MACD is taught in virtually every trading textbook, and yet most new traders use it wrong. They watch for every tiny crossover, chase every hint of divergence, and end up making bad decisions on random noise. The traders who make MACD work understand what it's actually measuring and use it as a filter rather than a trigger.
Direct Answer
**MACD (Moving Average Convergence Divergence)** is a momentum indicator developed by Gerald Appel in the late 1970s. It measures the relationship between two exponential moving averages (EMAs) of price — typically the 12-period EMA and the 26-period EMA.
The indicator has three components:
1. **MACD line**: 12-period EMA minus 26-period EMA. This is the primary line. 2. **Signal line**: 9-period EMA of the MACD line. Smooths the MACD line and generates crossover signals. 3. **Histogram**: MACD line minus signal line. The histogram is positive when the MACD is above the signal line (bullish momentum) and negative when below (bearish momentum).
MACD generates three types of signals:
- **MACD crossover**: when the MACD line crosses above or below the signal line. Bullish crossovers (MACD crossing above signal) suggest upward momentum; bearish crossovers suggest downward momentum.
- **Zero-line crossover**: when the MACD line crosses above or below zero. Crossing above zero means the 12-EMA is now above the 26-EMA (overall uptrend). Crossing below zero is the opposite.
- **Divergence**: when price makes a new high (or low) but MACD fails to confirm. Bullish divergence = price makes lower low, MACD makes higher low. Bearish divergence = price makes higher high, MACD makes lower high. Divergences are the strongest MACD signals but also the hardest to identify correctly.
MACD works on any timeframe (daily, 4-hour, 15-minute) and any instrument (stocks, forex, crypto, futures). Default settings are 12, 26, 9, but many traders adjust these for shorter or longer timeframes.
Screenshot any chart with MACD and Charted identifies the current signal type, confirms divergence with volume analysis, and explains the signal strength based on the broader price structure.
This content is for educational purposes only and does not constitute financial advice. Trading involves significant risk of loss.
How MACD Is Actually Calculated
Understanding the math helps you interpret the signals correctly.
Step 1: Calculate the 12-period EMA of closing prices.
EMA gives more weight to recent prices. The 12-EMA is more sensitive to short-term price changes.
Step 2: Calculate the 26-period EMA of closing prices.
The 26-EMA is slower and represents the longer-term trend.
Step 3: Subtract the 26-EMA from the 12-EMA.
This is the MACD line. When the 12-EMA is above the 26-EMA (short-term price above long-term average), MACD is positive. When below, MACD is negative.
Step 4: Calculate the 9-period EMA of the MACD line.
This is the signal line — a smoothed version of MACD that lags slightly.
Step 5: Calculate the histogram.
MACD line minus signal line. When MACD is above the signal line, histogram bars are positive (plotted above the zero line). When below, they're negative.
Why these specific numbers?
The 12-26-9 settings are the defaults Gerald Appel used when he invented the indicator in the late 1970s. They work across many markets and timeframes. Some traders adjust for different characteristics:
- **Faster (5-35-5)**: more signals, more noise. Used for short-term trading.
- **Slower (19-39-9)**: fewer signals, less noise. Used for longer-term position trading.
- **Zero-lag MACD**: variations that attempt to reduce the lag inherent in EMAs.
Most professional traders stick with 12-26-9 because the default provides good universality and the community recognizes it — when someone mentions a MACD crossover, they're referring to these settings.
Signal 1: MACD Line Crossover (Most Common Signal)
**Bullish MACD crossover**: MACD line crosses above the signal line.
- Suggests short-term momentum is turning up
- Generates a common buy signal in trading systems
- Often the first signal of a trend change
**Bearish MACD crossover**: MACD line crosses below the signal line.
- Suggests short-term momentum is turning down
- Common sell signal in trading systems
- Often the first warning of a correction
**Problems with crossover signals**:
1. **Lag**: by the time MACD crosses, price has often already moved significantly. The 9-period signal line adds an additional 9-period lag on top of the 12/26 EMAs.
2. **False signals in range-bound markets**: in sideways price action, MACD crosses back and forth repeatedly, generating many false signals. MACD is a trend-following indicator; it performs poorly in chop.
3. **Weak crossovers near zero line**: crossovers that happen right at the zero line (when MACD is barely positive or barely negative) are weaker than crossovers far above or below zero.
**How to use crossovers effectively**:
- Use as a confirmation tool, not a primary signal
- Require the crossover AND alignment with price structure (e.g., crossover after price broke above resistance)
- Use multiple timeframes — a crossover on the daily chart aligned with a crossover on the weekly is stronger than a daily crossover alone
- Avoid trading crossovers during sideways/range markets
Signal 2: Zero-Line Crossover (Trend Confirmation)
When the MACD line itself crosses above or below zero, it's a more structural signal than a MACD/signal crossover.
