Relative strength vs RSI: the difference, with examples
In short
Relative strength vs RSI is a naming trap: the two measure different things. Relative strength compares a stock with a benchmark, stock price ÷ index price, and what matters is the direction of the line. Wilder’s RSI (1978) uses only the stock’s own prices: 100 − 100 ÷ (1 + average gain ÷ average loss), usually over 14 sessions, on a 0 to 100 scale. A stock can have an RSI of 76 and still trail the index.
In 1967 the Journal of Finance published a paper by Robert Levy titled “Relative Strength as a Criterion for Investment Selection”: relative strength used to pick stocks, ranking them against one another. Eleven years later J. Welles Wilder published New Concepts in Technical Trading Systems and introduced an oscillator he called the Relative Strength Index, or RSI. The name was nearly the same, yet Wilder’s calculation uses only the stock’s own prices and compares it with nothing.
The two measures have shared the same charts ever since, and a note that says “strong” can mean two things that sometimes contradict each other. Take a hypothetical stock that rises from $50 to $53.60 over 14 sessions, with nine up days and five down days. Its 14-day RSI ends at 76, above the 70 line many traders call overbought. Over the same 14 sessions its benchmark index rises 10%.
The stock gained 7.2% while the index gained 10%. Its RSI reading of 76 describes its own price moves, while the comparison with the index shows underperformance.
If you trade your own capital you need both pieces of information, as long as you know which one you are reading. Our method starts with the index and only then gets to the stock, which is exactly the comparison relative strength measures.
What is relative strength in stocks?
Relative strength is the ratio of a stock’s price to the price of a benchmark, usually an index, calculated day by day. If a hypothetical stock goes from $100 to $110 while the index goes from 1,000 to 1,050, the ratio rises from 0.1000 to 0.1048. The level of the ratio means nothing on its own; the direction of the line is what counts, and here it rises because the stock gained 10% and the index 5%.
To make the line easier to read, rebase it to 100 on the first day. In the example it reaches 104.76: the stock beat the index by 4.76%, which is not the same as the 5-point gap between the two returns, because the right comparison is 1.10 ÷ 1.05 − 1. Some platforms instead use a percentile rank against a defined group of stocks. That rank depends on the group, lookback period and weighting, so it is a different calculation from the stock-to-index price ratio.
Can a stock outperform the index while losing money?
Yes: all it takes is for the stock to fall less than its benchmark. A stock that drops from $100 to $90 while the index drops from 1,000 to 800 loses 10% against the benchmark’s 20%. The ratio rises from 0.1000 to 0.1125, relative strength of +12.5%, and meanwhile your account holds $10 less for every $100 invested. A chart of relative strength alone doesn’t show that loss.
Now switch benchmarks. Against a sector index that fell only 5%, from 1,000 to 950, the same stock has relative strength of −5.3%: the ratio goes from 0.1000 to 0.0947. One stock, two answers, because the choice of index changes the question.
That is why the benchmark has to be chosen before you look at the result, and written down. A broad index answers “is it doing better than the market?”; a sector index or sector ETF answers “is it doing better than its peers?”, and a company can beat the market while trailing its sector, as we explain in our guide to sector correlation. Use the same dates, the same currency and the same treatment of dividends: comparing a dividend-adjusted stock with a price-only index adds a difference that comes from the data.
What is the RSI and how do you calculate it?
The RSI (Relative Strength Index) is an oscillator running from 0 to 100 that measures how much of a stock’s recent movement is made of gains. Wilder suggested 14 periods and treated readings above 70 as overbought and below 30 as oversold. It takes four steps: split the daily changes into gains and losses, average each, divide the average gain by the average loss (that ratio is called RS), and transform the result with 100 − 100 ÷ (1 + RS).
Here is the math on the hypothetical stock from the opening. The 15 closes are 50.00, 50.40, 50.10, 50.80, 51.30, 51.00, 51.60, 52.10, 51.70, 52.40, 52.90, 52.50, 53.20, 53.00 and 53.60, followed by two sessions at 53.10 and 52.60.
| Step | Calculation | Result |
|---|---|---|
| Changes over 14 sessions | +0.40, −0.30, +0.70, +0.50, −0.30, +0.60, +0.50, −0.40, +0.70, +0.50, −0.40, +0.70, −0.20, +0.60 | 9 gains, 5 losses |
| Sum of gains and losses | gains 0.40 + 0.70 + … + 0.60; losses 0.30 + 0.30 + 0.40 + 0.40 + 0.20 | 5.20 and 1.60 |
| Averages | 5.20 ÷ 14 and 1.60 ÷ 14 | 0.3714 and 0.1143 |
| RS and RSI | 0.3714 ÷ 0.1143 = 3.25; 100 − 100 ÷ 4.25 | 76.47 |
| Day 16, close 53.10 (−0.50) | gains (0.3714 × 13 + 0) ÷ 14 = 0.3449; losses (0.1143 × 13 + 0.50) ÷ 14 = 0.1418 | 70.86 |
| Day 17, close 52.60 (−0.50) | gains 0.3203; losses 0.1674 | 65.67 |
From day 16 on, update the gain and loss averages separately: multiply the previous average by 13, add today’s gain or today’s loss expressed as a positive amount, then divide by 14. Use zero for the side with no movement that day. Wilder’s smoothing has a factor of 1 ÷ 14, as explained alongside the SMA vs EMA moving averages. In this example two declines of $0.50 take RSI from 76.47 to 65.67. These RSI updates alone do not tell us how the benchmark moved.
