RSI (Relative Strength Index)
The Relative Strength Index (RSI) is a momentum oscillator developed by J. Welles Wilder that measures the speed and magnitude of a security's recent price changes on a scale of 0 to 100, used to identify overbought conditions (RSI above 70) and oversold conditions (RSI below 30) as well as momentum divergences between price and indicator that may signal impending reversals.
Key takeaways
- RSI is calculated from the ratio of average gains to average losses over a user-specified period (typically 14 periods), normalized to a 0–100 scale.
- RSI above 70 is conventionally considered overbought (potential sell signal); RSI below 30 is considered oversold (potential buy signal).
- Divergence between RSI and price—where price makes new highs but RSI does not—is considered a bearish reversal signal (and vice versa for bullish divergence).
- RSI is most effective in ranging/sideways markets; in strong trending markets, it can remain overbought or oversold for extended periods, leading to false signals.
- The 50-level acts as a regime indicator: sustained RSI above 50 signals bullish momentum; below 50 signals bearish momentum.
Explanation
The Relative Strength Index was introduced by J. Welles Wilder in his 1978 book 'New Concepts in Technical Trading Systems' and remains one of the most widely used technical indicators in financial markets. Unlike traditional 'relative strength' comparisons between two securities, the RSI measures the internal momentum of a single security by comparing the magnitude of recent price gains against recent price losses.
The RSI calculation begins by computing average gains and average losses over a specified lookback period (default 14 periods). The initial average gain is the simple average of all upward closes over the 14-period window; subsequent values use an exponential smoothing (Wilder's method: Previous Avg Gain × 13 + Current Gain) / 14. The Relative Strength (RS) is then computed as Average Gain / Average Loss, and RSI = 100 - (100 / (1 + RS)). This normalization ensures RSI always falls within the 0–100 range, making it an oscillator—an indicator that cycles within fixed bounds rather than trending indefinitely.
The overbought/oversold thresholds at 70 and 30 are the most commonly cited RSI signals, but experienced practitioners apply them contextually. In a strong uptrend, RSI may remain above 70 for weeks or months—an indicator that the trend is robust rather than a signal to sell. Wilder himself recommended adjusting thresholds to 80/20 in strongly trending markets. The most reliable RSI signals come from divergences: when price makes a new high but RSI fails to match it (bearish divergence), the rally lacks momentum confirmation, increasing the probability of a reversal. Conversely, when price makes a new low but RSI diverges upward (bullish divergence), sellers are exhausting their momentum.
RSI is applied across all timeframes—from one-minute intraday charts used by high-frequency scalpers to monthly charts used by long-term portfolio managers. The interpretation framework is timeframe-consistent: on daily charts, an RSI above 70 suggests the security has been strong over the past two-to-three weeks relative to its recent history; on monthly charts, the same reading reflects multi-month momentum. Many algorithmic trading systems use RSI as a filter or reversal trigger within larger systematic frameworks, combining it with trend indicators (moving averages), volume analysis, and price patterns to reduce false signals.
Quantitative finance researchers have tested RSI-based trading rules extensively. The evidence is mixed but generally shows that RSI-based contrarian strategies (buying oversold, selling overbought) produce positive returns in mean-reverting markets (equities in range-bound periods) but negative returns in trending markets. The 50-level crossover strategy—going long when RSI crosses above 50 and short when it crosses below—captures trend momentum more reliably than overbought/oversold strategies in many asset classes. Combining RSI with price-based stop-losses is a common risk management technique to limit losses when the contrarian signal fails.
Formula
RSI = 100 - (100 / (1 + RS)); RS = Average Gain (n periods) / Average Loss (n periods)
Example
Over 14 trading days, a stock closes with the following changes: +1.5%, +2.0%, -0.5%, +1.0%, -1.5%, +0.8%, +2.2%, -0.3%, +1.1%, -0.7%, +0.9%, +1.8%, -0.4%, +1.2%. Average Gain = (1.5+2.0+1.0+0.8+2.2+1.1+0.9+1.8+1.2)/14 = 12.5/14 = 0.893%. Average Loss = (0.5+1.5+0.3+0.7+0.4)/14 = 3.4/14 = 0.243%. RS = 0.893/0.243 = 3.68. RSI = 100 - (100/(1+3.68)) = 100 - 21.4 = 78.6. An RSI of 78.6 signals overbought conditions. If, over the subsequent two weeks, the stock continues rallying to new highs but RSI only reaches 72 (failing to match the prior high), this bearish RSI divergence—price new high but RSI not confirming—would alert a technical analyst to tighten stop-losses or consider a partial exit as momentum is waning.
Related terms
Algorithmic Trading Default Doji Head And Shoulders Pattern Overbought Oversold Point And Figure Chart Rally Relative Strength Reversal Simple Moving Average Speed