Momentum Investing Strategy in India: Lookback Periods, Rebalancing and Where It Fails

Momentum is the oldest, simplest idea in the market dressed up in a lab coat: stocks that have been going up tend to keep going up for a while. A momentum investing strategy in India takes that hunch and turns it into rules. You rank stocks by how strongly they’ve risen, buy the leaders, and swap them out on a fixed schedule. No story about the business, no target price. Just the tape.

That sounds too easy, and in a way it is. Momentum works until it suddenly doesn’t, and the way it fails is worth understanding before you put money behind it. In this guide we walk through how momentum is measured, how the NSE’s own momentum index builds its portfolio, how lookback periods and rebalancing change the result, and where the approach tends to break. We lean on the Nifty200 Momentum 30 factsheet from NSE Indices and Investopedia’s guide to momentum investing. Every stock and rupee figure in our worked examples is made up to show the arithmetic.

What momentum investing actually means

Investopedia describes momentum investing as buying securities that are already rising and selling them as the trend peaks, on the assumption that a trend in motion tends to stay in motion for some time. It points out that fund manager Richard Driehaus is often credited as the father of the approach, even though he wasn’t the first to use it.

There are two broad flavours, and people mix them up all the time.

• Time-series momentum looks at one stock or index on its own. Is it above where it was some months ago? Is its 50-day moving average above its 200-day? Investopedia gives that 50-day and 200-day crossover as a common signal: the shorter average crossing above the longer one is a buy, crossing back below is a sell.
• Cross-sectional momentum compares stocks against each other. You don’t ask “is this stock rising?” but “is it rising faster than the other 199 in my list?” Index products like the Nifty200 Momentum 30 work this way.

The raw measure is simple. Investopedia writes market momentum as M = V − V_x, where V is the latest price and V_x is the closing price x days ago. Most stock-ranking systems turn that into a percentage return over the lookback window so that a ₹100 stock and a ₹4,000 stock can be compared fairly. Our guide to the Rate of Change indicator covers that percentage version on a single chart.

Is there evidence behind it? Investopedia cites a 1993 study in the Journal of Finance which found that buying recent winners and selling recent losers produced significantly higher near-term returns than the overall US market between 1965 and 1989. It also notes that few professional managers use momentum on its own, preferring fundamental analysis such as discounted cash flows. So, real effect, plenty of sceptics.

How the Nifty200 Momentum 30 picks its stocks

The cleanest Indian template is the index NSE Indices runs. Its factsheet dated September 30, 2026 sets out the rules, and an HDFC Mutual Fund presentation on its Nifty200 Momentum 30 index fund spells out the scoring formula in more detail.

• Universe: stocks in the Nifty 200 at the time of review. The HDFC deck adds that a stock needs to be available in the F&O segment and have a listing history of more than one year.
• Score: each stock’s 6-month and 12-month price returns, each divided by its volatility. Volatility here is the annualised standard deviation of the stock’s daily (lognormal) returns over one year. That gives two “momentum ratios”.
• Standardise: each ratio is turned into a Z-score across the eligible stocks, then the two are blended 50:50 into a weighted average Z-score.
• Normalise: if the weighted Z-score is zero or positive, the Normalised Momentum Score is 1 plus that Z-score. If it’s negative, the score is 1 divided by (1 minus the Z-score), which keeps it above zero.
• Select and weight: the top 30 scores make the index. Each stock’s weight is its free-float market cap multiplied by its score, capped at the lower of 5% or five times the weight it would get on free-float market cap alone.
• Rebalance: twice a year, in June and December.

The factsheet also lists a base date of April 1, 2005, a base value of 1,000 and a launch date of August 25, 2020.

The volatility adjustment is the bit most DIY screens skip, and it matters. Without it, a momentum list in a hot market fills up with the wildest small names, simply because they moved the most.

A worked example: four stocks, one ranking

Let’s score four hypothetical stocks the index way. In the real index the Z-scores are calculated across the whole eligible universe. We’ll do it across just these four so the arithmetic stays on one screen.

Each ratio is just the return divided by volatility. Stock P: 24 ÷ 20 = 1.20 and 40 ÷ 20 = 2.00.

Now the Z-scores. The four 6-month ratios average 0.61 with a standard deviation of about 0.56. The 12-month ratios average 1.39 with a standard deviation of about 0.37. Subtract the average, divide by the standard deviation, blend 50:50, then normalise:

Look at Q. It has the biggest raw returns of the lot, 30% and 55%, yet it lands a distant second because it got there on a 40% volatility ride. P made less money but made it far more smoothly, and the score rewards that. R, with modest returns, isn’t far behind Q. S had a good year but a bad last six months, and the 6-month half of the score drags it to the bottom.

R’s score shows how the negative branch of the formula works. Its weighted Z-score is −0.09, so the score is 1 ÷ (1 + 0.09), roughly 0.92.

Weighting works the same way. Say P has a free-float market cap of ₹50,000 crore and Q has ₹80,000 crore. On market cap alone Q would carry more weight. Multiply by the scores, though, and P comes to 50,000 × 2.34 = 1,17,000 while Q comes to 80,000 × 1.10 = 88,000. The smaller, smoother stock now gets the bigger slice, subject to the caps.

A momentum score doesn’t ask how far a stock ran. It asks how far it ran for every unit of turbulence it put you through.

Choosing a lookback period

The lookback is the window over which you measure the return. There’s no magic number, only trade-offs.

• Short windows (around 3 months) catch new leaders early. They also catch every one-quarter wonder, a single result-day gap or an operator-driven spike. Expect more churn.
• Medium windows (around 6 months) are a middle path. Leaders need to hold up through a couple of result seasons to rank well.
• Long windows (around 12 months) are steadier and slower. By the time a stock tops a 12-month list, a good part of its move may already be behind it.

