RSI, MACD, and Bollinger Bands
Three common indicators, all derived from the price path: relative strength, the gap between two averages, and distance from normal volatility.
RSI measures how strong recent upward moves were relative to downward moves. It runs from 0 to 100. Values above 70 are called overbought, below 30 oversold.
MACD shows the gap between a short and a long average. As the gap widens, the move is strengthening. As it narrows, the move is losing steam.
Bollinger Bands place two lines around an average, spaced according to volatility. They show whether the price currently sits unusually far from its mean.
Important: all three are calculated from the price. They contain no additional information, they just present the same information differently. Ten indicators on a chart don't mean ten opinions, they mean one opinion shown ten ways.
RSI is calculated as 100 − 100/(1 + RS), where RS is the ratio of average upward to downward moves over the chosen period. The usual 70 and 30 thresholds are conventions with no theoretical derivation. In sustained trends, the indicator stays in the extreme zone for extended periods, which is why it systematically fails there as a reversal signal.
MACD is the difference between two exponential averages, plus a signal line that's a smoothed version of that difference. It's therefore a trend-following indicator in a different form and shares its properties, including the lag.
Bollinger Bands use a multiple of the standard deviation around a moving average. The common claim that a certain share of prices sits within the bands assumes normally distributed returns. Since financial-market returns have heavier tails, breaches occur considerably more often than that assumption predicts, weakening its use as an exception signal.
Summary
- All indicators are derived from price; none adds new information.
- In a trend, RSI stays permanently in the extreme zone and fails as a reversal signal.
- Bollinger Bands assume a distribution that doesn't actually hold.
Did you get it?
What does RSI measure?
The ratio of average upward to downward moves over a chosen period.
Why does RSI fail in strong trends?
Because it stays permanently in the extreme zone there and keeps signaling reversal.
What assumption is built into Bollinger Bands?
Normally distributed returns. In reality, the tails are heavier, so breaches happen more often.
Related
- What return actually meansStage 0
- Moving averagesStage 3
- Why indicators lagStage 3