Quick definitions of RSI and MACD
RSI (Relative Strength Index) and MACD (Moving Average Convergence Divergence) are technical indicators used to analyze market momentum and trend changes. They are often discussed together because they reflect different aspects of price behavior, but they are not direct measures of “future direction.”
RSI is a momentum oscillator. It summarizes recent price movements into a single score typically ranging from 0 to 100. MACD is derived from moving averages. It aims to show how the relationship between two averages is shifting, which can reflect changes in momentum and trend.
How RSI works (the simple mechanics)
RSI starts by looking at the recent sequence of price changes over a chosen lookback period (commonly 14 periods, but the period is an input you can vary). The calculation distinguishes between:
- Gains: periods where price increases
- Losses: periods where price decreases
RSI then converts the balance between gains and losses into a bounded score. A common interpretation is that values near the upper end suggest stronger recent gains, while values near the lower end suggest stronger recent losses. Traders sometimes refer to these areas as “overbought” or “oversold,” but these are descriptive labels, not guarantees.
A key modeling assumption is that the same lookback period and the same thresholds are used consistently. If you change the period length or the thresholds, the indicator’s behavior can change significantly.
How MACD works (the simple mechanics)
MACD uses moving averages to compare two time horizons:
- Compute a “fast” moving average and a “slow” moving average.
- Subtract the slow average from the fast average to form the MACD line.
- Compute a further smoothing of the MACD line to form a signal line.
- Often use the histogram as the difference between the MACD line and the signal line.
When the fast average pulls away from the slow average, the MACD line tends to move away from zero. When the averages converge again, MACD tends to shrink toward zero. This framework makes MACD primarily a moving-average-based description of momentum relative to a longer reference.
How they differ and why that matters in forex
RSI focuses on recent price change direction and magnitude (momentum of gains vs. losses). MACD focuses on the changing relationship between two smoothed price averages (momentum and trend shift via convergence/divergence).
Because they measure different things, they can disagree. For example:
- RSI may show strong momentum from recent swings, while MACD lags because moving averages smooth price.
- MACD may react to a trend shift even if RSI stays relatively stable due to the way RSI emphasizes the balance of recent gains and losses.
In forex specifically, both indicators can be applied to any chosen price series (such as closes), but the output will still depend on the selected timeframe and the exact data used. Historical patterns do not automatically become future expectations.
Evidence and a worked-through example (with explicit assumptions)
Assume you analyze a currency price series sampled once per period, and you compute:
- RSI using a fixed lookback period of 14 periods
- MACD using a fixed pair of moving-average windows and a fixed signal smoothing
Now consider a simplified scenario over 15 periods:
- Periods 1–13: price alternates, producing mixed gains and losses.
- Period 14: there is a strong upward move.
- Period 15: the price change is smaller but still positive.
Under these conditions, RSI is likely to rise sharply at period 14 because the “gains vs. losses” balance improves after a stronger positive move. If the moving averages used in MACD are longer, MACD may rise too, but it may do so more slowly because smoothing reduces the immediate impact of one large move.
This example illustrates the difference in sensitivity: RSI is directly shaped by recent gains and losses; MACD is shaped by smoothed averages and therefore often lags during transitions.
Limitations, failure modes, and uncertainty
Neither RSI nor MACD is a standalone prediction tool. Common failure modes include:
- Lag: MACD’s moving averages can delay response during rapid reversals. - Noise and whipsaws: RSI can move into extreme regions during choppy ranges, then quickly retreat. - Threshold dependence: “overbought/oversold” and zero-line interpretations are conventions. Changing thresholds can change conclusions.