Relative strength, and choosing what to own
The answer
Relative strength is one price divided by another. You divide your stock's price by an index's price and watch which way the result moves. When the line rises your stock is beating the index, and when it falls your money would have done better in the index instead.
Why this costs you money
You know which of your holdings are down. You almost certainly do not know which of them are losing to the market.
Those are different lists, and the second one is the one that matters. A stock that is up 6% in a year when the index is up 24% has not made you money. It has cost you 18 points against the simplest, cheapest thing you could have owned instead. On a holding of Rs 2,00,000 that is Rs 36,000 that quietly went somewhere else. Your app shows the position in green. Nothing warns you.
This is where a portfolio goes wrong. Not through disasters, which are obvious and get dealt with. Through a slow accumulation of positions that are neither bad enough to sell nor good enough to keep.
The specific mistake is this. People measure a holding against their own purchase price instead of against the market. Your purchase price is a fact about you. It has no effect on anything the company or the market will do next. Measuring against it produces exactly one decision — wait until it comes back — and that decision is how dead positions survive for years.
Relative strength replaces your purchase price with a benchmark that keeps moving. It is a 5-minute check and it usually finds 2 or 3 holdings you have been keeping out of hope.
How it works
Take your stock's closing price. Divide it by the index's closing price on the same day. Plot that number over time.
The level of the line means nothing. It depends on the units of both prices, and comparing the level of one ratio line to another tells you nothing at all. Only the direction carries information.
- The ratio line rises: your stock is outperforming the index.
- The ratio line falls: your stock is underperforming the index.
- The ratio line is flat: your stock is tracking the index, and you are taking single-company risk for index returns.
That third case is the one people miss. A flat ratio line over 2 years means you took the risk of one company going wrong and received, in exchange, what a diversified fund would have paid you for free.
The simpler version, which is enough
You do not need to plot anything. For ranking, subtraction works. Take the percentage change in your stock over a fixed window, take the percentage change in the index over the same window, and subtract. The result is called relative performance, and it is a single number you can sort a list by.
Use more than one window, because a single window will mislead you.
| Window | What it captures |
|---|---|
| 3 months | Recent leadership, and a lot of noise |
| 6 months | The usual working window for ranking |
| 12 months | Established leadership, slow to change |
A stock that is strong on all 3 windows is in a real trend. A stock that is strong on 3 months and weak on 12 is a bounce inside a decline, which is the single most common trap in this whole subject.
Relative strength is not RSI
These are 2 different things with almost the same name.
Relative strength compares one instrument to another. Your stock against the index.
RSI, the Relative Strength Index, compares an instrument to its own recent history. It measures the size of recent gains against the size of recent losses in the same stock. It never looks at the index at all.
RSI asks whether a stock has moved a lot lately. Relative strength asks whether owning it beat owning the market. Say the full name of whichever one you mean.
Top-down: market, then sector, then stock
Money does not arrive in a market evenly. It concentrates. In most periods a small number of sectors carry most of the return, and inside those sectors a small number of companies carry most of the sector. That gives an order of work.
- The market. Is the broad index in an uptrend? If it is not, expect fewer candidates and smaller positions. Relative strength in a falling market identifies the stocks falling least, which is useful information and is not the same as a reason to buy.
- The sector. Rank the sector indices by 6-month relative performance against the broad index. This narrows thousands of companies to a handful of areas in about 4 minutes.
- The stock. Inside the 2 or 3 leading sectors, rank the companies by the same measure.
This is sector rotation — leadership moving between sectors over months and years rather than staying still. You do not need a theory of why it moves. You need to notice that it has.
Building a watchlist you can actually use
Ranking produces a long list. A list is not a watchlist. Three filters turn one into the other.
- Liquidity. Divide your intended position by the average daily volume. If it is over 5%, exiting will take days and will move the price against you. Strong relative strength in a stock you cannot sell is a trap. This filter removes more names than any other.
- Its own trend. A stock can outperform a falling index while falling itself. Check that it is above its own rising moving average, or making higher lows, before treating strength as a buy signal.
- A known reason to be wrong. If you cannot name a price at which you would admit the idea failed, the name does not go on the list.
What it tells you, and what it does not
Relative strength tells you where money has been going. That is a fact about the past and it is a genuinely useful fact, because flows into an asset tend to persist for months.
It does not tell you why. A stock can lead because earnings are accelerating, or because index funds are buying it after an inclusion, or because a small number of participants are pushing it. The ratio line looks the same in all 3 cases.
It does not tell you valuation. The strongest stock in a ranking is often the most expensive. Relative strength and cheapness are close to opposite measures, and a method that uses only one of them is taking a known risk.
