Chaikin Oscillator Settings the Vendor Documents Two Ways
This page was written against six published explanations of the oscillator, each one read end to end. All six give the same formula, and the company that ships the indicator to most of the readers who will load it publishes that formula in one document and contradicts it in another. Which of the two your platform follows decides what the line on your screen does.
Key takeaways
- The MetaTrader 5 help page states the calculation as an exponential average of 3 against an exponential average of 10, both taken of the Accumulation/Distribution line.
- The MQL5 programming reference for the same indicator requires a smoothing type as an argument, and it accepts four of them: simple, exponential, smoothed and linear-weighted.
- Not one of the six explanations read for this page uses the phrase smoothing method at all, so a reader following any of them will not know the choice exists.
- Three separate tools carry the Chaikin name and MetaQuotes exposes each as its own function, so a page answering one of them may be answering a different tool than the one you loaded.
- The 3 and the 10 have a citable home in the vendor help page; a slower alternative pair repeated by two of the explanations has none anywhere that could be found, so this page does not carry those numbers.
- Five of the six explanations show no publication date and none shows a modified date.
Table of contents
What the Oscillator Is Built On
The oscillator is not calculated from price. It is calculated from a second indicator, the Accumulation/Distribution line, and reports the gap between two averages taken of that line.
That matters more than it sounds. Anything the underlying line cannot see, the oscillator cannot see either, and any setting that changes the underlying line changes every reading above it. The oscillator inherits its inputs rather than choosing them.
MetaQuotes exposes the underlying line as its own function in the MQL5 reference, with the symbol, the timeframe and a volume type as its arguments, and states that it returns a single buffer. The oscillator function sits separately in the same list and adds three arguments of its own: a fast period, a slow period and a smoothing type.
The line those arguments are applied to is a running total rather than a bounded reading. Each bar contributes an amount weighted by where the close finished inside that bar range, so a close near the high adds and a close near the low subtracts, and the contributions accumulate without limit.
A running total has no natural level, which is the reason a second step exists at all. Subtracting a slow average of the total from a fast one strips the level out and leaves the rate of change behind.
So the honest description of the tool is a subtraction performed on a derived series. It is closer in construction to the difference-of-averages family than to anything that reads price directly, which is also why it belongs with the lookback settings that do the real work rather than with pattern reading.
The Vendor Documents the Smoothing Two Different Ways
Every one of the six explanations presents the calculation as a single fixed recipe: an exponential average of 3 minus an exponential average of 10, taken of the Accumulation/Distribution line. The MetaTrader 5 help page agrees with them and prints exactly that formula in its calculation section.
The MQL5 reference for the same indicator does not. Its function signature requires a smoothing type to be passed in, and the enumeration it draws from lists four values: simple averaging, exponential averaging, smoothed averaging and linear-weighted averaging. The averaging type is an input, not a constant.
Two documents from the same company therefore describe the same named indicator differently. One states the maths as settled, the other makes a third of it a decision the caller makes. Neither is wrong about its own scope, and the reader is the one holding both.
The consequence is concrete. Two traders can run the Chaikin Oscillator on the same pair, on the same timeframe, with the same 3 and the same 10, and see two different lines, because one of them is averaging exponentially and the other is averaging simply. Crossings of the zero line will not occur on the same bars. Divergences will not begin at the same point.
How far apart the two lines run is not a matter of taste either. A simple average gives every bar in its window the same weight and drops the oldest one entirely when the window moves on, while an exponential average keeps a decaying share of every bar it has ever seen and weights the newest most heavily.
On a fast window of 3 the difference is small and quick. On a slow window of 10 it is large enough to move where the subtraction lands, and the subtraction is the whole indicator.
Nothing in the six explanations prepares a reader for that. The phrase smoothing method appears in none of them, so the setting is invisible to anyone who learned the indicator from the first page of results and then went looking for it in a settings dialogue. The wider point about which setting on an indicator actually carries the weight is covered in our guide to how these tools fit together.
The practical instruction is short. Before comparing your reading of this oscillator with anyone else, confirm which averaging type your platform loaded, because agreeing on the two periods is not agreeing on the calculation.
Three Tools Carry the Chaikin Name
The second confusion is not about a setting. It is about which tool is being discussed at all.