The NSE index hedges this by blending 6 and 12 months equally. That’s a sensible default for anyone building their own screen. If you go shorter, expect more trades. If you go longer, expect to be late on both entry and exit.

One more practical point. Use the same lookback for every stock, and decide it before you look at results. Tweaking the window until a backtest looks pretty is the fastest way to build a strategy that only works on the past.

How often to rebalance, and what it costs

Rebalancing is where theory meets brokerage bills. Every review sells the names that dropped out and buys the new entrants, and each trade carries brokerage, statutory charges and slippage. Here’s a rough illustration using assumed numbers, not any broker’s actual charges.

Take a ₹10 lakh momentum portfolio and assume all-in trading costs of 0.5% of the value traded.

• Semi-annual review, 40% of the portfolio replaced each time: ₹4 lakh sold plus ₹4 lakh bought is ₹8 lakh traded. At 0.5%, that’s ₹4,000 per review, or ₹8,000 a year, about 0.8% of the portfolio.
• Monthly review, 15% replaced each time: ₹1.5 lakh sold plus ₹1.5 lakh bought is ₹3 lakh traded, ₹1,500 per review and ₹18,000 a year, about 1.8% of the portfolio.

Monthly feels more responsive, and sometimes it is. But in this example it costs more than twice as much a year, before any tax on gains booked along the way. Frequent selling also means more gains booked within a year of buying, so check the current capital gains rules with a tax adviser before you pick a schedule.

Two habits help here. First, use a buffer: don’t sell a holding just because it slipped from 30th to 34th. A simple way to cut pointless churn is to sell only once a stock falls well outside the top group. Second, keep the schedule fixed. Rebalancing “when it feels right” quietly turns a rules-based system back into gut trading. Our pillar on risk management in trading covers position sizing, which matters just as much once you hold 20 or 30 names.

Where a momentum investing strategy in India fails

This is the part most momentum pitches skip.

It lags turning points. A score built on 6 and 12 months of returns will still be full of last year’s winners when the market turns. With a semi-annual review, the portfolio can sit in faded leaders for months. The factsheet’s own numbers show how the strategy behaves through rough patches. As of September 30, 2026, the Nifty200 Momentum 30 price return index was down 5.23% year-to-date and 1.09% over one year, against a since-inception compound annual return of 17.04% and a five-year CAGR of 8.32%.

It concentrates. Momentum piles into whatever has been working. On the same factsheet, Capital Goods (24.11%), Financial Services (21.17%), Metals & Mining (18.66%) and Power (14.19%) together make up about 78% of the index. Tata Steel, Hindalco, Vedanta, NTPC and Adani Power all sit in the top ten. That’s not a diversified portfolio. It’s a bet on the current leadership.

It’s more volatile than the broad market. The factsheet puts the index’s one-year beta against the Nifty 50 at 1.18 and its one-year annualised standard deviation at 18.60%. A beta above 1 means bigger swings than the Nifty in both directions, on average.

It reverses hard. When a crowded trend breaks, everyone holding the same leaders heads for the exit together. Investopedia lists sudden price drops and market swings as the main risks, and warns that momentum traders need to watch liquidity and trading volume. In mid and small caps that warning bites, because the exit door is narrow.

It ignores the business. A stock can rank high because of a one-off order, a sector re-rating or plain speculation. The score can’t tell the difference.

Momentum doesn’t fail slowly. It works for a long stretch, then gives a chunk of it back in a few ugly months, usually when the leaders everyone owns all turn at once.

Building a simple momentum checklist

If you want to try this yourself rather than through an index fund, keep it boring and written down.

1. Fix the universe. A liquid list such as the Nifty 200 keeps you away from stocks you can’t exit. The NSE index also requires F&O availability and a year of listing history.
2. Fix the score. Blend 6- and 12-month returns and divide by volatility, as the index does. Write the formula down.
3. Fix the size of the portfolio. Twenty to thirty names spreads single-stock risk without turning you into a fund manager.
4. Fix the calendar. Pick review dates in advance and stick to them. Add a buffer rule so you aren’t trading every small rank change.
5. Cap single positions and watch sector piles. The index caps a stock at 5%. Decide your own cap, and notice when three sectors make up most of your book.
6. Decide what you do in a falling market. Some investors pause new buying when the broad index trades below a long moving average, the kind of 200-day signal Investopedia describes. Others just ride it out. Choose before the fall, not during it. A price-channel tool like the Donchian channel is another way to define “trend broken” in advance.

The bottom line

Momentum is one of the few market ideas that has both a long research record and a ready-made Indian template in the Nifty200 Momentum 30. The rules are clear: rank on risk-adjusted 6- and 12-month returns, hold the top names, rebalance on a schedule. The hard part is behavioural. You’ll own concentrated, expensive-looking leaders, pay real trading costs, and sit through stretches where the strategy trails the plain Nifty. If you can live with that and keep to the rules, momentum can earn its place in a portfolio. If you’re likely to abandon it after one bad half-year, you’ll probably get the costs without the benefit.

Disclaimer: This article is for information and education only and is not investment advice or a recommendation to buy or sell any security, fund or index product. Stocks P, Q, R and S, and every return, volatility, market cap, portfolio value and cost figure in the worked examples, are hypothetical and used only to show the calculations. Index figures are taken from the NSE Indices factsheet dated September 30, 2026, and past performance does not indicate future returns. Investing involves risk, including the loss of capital. Please consult a SEBI-registered investment adviser before making investment decisions.