What Is Momentum?

Momentum as a strategy was first documented by Jegadeesh and Titman in 1993. Their finding was simple but powerful — stocks that performed well over the past 12 months continued to outperform over the next 3 to 6 months. This was not a small or fleeting effect. It was statistically significant, economically meaningful, and persistent across decades of data..

The effect has been observed across asset classes — equities, bonds, commodities, and currencies — and across geographies from the US to India to emerging markets. What began as an academic curiosity in the early 1990s has since become one of the most widely accepted and exploited anomalies in quantitative finance.

What makes momentum particularly interesting is that it should not exist in a truly efficient market. If all information is priced in immediately, there should be no reason for yesterday's winners to keep winning. Yet they do — and they have, consistently, for over three decades of out-of-sample evidence. The persistence of momentum is itself evidence that markets are not fully rational, and that human behaviour leaves exploitable patterns in price data.

At its core, momentum is a bet on continuation. It says that trends, once established, are more likely to continue than to reverse — at least over the medium term of 3 to 12 months. Beyond that horizon, evidence suggests that long-term reversal takes over, and yesterday's winners eventually become tomorrow's mean-reversion candidates.

Understanding momentum is therefore not just about finding a trading signal. It is about understanding the deeper mechanics of how prices move, how information diffuses through markets, and how collective investor psychology creates patterns that persist long enough to be traded profitably.

Why Does Momentum Work?

Markets are not fully efficient. Price trends persist because human behaviour creates predictable patterns — anchoring, herding, and slow information diffusion all contribute to momentum. These are not random glitches in the system. They are systematic, repeatable biases that show up across markets, time periods, and investor types.

When a company reports better-than-expected earnings, the stock price rises. But not all investors react at the same speed. Institutional investors with dedicated research teams update their models within hours. Retail investors, reading about it days later, update more slowly. This staggered reaction means the full information is not priced in immediately — the stock continues to drift upward for weeks as more participants catch up. This is the information diffusion channel of momentum, and it is one of the most robust explanations in the academic literature.

Anchoring Bias

Investors anchor to recent prices and are slow to update their views when new information arrives. A stock reporting strong earnings continues to be undervalued for weeks as the market slowly digests the news.

Herding Behaviour

As a stock rises, more investors notice and pile in. This self-reinforcing cycle pushes prices further in the direction of the initial move — until it overshoots. The momentum trade works precisely because of this overshoot. By the time the broad market fully recognises the strength of a trend, the early momentum investor has already built a position and is riding the wave of late entrants pushing prices higher.

This dynamic is not unique to individual stocks. It plays out across sectors, indices, and even entire asset classes. When global investors rotate into Indian equities, the initial buyers attract more buyers. When a commodity enters a bull cycle, the trend feeds on itself as more speculators, hedgers, and passive allocators increase exposure. The mechanism is the same — a price signal attracts attention, attention attracts capital, and capital drives prices further in the same direction.

Markets are not fully efficient. Price trends persist because human behaviour — anchoring, herding, slow information diffusion — creates momentum that can last months.

Momentum in Indian Markets

In Indian markets, momentum has historically worked well in mid and small cap segments. The Nifty 200 Momentum 30 Index has outperformed the Nifty 200 by over 6% annually over a 10-year period. This is a substantial and consistent alpha — not a one-off result driven by a single lucky year, but a structural outperformance that has persisted through multiple market cycles, regime changes, and macro shocks.

The outperformance is concentrated in the mid and small cap universe because these segments have more inefficiency. Analyst coverage is thinner, institutional participation is lower, and information diffuses more slowly. This creates fertile ground for momentum — prices take longer to fully reflect new information, and trends persist for longer before mean reversion sets in. In large caps, where every major institution has a research team covering every stock, the momentum effect is present but weaker, as prices adjust more quickly to new data.

However, momentum strategies are prone to sharp drawdowns during market reversals. The COVID crash of March 2020 saw momentum portfolios lose 40% in weeks — a phenomenon known as a momentum crash. What makes these crashes particularly painful is their speed and their counterintuitive nature. Momentum portfolios, by construction, are concentrated in stocks that have recently performed well. When the market reverses, these are often the stocks that sell off the hardest, as investors rush to lock in gains and reduce risk. The same crowding that amplified returns on the way up becomes a liability on the way down.

Practical Implementation

The strategy requires monthly rebalancing, disciplined position sizing, and strict risk management. Transaction costs can erode returns significantly if rebalancing frequency is too high. This is one of the most underappreciated aspects of momentum investing — the gross returns look compelling in backtests, but the net returns after accounting for brokerage, impact cost, and taxes tell a more sobering story. For retail investors trading through a standard brokerage account, frequent rebalancing can eat 1 to 2 percentage points of annual return, meaningfully reducing the edge.

Position sizing is equally critical. A common mistake is to run a concentrated momentum portfolio — betting heavily on the top 5 or 10 names — in pursuit of maximum returns. In calm markets this works. But when a momentum crash arrives, concentration becomes catastrophic. The standard approach among systematic funds is to equally weight positions across 20 to 30 names, accept slightly lower returns in good years, and in exchange gain meaningfully better drawdown protection when the cycle turns. For most investors, a rules-based, diversified implementation through a momentum index fund is more practical and more robust than attempting to run a concentrated stock-picking version of the strategy.

When Momentum Breaks

Momentum crashes happen fast and without warning. They typically occur during sharp market reversals — when the market regime shifts suddenly from risk-on to risk-off. Unlike a slow fundamental deterioration that gives value investors time to reassess, a momentum crash is a discontinuous event. One day the portfolio is working, the next it is down 10% with no obvious catalyst. By the time most investors recognise what is happening, the damage is already done.

