MACD

MACD (Moving Average Convergence Divergence) is a momentum indicator built from the spread between two exponential moving averages, developed by Gerald Appel in the late 1970s. MACD, also known as line/signal/histogram, signal cross, zero-line cross, histogram inflection, is a Momentum & Oscillators concept. The Library holds 3 implementations, each one ready to run on Vela.

Top MACD indicators

The top custom implementations, built on the original standard MACD formula.

3 total

Want to trade MACD? Put the top implementation below on a live chart in Vela, and Quant will help you build a strategy on it and backtest it.

Open in Vela
MACD Based Price Forecasting preview

MACD Based Price Forecasting

Indicator

Adaptive MACD preview

Adaptive MACD

Indicator

The standard MACD indicator

MACD exactly as classically defined: the faithful reference build of the original formula, free to run in Quant.

MACD preview

MACD

Indicator

What is MACD?

MACD (Moving Average Convergence Divergence) is a momentum indicator built from the spread between two exponential moving averages, developed by Gerald Appel in the late 1970s. The MACD line is the 12-period EMA of price minus the 26-period EMA; the signal line is a 9-period EMA of the MACD line itself; and the histogram, added by Thomas Aspray in 1986, plots the gap between the two. The result is unbounded and scaled in price units.

The three components give three readings at different speeds. The zero-line cross is the slowest: MACD above zero simply means the 12 EMA is above the 26 EMA, a trend statement equivalent to a moving average crossover. The signal cross is faster and marks momentum turning relative to its own recent path. The histogram inflection is the earliest: shrinking bars show the spread narrowing before any cross prints.

MACD matters because it packages trend and momentum in one pane, and its line/signal/histogram grammar became the template for a whole lineage of oscillators. Its main structural limit is the price scaling: a MACD value of 2 means nothing across instruments, or across one instrument's history at different price levels, which is what percentage and volatility-normalized variants like the PPO were built to fix.

How to calculate MACD

Three EMAs and two subtractions specify the whole study; 12, 26, and 9 are the conventional defaults.

  1. 1Compute two EMAs of the source (usually the close): a fast one, default 12 periods, and a slow one, default 26.
  2. 2Subtract: the MACD line equals the fast EMA minus the slow EMA. Positive means the fast average is above the slow, and distance from zero measures the width of the spread.
  3. 3Smooth the result: the signal line is an EMA of the MACD line, default 9 periods.
  4. 4Take the difference again: the histogram equals the MACD line minus the signal line. Bars shrinking toward zero show convergence ahead of a signal cross; growing bars show the move accelerating.

How it's calculated

MACD measures the gap between a fast and a slow EMA of price, with a signal line and histogram derived from that gap.

MACD⁡t=EMA⁡(C,nfast)t−EMA⁡(C,nslow)t\operatorname{MACD}_t = \operatorname{EMA}(C, n_{\text{fast}})_t - \operatorname{EMA}(C, n_{\text{slow}})_t
Signal⁡t=EMA⁡(MACD⁡,nsig)t\operatorname{Signal}_t = \operatorname{EMA}(\operatorname{MACD}, n_{\text{sig}})_t
Hist⁡t=MACD⁡t−Signal⁡t\operatorname{Hist}_t = \operatorname{MACD}_t - \operatorname{Signal}_t
EMA recursion: EMA⁡(x,n)t=α×xt+(1−α)×EMA⁡(x,n)t−1\text{EMA recursion: } \operatorname{EMA}(x, n)_t = \alpha \times x_t + (1 - \alpha) \times \operatorname{EMA}(x, n)_{t-1}
α=2n+1\alpha = \frac{2}{n + 1}
C: close price series
t: bar index
n_fast: fast EMA length (default 12)
n_slow: slow EMA length (default 26)
n_sig: signal EMA length (default 9)
MACD_t: MACD line at bar t
Signal_t: signal line at bar t
Hist_t: histogram value at bar t
EMA(x, n)_t: exponential moving average of series x with length n, at bar t
x: input series being averaged (price for the MACD line, the MACD line for the signal)
alpha: EMA smoothing factor
n: length of whichever EMA is being computed

Defaults are 12, 26, 9 on closing prices.

Signal crosses, zero-line crosses, and histogram inflections are readings taken from these three series, not separate calculations.

How traders use it

  • Signal-line crosses as momentum triggers: buying the MACD line crossing above the signal and selling the cross down. Raw crosses whipsaw badly in a trading range, so most rules add a trend or regime filter on top.
  • Zero-line crosses as trend confirmation: above zero as bullish regime, below as bearish, often applied as a higher-timeframe filter that gates faster signals taken on lower timeframes.
  • Histogram inflections as early warnings: the first shrinking bar after a run of expansion flags fading thrust before either cross, useful for tightening stops or scaling out rather than reversing outright.
  • Divergence: price making a new extreme while the MACD line or histogram makes a lesser one is a classic regular divergence read, generally traded only with further confirmation.

MACD vs related oscillators

PPO: Structurally identical, but PPO divides the EMA spread by the slow EMA and quotes it as a percentage. That makes readings comparable across instruments and across history, which raw MACD's price scaling cannot do. Signals have the same shape; only the units change.

OsMA: OsMA is MACD's histogram extracted as its own indicator: the MACD line minus the signal line. Platform labels for it vary, so check the formula rather than the name. If you trade histogram inflections, you are trading OsMA by another name.

MACD-V: A volatility-normalized MACD that scales the EMA spread by average true range, so a given reading means roughly the same thing in any instrument or regime. The normalization restores meaningful overbought and oversold zones, which raw MACD lacks.

RSI: RSI is bounded 0-100 and measures stretch by comparing average gains to losses; MACD is unbounded and measures the spread between trend-following averages. RSI saturates in strong trends while MACD keeps extending with them, which is where the two disagree most usefully.

MACD FAQ

Turn MACD into a trading strategy.

Open the top implementation from this page in Vela. Quant, our coding agent, builds the strategy with you and backtests it on real data.

Build a strategy in Vela
Browse this familyMomentum & Oscillators85 concepts mapped · 85 in the Library