Relative strength vs RSI: what is the difference?
In the opening example RSI is about 76, while the stock gained 7.2% against the index’s 10%. The relative return is 1.072 ÷ 1.10 − 1, or about −2.5%. The stock rose in absolute terms and still underperformed its benchmark.
| Relative strength | RSI | |
|---|---|---|
| What it compares | the stock with a benchmark | the stock’s gains with its losses |
| Formula | stock price ÷ index price | 100 − 100 ÷ (1 + RS) |
| Scale | open-ended, read the direction | 0 to 100 |
| Write down | index, dates, adjusted data or not | period and platform definition |
| Question | is it beating the market? | how much of the recent move is up? |
The reverse can happen too. In the figure below, on simulated data, the market falls 14.2% over 19 sessions and the stock only 7.3%. Relative strength rises 8.0%; the same stock’s RSI, which looks only at the stock’s own sessions, mostly down ones in this stretch, falls from 70 to 28 and lands in what traders call oversold territory.
What do an RSI of 70 and 30 mean?
They are conventional thresholds proposed by Wilder. An RSI above 70 means smoothed average gains exceed smoothed average losses by a wide margin. After initialization, a 14-period RSI retains weight from earlier observations. The reading does not predict when a reversal will occur. In a strong trend RSI can remain above 70 for weeks, or below 30 during a prolonged decline. Figure 2 shows RSI below 30 while the stock outperforms the market.
The two measures become useful when you read them together, each for the question it answers. There are four possible combinations.
Does relative strength actually help you pick stocks?
Academic work on relative strength goes back to the 1960s. In 1993 Narasimhan Jegadeesh and Sheridan Titman published a study of US stocks in the Journal of Finance: strategies that bought the stocks that had done best over the previous months and sold those that had done worst generated significant positive returns over three- to twelve-month holding periods. The same study noted that part of those excess returns faded over the following two years.
These are results on portfolios of many stocks over past periods, and nothing ensures they repeat on a single chart. If you build a watchlist you review every week, relative strength is still a sensible filter for putting at the top the stocks the market is already rewarding, read on a daily and a weekly chart, with the benchmark always written next to it.
Keep the two indicators in separate panels with their settings written underneath: numerator, denominator and dates for relative strength, period and platform for the RSI. In the TC2000 setup we use every chart has its settings fixed in the layout, so today’s screen can be compared with one from a month ago.
The first RSI average uses fourteen changes; Wilder’s subsequent updates retain a weight from earlier history. “Fourteen periods” therefore does not mean that all previous observations disappear. When both average gains and average losses are zero, the sheet uses 50 as a neutral convention; a platform may display an unavailable value instead, so compare that setting too.
Frequently asked questions
What is the RSI in simple terms?
It is a momentum indicator from 0 to 100, proposed by J. Welles Wilder in 1978. Values near 100 indicate that smoothed average gains dominate smoothed average losses, and near 0 the reverse. It uses only the stock’s own price changes, without comparing them with an index.
What does an RSI above 70 mean?
That smoothed average gains outweigh smoothed average losses by a wide margin. By convention it is called overbought. RSI can remain above 70 during a strong trend, so crossing that threshold alone does not establish a sell signal.
How do you calculate the relative strength of a stock?
Divide the stock’s price by the benchmark index’s price every day and rebase the series to 100 on the first day. If the line rises, the stock is beating the index; if it falls, the stock is trailing, even when it is rising in absolute terms.
What is the best RSI period?
Wilder suggested 14 periods, and it is the default almost everywhere. Shorter periods make the indicator jumpier, longer ones make it slower. Using the same period every time, and writing it down, matters more than the number itself.
Which index should you use for relative strength?
The one that answers your question. To see whether a stock is beating the US market, use a broad index such as the S&P 500; to compare it with its peers, use an index or ETF for its sector. Pick the benchmark before you look at the result and write it next to the chart.
Sources and further reading
- StockCharts ChartSchool: Relative Strength Index (RSI), calculation and Wilder smoothing
- Fidelity: RSI
- Fidelity: relative strength comparison
- Levy, “Relative Strength as a Criterion for Investment Selection,” The Journal of Finance, 1967
- Jegadeesh and Titman, “Returns to Buying Winners and Selling Losers,” The Journal of Finance, 1993
Use the same dates before comparing two charts
A stock chart taken from an old research article and today’s index chart do not form a valid relative-strength comparison. Choose the same observation window and consistent price data for both series. Write down the benchmark and whether your data adjusts for dividends and splits. Then interpret RSI separately; it measures the stock’s own recent price moves, not its performance against that benchmark. Fidelity’s indicator reference explains the price-ratio calculation. This date check matters whenever you move from a historical chart to a current watchlist.