It does not survive a turn. Momentum works for months and then reverses violently, usually when the market as a whole changes direction. Daniel and Moskowitz described these momentum crashes — the strategy's worst periods come immediately after a market bottom, when the previous losers rebound hardest. Ranking by strength puts you on the wrong side of that.
And it is sensitive to the window. The academic literature usually skips the most recent month, because very short-term moves tend to reverse. If you rank on the last 4 weeks you will systematically buy what is about to fall back.
The decision rule
Rank against the index, not against your purchase price.
For anything you already own:
- Outperforming over 6 and 12 months: the thesis is working. Leave it alone.
- Underperforming by less than 5 points over 6 months: noise. No action.
- Underperforming by more than 10 points over 6 months: write 1 sentence
saying why you still own it. If the sentence contains "will recover", "eventually" or "my price", you are holding it out of hope.
For anything you are thinking of buying: strong on 6 and 12 months, in a sector that is also strong, and small enough against daily volume that you could leave.
Unless the market has just turned upward after a long fall. In that one condition the ranking inverts for a period, and the previous leaders lag while the previous losers run.
Try this now
Five minutes, on your own holdings, and most people find something they did not want to find.
- Open your holdings list. Write down each stock and its price today.
- For each one, find the price 6 months ago. Every broker app and every free chart site shows a 6-month chart with a starting value. Compute the percentage change:
(today − then) ÷ then × 100. - Do the same for your home index over exactly the same 6 months. Use the Nifty 50 or the Sensex in India, and the S&P 500 in the United States.
- For each holding, subtract the index's percentage change from the stock's. You now have a single number per holding.
- Sort the list from highest to lowest.
- For every holding more than 10 points below the index, write 1 sentence saying why you still own it. Write it down. Do not think it.
What you should see. The list will not be even. Typically 2 or 3 holdings are doing all the work, several are close to the index, and 2 or 3 are a long way below it. The last group is the point of the exercise.
Read your own sentences back. The ones that describe the business — a new plant, a contract, a margin recovery you are waiting for — are theses, and a thesis is allowed to take time. The ones that describe the price are not. "It will come back" and "I am down too much to sell now" are the same sentence, and neither is about the company.
Then add up the value of the holdings in the bottom group. That total is how much of your money is allocated to positions losing to a fund you could buy in 1 click.
Three real cases
1. The energy sector in the United States, calendar year 2022 — relative strength that was visible for months 2022 was a bad year for American shares. The S&P 500 fell sharply across the year. Energy was the exception and rose strongly, finishing as the only major sector with a large positive return. The important detail is not the size of the gap. It is that the energy sector's ratio line against the S&P 500 had been rising since early 2021, more than a year before the headline year. A reader ranking 11 sector ETFs once a month would have seen it near the top for many consecutive months.
2. Nifty PSU Bank against Nifty IT, calendar year 2022 (India) — rotation inside one market In the same year the 2 ends of the Indian sector table swapped places. Nifty IT fell heavily as global technology spending slowed, while Nifty PSU Bank rose strongly as bank profits recovered. Both indices are published daily by the NSE and are free. An investor holding technology shares could have seen relative performance turn negative early in the year. Many did not look, because the holdings were still profitable positions from 2020 and 2021, and a profitable position does not feel like a problem.
3. Silicon Valley Bank, March 2023 (United States) — the case where relative strength gave no useful warning SVB Financial Group's bank subsidiary was closed by regulators on 10 March 2023 after a deposit run. Signature Bank was closed on 12 March 2023. The collapse took days. Regional bank shares as a group had been weak against the S&P 500 before this, but that weakness was shared by dozens of banks, most of which did not fail. Ranking would have told you the sector was lagging. It would not have told you which institution was about to disappear, and it gave no warning of the speed. This case is here on purpose. Relative strength is a ranking tool for gradual flows. It has nothing to say about a sudden failure.
The question that resolves it
A novice asks: is this stock going up?
An expert asks: is it going up faster than the thing I could own instead for almost no cost?
The second question has a free benchmark, an unambiguous answer and no room for a story. That is why it is the useful one.
What would make this wrong
If ranking shares by their past 6 to 12 month performance produced no difference in what happened next, relative strength would be worthless.
That claim has been tested many times. Jegadeesh and Titman documented in 1993 that buying past winners and selling past losers produced abnormal returns in United States data from 1965 to 1989. The effect has since been found in other countries and other decades. It is one of the most replicated results in finance.
Here are the honest limits.
The effect is modest and it is shrinking. Published anomalies tend to weaken once they are published and traded. Momentum has behaved that way. It is a small edge, not a system.
