the original construction claimed is a separate question, and none of the four asks it. The general version of this problem, that the lookback is the setting doing the real work, is set out in our page on why the leading and lagging labels are about use.
What the Three Numbers Actually Do
The construction has two stages. Two rate-of-change readings are taken over different lookbacks and added together, and the sum is then smoothed by a weighted moving average, which is the kind that gives recent bars more weight than older ones by a fixed schedule rather than by decay.
The three settings below are the defaults of the third-party build named above, taken from that distributor’s own page for its own software. They are not a vendor standard, because no vendor publishes one.
| Input in that build | Default | What it governs |
|---|---|---|
| First rate-of-change lookback | 14 | How far back the longer of the two comparisons reaches |
| Second rate-of-change lookback | 11 | How far back the shorter comparison reaches |
| Moving-average length | 10 | How heavily the summed series is smoothed before it is drawn |
Two of the four explanations attach an origin story to the choice of the first two numbers. Neither traces it to a document, and the encyclopedia entry that repeats it points at an archived glossary rather than a primary source, so this page records that the story exists and stops there. How a weighted average differs from the alternatives is covered in our page on how a weighted moving average is built.
Nobody Sources the Performance Claims
The most widely repeated claims about this indicator are success rates in rising and falling markets, and a threshold said to identify a particular kind of rally. Both appear on the encyclopedia entry, and both are flagged on that page as needing a citation, by the encyclopedia itself.
Because no official source confirms either, the figures are not reproduced here. A number that the page carrying it marks as uncited is not evidence, and repeating it on a seventh page would only make it look better travelled than it is.
The other detailed explanation lists specific historical signals and a count of how many have occurred since the late nineteen-eighties. Those are stated without a dataset, a method or a source, and its most recent example is from 2008. There is no published, independent evaluation of this indicator to consult, which is a narrow and important fact rather than a criticism of the tool.
The Sources Do Not Agree on the Year
Two of the four date the first publication to 1962 and one dates it to 1965. The encyclopedia entry is the only one that supplies anything checkable behind its date, naming the magazine, the issue date and a 1964 thesis as the reference, and that entry is itself a secondary source rather than the original.
One of the four also gives the author a different first name from the one the encyclopedia entry records. The disagreement is small, and it matters only in one way: four pages describing one tool cannot agree on when it was published or who published it, which is a reasonable measure of how carefully the rest was checked.
Who Should Not Load It
Anyone trading a currency pair intraday is applying a monthly-horizon construction to a series it was never specified over, and no source consulted here claims that transfer works. Anyone who wants a checkable formula will not find one on MetaTrader, because the platform does not publish it.
If the interest is in the mechanism rather than the name, the two components are each documented on their own, and the more useful reading is to go to them directly. Our pages on how a momentum reading is constructed and on how these tools fit together carry that ground, and both describe calculations the platform actually ships and documents.
Risk notice. This page is educational and describes how one technical indicator is constructed and what documentation exists for it. Nothing here is a recommendation to buy or sell any instrument, no reading of any indicator is a forecast, and no figure above is presented as a result anyone should expect. Leveraged trading carries a high risk of loss.