What it is
The Relative Strength Index is a bounded momentum oscillator published by J. Welles Wilder in 1978. It compares the average size of recent up closes with the average size of recent down closes and expresses the result on a 0-100 scale. Despite the name it measures nothing about strength relative to another asset or to a benchmark: the 'relative strength' in the title refers only to the ratio of gains to losses inside the same price series.
How it works
Calculation starts with the period-over-period change in the closing price. Gains and losses are separated, then each is smoothed with Wilder's moving average, an exponential average using a smoothing factor of 1/N rather than the more common 2/(N+1). Relative strength RS is average gain divided by average loss, and RSI = 100 - 100 / (1 + RS). The default lookback is 14 periods. Shorter settings such as 2 or 5 produce a fast, noisy oscillator favoured by short-horizon mean-reversion systems, while 21 or more produces a slow line that behaves closer to a trend filter.
How traders use it
The textbook reading is that values above 70 mark an overbought market and values below 30 an oversold one, but experienced users lean on three other behaviours. The first is divergence, where RSI fails to confirm a new price extreme. The second is the centreline at 50, a rough dividing line between bullish and bearish momentum. The third, and most useful, is the range the oscillator occupies: in a healthy uptrend RSI tends to hold above roughly 40 and reach 80, while in a downtrend it caps out near 60 and visits 20. Where the oscillator refuses to go is usually more informative than any single threshold.
Where it breaks down
The most common misuse is treating an overbought reading as a sell signal. Momentum is persistent, so in a strong trend RSI can sit above 70 for weeks while price keeps advancing, and traders who fade every extreme end up systematically short the strongest part of the move. Divergences are equally unreliable in isolation: they can repeat several times before price responds, and naive divergence rules tested without a trend filter or a defined invalidation level rarely show a durable edge. Treat RSI as a description of the character of a move rather than as an independent trade trigger.
Educational reference. This entry describes how a concept is defined and used. It is not investment advice, not a recommendation, and not a signal. Any rule you build from it should be tested with realistic costs before it is traded, and no historical result guarantees a future one.