MACD Strategy: Real NSE Backtest Across Multiple Timeframes

Moving Average Convergence Divergence (MACD) was developed by Gerald Appel in the late 1970s. Designed to track both trend direction and momentum shifts, the indicator combines three distinct components:

• The MACD Line: Calculated by subtracting the 26-period exponential moving average from the 12-period exponential moving average. When short-term momentum rises faster than long-term momentum, the MACD line moves upward.
• The Signal Line: A 9-period exponential moving average of the MACD line itself. This serves as a trigger line to identify cyclical turning points.
• The MACD Histogram: The vertical visual distance between the MACD line and the Signal line. When the MACD line crosses above the Signal line, the histogram flips from negative to positive.

Traders on the National Stock Exchange (NSE) widely deploy a macd strategy assuming that these mathematical relationships provide a reliable statistical edge.

In quantitative trading, there are two primary methods to generate signals from MACD:

• Signal Line Crossover: Enters long as soon as the MACD line crosses above the 9-period Signal line. This setup triggers relatively early in an emerging price move, but also generates frequent false signals during choppy consolidation.
• Zero Line Crossover (Centerline): Enters long only when the MACD line crosses above the zero line. This requires the 12-period EMA to trade higher than the 26-period EMA, confirming that medium-term trend direction has flipped positive. While it generates fewer false trades, it suffers from substantial lag.

We tested MACD strategies across a 12-month period on a portfolio of 15 large-cap NSE liquid stocks including Reliance, TCS, HDFC Bank, and Infosys. The account was ₹1,00,000, each trade used 20% of it, and at most 5 trades were open at once. Full Indian transaction charges—including STT, brokerage, exchange fees, SEBI charges, stamp duty, GST, and execution slippage—were deducted from every trade.

None of the four setups produced a positive net return after charges:

To evaluate whether a MACD bullish crossover provides any intrinsic statistical predictive power outside of specific stop-loss rules, we analyzed all NIFTY 100 constituent stocks across 9 years of daily trading data from July 2017 to October 2026.

We tracked close-to-close returns 1, 5, and 10 trading sessions after a bullish signal line crossover on daily charts.

In the event study, stocks were positive 5 days after a bullish cross 53.5% of the time, modestly higher than the baseline rate of 52.7% across all trading days. By 10 days, the win rate reached 54.4% (baseline 53.7%). While an underlying directional drift exists, our backtest proves that trading it mechanically with standard stops and targets is insufficient to overcome round-trip Indian equity costs.

Across the 1-hour test alone, the system generated 725 trades over 12 months. Raw gross price moves totaled +₹1,108. However, paying ₹13,456 in cumulative statutory charges turned the final result into a net loss of -₹14,880.

Because MACD is a lagging indicator built entirely from moving averages, it responds slowly to rapid reversals. In trendless or sideways markets, it consistently buys at local highs and sells at local lows, triggering whipsaws that rapidly bleed trading capital through repeated brokerage and STT friction.