It reverses at turning points, and it reverses hard. The worst outcomes come right after a market low. A ranking system with no market filter will hand you its largest loss at the moment you can least afford it.
Costs eat a real share of it. Ranking implies turnover, and turnover is expensive on both sides in both countries. Much of the measured momentum effect in academic studies also sits in small, illiquid companies, which are exactly the companies where your own order moves the price. That is the difference between a number in a study and a number in your account.
In India
The NSE publishes a full set of sectoral indices, and they are the raw material for this method. Nifty Bank, Nifty IT, Nifty Auto, Nifty FMCG, Nifty Pharma, Nifty Metal, Nifty PSU Bank, Nifty Realty, Nifty Energy and others. Values are free on the exchange website and on every chart platform. Ranking them by 6-month change against the Nifty 50 takes a few minutes.
Three Indian details change how the results should be used.
The benchmark is concentrated. The Nifty 50 is weighted by free-float market value, and financial services alone is a very large share of it. So "relative strength against the Nifty 50" is largely relative strength against a small number of very large companies. On a day when 1 heavyweight moves 4%, every other stock's relative performance shifts for reasons unrelated to any of them.
You often cannot buy the sector. Sector ETFs exist in India, but the range is narrow and several trade thinly, with wide spreads and prices that drift from the value of the underlying holdings. An Indian ranking usually cannot be acted on directly. You rank sectors, then pick companies inside the winning sector, which reintroduces single-company risk the sector ranking never measured.
Surveillance rules distort the picture. A stock placed under the Additional Surveillance Measure or Graded Surveillance Measure framework can be moved to 100% margin or to trade-for-trade settlement. Its volume collapses and its price behaviour changes for administrative reasons, not economic ones. Check the exchange's surveillance list before you trust a ranking on a small company.
Costs matter too. Securities transaction tax applies on the sell side of a delivery trade, and short-term capital gains on listed equity are taxed at a higher rate than long-term gains, where long-term means a holding period of more than 12 months. A ranking method with high turnover pays that difference repeatedly.
In the United States
The United States has the cleanest version of this method anywhere, because the sector conclusion is directly purchasable.
Shares are classified into 11 sectors under the Global Industry Classification Standard: energy, materials, industrials, consumer discretionary, consumer staples, health care, financials, information technology, communication services, utilities and real estate. Each has a large, liquid, low-cost ETF. Ranking 11 of them by 6-month performance against the S&P 500 takes 3 minutes and can be executed the same afternoon at a spread of a cent or two.
That changes the character of the exercise. An American investor can hold a sector view without holding a company view. The ranking says energy is leading; the investor buys the energy sector and takes no position on which energy company survives. This removes a large source of risk that Indian investors cannot remove the same way.
Breadth is measurable through relative strength here too. Comparing an equal-weighted version of the S&P 500 to the standard capitalisation-weighted version tells you whether the average share is participating or whether a handful of very large companies are carrying the index. When that ratio falls for months, "the market is up" is a statement about a few companies.
Where they differ, and what that tells you
The difference is not in the analysis. It is in whether the analysis can be executed.
In the United States, a relative strength ranking of sectors ends in a purchasable instrument. The reader ranks 11 ETFs, buys the top 2, and owns exactly the thing the ranking identified. The risk taken matches the risk measured.
In India, the same ranking ends in a list of sectors and a second problem. Sector ETFs are limited and often thin, so the next step is to pick 2 or 3 companies from inside the leading sector. The moment you do that, you are taking company risk that the sector ranking never examined. The sector can be right and the company can still lose money.
What that tells you is practical. An Indian investor needs an extra step that an American investor does not: after choosing the sector, check that the company is also strong against its own sector index, not only against the Nifty 50. Two ratio lines, not one. Stock against sector, and sector against market. If a stock lags its own sector while the sector leads the market, you have bought the sector's weakest member and the single ranking told you the opposite.
There is a second difference. The American benchmark holds 500 companies across 11 sectors. The Indian benchmark holds 50, weighted heavily toward financials. Relative performance against a narrow benchmark is a noisier measurement. Indian readers should therefore act only on large gaps. A 3-point gap over 6 months against the Nifty 50 is close to nothing. A 20-point gap is information.
Carry this
- Relative strength is your stock's change minus the index's change over the same window. Nothing more complicated is needed.
- Direction of the ratio matters. Level of the ratio means nothing.
- Rank on 6 and 12 months. Ignore the last 4 weeks, which mostly reverse.
- A holding more than 10 points below the index for 6 months needs a written sentence, and the sentence cannot be about your purchase price.
- Strong and illiquid is not an opportunity. Check volume before you check anything else.