Oscillators and Momentum
RSI, MACD, and stochastics — measuring the speed of a price move rather than its direction, and reading divergence.
Oscillators are introduced as a different category of tool from trend-following indicators like moving averages — where a moving average measures direction, an oscillator measures the speed or momentum behind a price move, typically bounded within a fixed range (0 to 100 for the Relative Strength Index, for example) that makes "overbought" and "oversold" conditions directly readable. Murphy covers three in depth: the Relative Strength Index (RSI), which compares the magnitude of recent gains to recent losses; MACD (Moving Average Convergence Divergence), built from the difference between two exponential moving averages; and the stochastic oscillator, which compares a closing price to its recent trading range.
A consistent warning across all three tools is that "overbought" does not mean "sell immediately" and "oversold" does not mean "buy immediately" — a strongly trending market can stay at oscillator extremes for a long stretch precisely because the trend is strong, and treating an overbought reading as an automatic reversal signal during a genuine strong trend is presented as one of the most common ways new traders misuse these indicators.
Murphy positions oscillators as most useful in a range-bound or choppy market — exactly the condition where the trend-following tools covered earlier in this course (moving averages, trendlines) struggle most, since those tools are built to identify and follow direction, and a market without a clear direction generates mostly false signals from them. This complementary relationship — trend tools for trending markets, oscillators for range-bound ones — is presented as a reason to read market condition first and pick the appropriate tool category second, rather than applying the same toolkit indiscriminately regardless of what kind of market is actually in front of the trader.
| Oscillator | What it measures | Typical range |
|---|---|---|
| RSI | Magnitude of recent gains vs. recent losses | 0 to 100 |
| MACD | Difference between two exponential moving averages | Unbounded, centered on zero |
| Stochastic | Closing price's position within its recent trading range | 0 to 100 |
Across all three oscillators, the single signal the book treats as most reliable is divergence: price making a new high (or low) while the oscillator fails to make a corresponding new high (or low). This is read as the underlying momentum weakening even while price is still nominally extending the trend — the same kind of participation-fading signal volume gives in the earlier chapter, just measured a different way. A bearish divergence (price higher, oscillator lower) ahead of a top and a bullish divergence (price lower, oscillator higher) ahead of a bottom are both presented as warning signs worth taking seriously specifically because they represent the market's own momentum quietly disagreeing with its own price action.
Imagine a stock making three successive new highs over several weeks, while its RSI reading makes a lower high at each successive peak. Price alone looks unambiguously bullish; the divergence is the market quietly telling a different story — each new high is being made with less underlying force than the one before, exactly the kind of fading-conviction warning volume gives in a different form.
Although RSI, MACD, and the stochastic oscillator are built from different underlying math, they tend to move in broadly similar ways because they are all, at bottom, measuring some version of the same thing: how forcefully price has been moving recently relative to its own recent history. Murphy treats general agreement across two or three oscillators as mild additional confirmation, in keeping with the book's broader instinct to trust a signal more when independent tools agree — the same logic that ran through Dow Theory's two-average confirmation requirement and the volume-plus-pattern reads earlier in this course.
When the oscillators genuinely disagree — RSI showing a bearish divergence while stochastic does not, for instance — the disagreement is usually a function of their different lookback sensitivities rather than a sign one of them is "wrong." A shorter-lookback oscillator reacts faster and will show divergence earlier than a longer-lookback one; the practical lesson is to understand what timeframe a given oscillator is actually measuring rather than treating all oscillator readings as interchangeable.
- Oscillators measure the speed of a price move, not its direction — a separate category of tool from trend-following indicators like moving averages.
- Overbought and oversold readings are not automatic reversal signals — a strongly trending market can stay at an extreme for a long stretch.
- Divergence between price and an oscillator — price extending a trend while the oscillator fails to confirm — is the single most reliable oscillator signal the book identifies.
- Oscillators are most useful in range-bound markets, complementing rather than replacing trend-following tools, which work best when a market actually has direction.
- When multiple oscillators disagree, the cause is usually differing lookback sensitivity, not one of them being simply wrong.