**MACD crossing above zero**: the 12-EMA is now above the 26-EMA. The short-term trend has overtaken the longer-term average — typically interpreted as confirmation of an uptrend.
**MACD crossing below zero**: the 12-EMA is now below the 26-EMA. Downtrend confirmation.
**Why zero-line crossovers matter**: they're slower than MACD/signal crossovers (so they filter out more noise) but indicate a more significant trend shift. Many traders use zero-line crossovers as a trend filter — only taking long trades when MACD is above zero, only taking short trades when below.
**Practical application**:
- **Trend filter**: require MACD > 0 before entering longs, MACD < 0 before entering shorts
- **Reduce false signals**: by requiring both a MACD/signal crossover AND MACD to be on the correct side of zero, you filter out many weak signals
- **Structural reference**: a security with MACD consistently far above zero is in a strong uptrend. A security with MACD hovering around zero is in chop.
Signal 3: Divergence (Strongest Signal, Hardest to Use)
Divergence occurs when price and MACD disagree about the strength of the move. It's the most powerful MACD signal — but also the easiest to misidentify.
**Bullish divergence**: price makes a LOWER low, but MACD makes a HIGHER low (or a less-negative low). This suggests downward momentum is weakening even though price is still falling. Often precedes a reversal to the upside.
**Bearish divergence**: price makes a HIGHER high, but MACD makes a LOWER high (or a less-positive high). This suggests upward momentum is weakening even though price is still rising. Often precedes a reversal to the downside.
**Example — bullish divergence**:
Say a stock trades: - Swing low 1: $50 on March 1, MACD at -2.5 - Swing low 2: $48 on March 20, MACD at -1.8
Price made a lower low ($48 < $50) but MACD made a higher low (-1.8 > -2.5). This is bullish divergence. The downward momentum is fading even though price itself is still making new lows. Often a reversal signal.
**Example — bearish divergence**:
- Swing high 1: $100 on February 1, MACD at 3.5
- Swing high 2: $105 on February 25, MACD at 2.2
Price made a higher high ($105 > $100) but MACD made a lower high (2.2 < 3.5). Bearish divergence. The rally is losing momentum.
**Why divergences are powerful**: they often precede trend reversals by days or weeks. When price action looks strong but momentum is quietly fading, divergence is one of the few indicators that catches it.
**Why divergences are hard**:
1. **False divergences are common**: not every divergence produces a reversal. In strong trends, divergences can appear and resolve without any reversal at all.
2. **Timing is vague**: divergences can persist for weeks before price actually reverses. Trading divergence at the first sign is premature; waiting too long means missing most of the move.
3. **Requires swing identification**: you need to correctly identify which highs or lows count as 'the' swing high/low. Noise in price action produces minor swings that aren't meaningful for divergence analysis.
**How to use divergences effectively**:
- Require divergence + a confirmation signal (price breaking a trend line, MACD/signal crossover, candlestick reversal pattern)
- Look for divergences at clear support or resistance levels — those are the highest-probability setups
- Use multiple timeframes — daily divergences in the direction of a weekly trend are weaker than divergences against a weak weekly trend
- Avoid divergences in strong trending markets (strong uptrend + bearish divergence often fails to reverse)
Common MACD Mistakes
Mistake 1: Treating every crossover as a signal
MACD generates many crossovers, especially on shorter timeframes. Not all are tradeable. Require additional confirmation (price structure, volume, alignment with higher timeframe).
Mistake 2: Ignoring the bigger picture
MACD on a 15-minute chart telling you to buy while the daily chart shows a strong downtrend is usually a bad trade. Always align with higher-timeframe context.
Mistake 3: Using MACD alone
MACD works best combined with other tools: price action, support/resistance, volume, and broader market context. Using MACD as a solo signal produces inconsistent results.
Mistake 4: Adjusting settings obsessively
Switching from 12-26-9 to 10-20-7 to 8-16-5 based on recent performance is usually curve-fitting. Stick with standard settings and focus on reading the signals correctly.
Mistake 5: Trying to catch every divergence
Divergences are powerful but not magical. Many divergences resolve without reversal. Treat divergence as a warning sign that invites further investigation, not an automatic trade trigger.
Screenshot your MACD chart and Charted analyzes the current signal (crossover, zero-line position, or divergence), identifies potential false signals based on the broader price structure, and shows the historical reliability of similar setups.