The speed of momentum crashes is a direct consequence of crowding. When many participants run similar momentum strategies — buying the same winners, avoiding the same laggards — the portfolio becomes fragile. Any trigger that causes a simultaneous exit, whether a macro shock, a liquidity event, or simply a sentiment shift, creates a cascade. Sellers beget more sellers. Stop-losses trigger automatically, adding fuel to the decline. What begins as a modest correction can become a rout within days.

Risk management is therefore non-negotiable. Always size momentum positions with defined stop-losses and never concentrate more than 5% in a single name. These rules feel conservative during a bull run when momentum is working and every position seems to go up. But they are not designed for the good times — they are designed for the moment when everything breaks at once. A 5% cap per position means that even a complete loss in one name costs the portfolio 5% — painful, but survivable. Without a cap, a single concentrated bet gone wrong can permanently impair capital in a way that no subsequent recovery can fully repair.

Terms

Momentum (12-1) — The standard momentum signal, measured as a stock's return over the past 12 months excluding the most recent month. The exclusion of the last month removes the short-term reversal effect that would otherwise dilute the signal.

Cross-Sectional Momentum — A relative strategy that ranks stocks against each other and goes long the top decile while avoiding (or shorting) the bottom. The signal is about relative strength within a universe, not absolute direction.

Time-Series Momentum — An absolute strategy that asks whether an asset is trending up or down relative to its own history. Used more in multi-asset portfolios and trend-following hedge funds than in equity-only strategies.

Information Diffusion — The speed at which new fundamental information is absorbed into a stock's price. Slower diffusion means longer, more exploitable momentum trends — the primary reason mid and small caps show stronger momentum than large caps.

Momentum Crash — A rapid, severe drawdown specific to momentum portfolios, typically triggered by a sudden market regime change. Characterised by a simultaneous unwind of crowded long positions, producing losses far larger and faster than standard drawdown models predict.

Crowding — When many systematic participants hold similar positions. Amplifies returns during trending markets and amplifies losses during reversals. The primary source of tail risk in momentum strategies.

Factor Decay — The gradual erosion of a factor's alpha as more capital chases the same signal. Momentum has shown resilience to factor decay compared to value and size, partly because the behavioural biases driving it are structural.

Rebalancing Drag — The performance cost of frequently turning over a portfolio — brokerage, market impact, and taxes. For momentum strategies with monthly rebalancing, this can consume 1–2% of gross annual return for a retail account.

Equal Weighting — Allocating the same capital to each position regardless of signal strength or market cap. Standard practice in institutional momentum portfolios to avoid concentration risk without sacrificing diversification.

SB Finance Findings

Indian mid and small cap momentum is structurally stronger than large cap. NSE data from 2014–2025 confirms the academic finding with greater magnitude than developed markets. The Nifty 200 Momentum 30 has delivered alpha in the range of 6–8% annually over the Nifty 200 across most rolling 10-year windows in this period — with the spread widening during the 2020–2024 bull cycle. The same signal applied to the Nifty 50 universe produces materially weaker alpha, net of costs. The gap is explained by analyst coverage density: mid and small cap stocks have thinner institutional research, slower price discovery, and longer-lasting momentum trends before mean reversion sets in.

Momentum crashes in India cluster around three identifiable regime types. Reviewing drawdowns exceeding 20% in momentum portfolios since 2008, we find they occur almost exclusively during: (1) sudden global risk-off shocks — GFC 2008, COVID March 2020; (2) domestic policy discontinuities — demonetisation November 2016; and (3) sharp sector rotation driven by abrupt FII flow reversals. The regime type matters more than the drawdown depth. Global risk-off crashes have historically recovered within 9–14 months for a patient holder. Domestic policy shocks have taken longer and have more frequently required a deliberate re-entry signal rather than a hold-through.

Monthly rebalancing is the practical optimum for Indian retail accounts. Weekly rebalancing erodes a meaningfully larger share of gross alpha through Securities Transaction Tax (STT), brokerage, and market impact — costs that are structurally higher in India than in the US-centric academic literature on momentum. The NSE's own Nifty factor index methodology uses monthly reconstitution for the same reason. For most retail investors, aligning with this cadence eliminates unnecessary cost drag without sacrificing signal quality.

The 5% single-name cap is not a conservative choice — it is the structurally optimal one. Concentrated momentum portfolios (top 10 names) produce higher peak returns in trending markets but carry maximum drawdowns roughly 20–25 percentage points deeper than diversified constructions (top 25–30 names, equal-weighted). For a strategy where the compounding edge accrues over years, not months, a smaller but more survivable drawdown is strictly better for most holding horizons beyond three years. The math of recovery — a 50% drawdown requires a 100% gain to break even — makes concentration a poor trade even when the gross return looks attractive.

Conviction score alone is an incomplete signal — trend duration is the missing variable. Stocks entering the top momentum decile for the first time after six or more months of absence carry significantly different risk profiles than stocks that have resided in the top decile for three or more consecutive months. The latter group carries elevated crowding risk: many systematic participants hold the same names, and a simultaneous exit triggers amplified drawdowns. This distinction — fresh momentum versus mature momentum — is central to how Signal weights and classifies trend strength.

Conclusion

Momentum is real, persistent, and exploitable — but it requires discipline. For retail investors, the simplest exposure is through index funds tracking the Nifty 200 Momentum 30, available via several AMCs at low expense ratios. This removes the execution complexity, eliminates individual stock risk, and ensures systematic rebalancing without emotional interference.

The factor has survived decades of scrutiny, multiple market cycles, and the attention of thousands of quantitative researchers trying to arbitrage it away. It persists because its source — human behavioural bias — is not going away. As long as investors anchor, herd, and react slowly to new information, momentum will continue to offer a return premium to those patient and disciplined enough to harvest it systematically.