MACD: the indicator and the automatable strategy.

The MACD looks intimidating at first — three elements, two numbers, a histogram — but the core idea is simple: it measures whether the trend is accelerating or decelerating, by comparing two moving averages. After this guide, you'll understand each part and have the code to automate the classic strategy.

01What the MACD is

MACD stands for Moving Average Convergence Divergence, created by Gerald Appel in the late 1970s. It's both a trend and a momentum indicator. It has three components:

• MACD line: the difference between the 12-period exponential moving average (EMA) and the 26-period one. When the fast one is above the slow one, it's positive (bullish momentum).
• Signal line: a 9-period EMA of the MACD line itself. It acts as a trigger.
• Histogram: the difference between the MACD line and the signal line. It shows the strength of the move visually.

02The MACD signals

1. Line crossover (the classic)

The most-used signal: when the MACD line crosses above the signal line → buy. When it crosses below → sell. It's the basis of the automatable strategy.

2. Zero-line crossover

When the MACD line crosses above zero, the fast average has overtaken the slow one — confirmation of an uptrend. Below, a downtrend. Slower, but more reliable than the line crossover.

3. Histogram

The histogram grows when momentum accelerates and shrinks when it decelerates. A shrinking histogram (even while still positive) warns that the rally is losing strength — it anticipates the crossover.

4. Divergences

As with the RSI, divergence between price and MACD is powerful: price makes a higher high, MACD makes a lower high → weakening, possible reversal.

03Coding the MACD

//@version=5
strategy("MACD Crossover", overlay=false)

fast = input.int(12, "Fast EMA")
slow = input.int(26, "Slow EMA")
sig  = input.int(9,  "Signal Line")

[macdLine, signalLine, hist] = ta.macd(close, fast, slow, sig)

plot(macdLine, "MACD", color=color.blue)
plot(signalLine, "Signal", color=color.orange)
plot(hist, "Histogram", style=plot.style_histogram,
     color=hist >= 0 ? color.green : color.red)

// strategy: line crossover
if ta.crossover(macdLine, signalLine)
    strategy.entry("Buy", strategy.long)
if ta.crossunder(macdLine, signalLine)
    strategy.close("Buy")

import pandas as pd

def compute_macd(prices, fast=12, slow=26, signal=9):
    ema_fast = prices.ewm(span=fast, adjust=False).mean()
    ema_slow = prices.ewm(span=slow, adjust=False).mean()
    macd_line   = ema_fast - ema_slow
    signal_line = macd_line.ewm(span=signal, adjust=False).mean()
    histogram   = macd_line - signal_line
    return macd_line, signal_line, histogram

macd, signal_l, hist = compute_macd(df["close"])

# detect a bullish crossover on the last candle
crossed_up = (macd.iloc[-2] < signal_l.iloc[-2]) and \
             (macd.iloc[-1] > signal_l.iloc[-1])
if crossed_up:
    print("MACD crossed up — buy signal")

04Improving the strategy

A pure MACD crossover generates many false signals. Filters that improve it a lot:

• Trend filter: only buy if price is above the 200 moving average and MACD crossed up.
• Zero-line filter: only validate the crossover if it happens above zero (already in an uptrend).
• RSI confirmation: MACD crossover + RSI above 50 = a stronger signal.

05Frequently asked questions