What MACD measures
The MACD line is the difference between a fast and a slow exponential moving average. With periods of 12 and 26, it is calculated as the 12-period EMA minus the 26-period EMA. A positive value means the fast average is above the slow one; a negative value means the opposite. MACD is expressed in the instrument’s price units, so it has no universal thresholds comparable across assets.
The signal line applies an average to MACD. Where the histogram is defined as MACD minus the signal, it measures the gap between the two lines. Check the platform’s legend: drawing a series as bars does not identify its formula.
The parameters do not describe the full implementation
The notation 12, 26, 9 specifies periods, but not necessarily the signal’s averaging method. MetaTrader 5 documentation, for example, uses a 9-period simple average of MACD; other implementations allow an EMA. The input price, averaging method and initialisation can produce differences.
Before comparing TotalTrade with another chart, check the actual settings on both. MACD appearing in the catalogue does not imply that every variation on every platform has identical parameters.
Signal crossovers and zero crossings
MACD can cross its signal while both lines are still below zero. That differs from MACD crossing above zero. The first condition compares MACD with its average; the second directly compares the fast and slow EMAs.
One rule to investigate is MACD being below the signal on the previous bar and above it on the bar that has just closed. Then define execution on the following bar, a stop and an exit. Requiring MACD to be above zero adds a filter: record its effect on trades separately.
Making the test repeatable
Fix the asset, timeframe, averaging method, applied price and reading time. Allow enough initial data for the averages to be calculated, and do not change the starting period between variations without considering the effect on early values.
Compare trade count, average net result and drawdown. Check whether frequent crossovers in sideways markets erode the advantage through costs. If you add a volatility or session filter, test it as a new hypothesis and check the result outside the period used to choose it.
