RSI
The Relative Strength Index: a 0–100 score of how strongly the price has been rising versus falling lately (usually over 14 days).
4 min readLesson 41 of 57Interactive
In short
Above 70 is often called overbought and below 30 oversold, but a strong trend can stay there a long time. It's a measure of momentum, not a promise of a turn.
Why it matters
Moving averages tell you which way the price has been heading. They don't say how hard it's been pushing. A stock that rose a little on most days and one that rose a lot on every day can both sit above their averages.
The Relative Strength Index measures that push. It compares the size of recent up days with recent down days and turns the result into a score from 0 to 100. It was introduced by J. Welles Wilder in 1978, and it's now one of the most common indicators on charting sites. Traders use it to spot moves that look stretched, though stretched moves can keep stretching.
How it works
Start with the daily changes in the close. Split them into gains (up days) and losses (down days, counted as positive numbers). Over the period, usually 14 days, work out the average gain and the average loss. Days with no gain count as zero gain, and the same for losses.
Divide the average gain by the average loss to get the relative strength, RS. Then squash it into a 0–100 score: RSI = 100 − 100 ÷ (1 + RS). If gains and losses are equal, RS is 1 and the RSI is 50. If there were only gains, the RSI is 100; only losses, 0.
After the first value, the averages are updated with Wilder's smoothing rather than recalculated from scratch: the new average gain is (previous average × 13 + today's gain) ÷ 14, and the same for losses. MarketLab does exactly this, starting from the simple averages of the first 14 changes, so it needs 15 closes before the first RSI appears. Some charting sites use simple averages throughout, which gives slightly different numbers.
RSI = 100 − 100 ÷ (1 + average gain ÷ average loss)
How big recent up days were compared with recent down days, as a score from 0 to 100.
A worked example
Over the last 14 trading days, a stock's up days added $14.00 in total and its down days took away $7.00. Work out the 14-day RSI, then update it after a day when the stock rises $2.00.
- Average gain: $14.00 ÷ 14 = $1.00. Average loss: $7.00 ÷ 14 = $0.50.
- Relative strength: RS = 1.00 ÷ 0.50 = 2.
- RSI = 100 − 100 ÷ (1 + 2) = 100 − 33.3 = 66.7.
- Next day, the stock rises $2.00, so the gain is $2.00 and the loss is $0. New average gain: ($1.00 × 13 + $2.00) ÷ 14 = $15.00 ÷ 14 ≈ $1.0714. New average loss: ($0.50 × 13 + $0) ÷ 14 = $6.50 ÷ 14 ≈ $0.4643.
- New RS = 1.0714 ÷ 0.4643 ≈ 2.31, so RSI = 100 − 100 ÷ 3.31 ≈ 69.8.
Result: The RSI went from 66.7 to 69.8, just under the 70 line that's often called overbought. One more strong day could push it above, but that alone says nothing about whether the rise will continue.
Try it
Reading the RSI
The stock's price over a year.
Its 14-day RSI was above 70 on 3 days and below 30 on 0 days.
Days above 70
3
Days below 30
0
RSI now
47
Shorter lengths react faster and touch the extremes more often; longer ones rarely do.
Generated prices for illustration, not market data.
How to read it
- Above 70
- Conventionally called overbought: gains have far outweighed losses lately. In a strong uptrend, the RSI can stay above 70 for weeks.
- 30 to 70
- The normal range. Around 50, gains and losses have been roughly balanced.
- Below 30
- Conventionally called oversold: losses have dominated. A falling stock can stay below 30 while it keeps falling.
- Price and RSI disagree
- If the price makes a new high but the RSI makes a lower high, traders call it a divergence: the push behind the rise may be fading. It's a warning sign that often comes to nothing.
In the real world
- The 70 and 30 levels are the conventional ones, from Wilder's original work. Some traders use 80 and 20 to get fewer, more extreme signals, and some use a shorter period, like 2 or 5 days, for very short-term trading.
- RSI is mostly used for timing, not for choosing what to own. A trader who already likes a stock might wait for the RSI to dip before buying, or avoid chasing it when the RSI is very high.
- Its biggest weakness is strong trends. Buying because the RSI says oversold is buying a stock that's been falling hard, and sometimes it keeps falling for a long time. Selling at 70 can mean selling early in a long rally. Evidence on whether RSI rules beat the market after costs is mixed, and depends heavily on the period tested.
Common mistakes
- Treating overbought as "will fall" and oversold as "will rise". They describe recent momentum, not what comes next.
- Using RSI rules on their own in a trending market. Many traders pair them with a trend filter, like requiring the price to be above its 200-day average before buying an oversold dip.
- Expecting your RSI to match every site exactly. Smoothing methods and starting points differ.
- Tuning the period and levels until a backtest looks great. That's Overfitting; check that neighboring settings also work.
See it in MarketLab
Test yourself
Optional: 4 quick questions on the words, using the idea, and how it plays out in real markets.
Loading your answers…