MetaQuotes lists the Chaikin Oscillator, the Accumulation/Distribution line and the Money Flow Index as three separate functions in its technical indicator reference, each returning its own handle. The MetaTrader 5 help files them in two different sections as well: the oscillator sits under Oscillators, while the Accumulation/Distribution line sits under Volume Indicators.
Chaikin Money Flow is a fourth name in the same family and it is not the oscillator either. The distinction is not academic on a search results page, where an explanation of one appears among the results for another, and where the questions Google itself attaches to this oscillator ask about money flow instead.
| Tool | What it reports | Where MetaTrader 5 help files it |
|---|---|---|
| Chaikin Oscillator | The gap between a fast and a slow average of the Accumulation/Distribution line | Oscillators |
| Accumulation/Distribution | A running total, the series the oscillator is calculated from | Volume Indicators |
| Money Flow Index | A bounded reading on its own scale, not built on the line above | Oscillators |
What each of the last two is used for on a MetaTrader chart is set out separately in our page on the volume tools the platform ships with, and the Money Flow Index has its own treatment in how a liquidity reading is put together. This page stays with the oscillator.
Where the 3 and the 10 Actually Come From
All six explanations state 3 and 10 as the default periods. Not one of them says where the numbers came from, and two of the six add a slower alternative pair described as a smoother setting, also with nothing attached.
The 3 and the 10 do have a home. The MetaTrader 5 help page for the oscillator prints them inside its calculation section, which makes them the vendor’s stated figures for the vendor’s own implementation rather than folklore. That is worth saying plainly, because a number with a document behind it can be checked and a number without one cannot.
The slower alternative pair is a different case. It was repeated by two of the explanations read here and traced to nothing in the vendor documentation or in any standards body publication that could be reached, so the numbers themselves are not reproduced on this page. A period setting that no document supports is not a setting a reader can check, and reprinting it here would only add a seventh source stating it.
There is a further wrinkle worth naming. The MQL5 function requires the two periods to be supplied by the caller and documents no default of its own, so an indicator loaded through a program rather than through the terminal takes whatever the program passed. A default is a property of an interface, not of the indicator. The same distinction decides how the True Strength Index is read, because the platform reference for that study names its period inputs and sets none of them.
Reading the Zero Line Without Adding a Claim to It
A cross above zero means the fast average of the underlying line has risen above the slow one. That is a statement about two averages, and nothing more is contained in it.
Reading the cross as an instruction imports a claim the subtraction itself never makes, and that claim has to be supported from somewhere outside the indicator. None of the six explanations supplies any support for it, and the illustrative counts in the most detailed of them are drawn from equity charts around 2010 with no test behind them.
One caveat has to be stated once rather than assumed. The Accumulation/Distribution line takes a volume input, and on a spot currency symbol what the platform can hand it is a tally of how often the quote moved rather than how much changed hands, which our page on what the volume figure on a forex chart counts sets out in full. Everything above it inherits that limit.
A reading that survives all of this is narrower. Note how long the oscillator has held one side of zero rather than acting on the moment it changes sides, since a series that flips every few bars is describing indecision in the underlying line and one that holds is describing a stretch of one-way accumulation in it.
None of the Widely Read Explanations Carries a Date
Of the six pages read for this one, exactly one shows a publication date, in February 2024, and it shows no modified date. The other five show neither, and one of the five carries a site-wide copyright year that a reader could easily mistake for a freshness signal.
The most detailed of them illustrates its argument with equity examples whose price action belongs to 2010. That is fifteen years of platform change, instrument coverage and data availability sitting between the example and the reader, with nothing on the page to signal it.
An undated page is not automatically wrong. It is unverifiable, which is a different problem and a worse one on a topic where the vendor documentation has moved and the search results have not.
Who Should Not Load This Indicator
Three groups get nothing from it. Anyone who wants a level rather than a direction is asking a question it cannot answer, because it reports a difference and never a price. Anyone trading a symbol whose volume series is a tick count is reading an underlying line built on activity rather than participation. And anyone who intends to compare readings with another trader is comparing two calculations until both have confirmed the averaging type.
That leaves a narrower group it does suit. If the question is whether the Accumulation/Distribution line has been rising or falling faster than its own recent average, and the answer is wanted as one series rather than as a signal, the oscillator is a reasonable way to see it, provided the averaging type is set deliberately and recorded.
If the question is when to act, the tool does not contain the answer, and no source consulted for this page claims that it does.
Risk notice. This page is educational and describes how one technical indicator is constructed and how its documentation differs between sources. 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.
