What Is Momentum Trading?

Most investment strategies are built on one of two premises. Value investing says: find assets that are cheap relative to their intrinsic worth, buy them, and wait for the market to recognize their value. Growth investing says: find companies growing faster than the market expects, pay a premium, and let compounding do the work.

Momentum investing starts from a different premise entirely. It does not ask whether a stock is cheap or whether the company is growing. It asks: which stocks have been going up, and is that upward movement likely to continue? The core observation is simple – stocks that have outperformed over the past three to twelve months tend to continue outperforming over the following three to twelve months. Stocks that have underperformed tend to continue underperforming.

This seems to violate everything we've been taught about efficient markets. If markets are rational and prices reflect all available information, how can past performance predict future performance? The answer lies in the messy, deeply human psychology of markets – and in a body of academic research that has spent three decades trying to understand why momentum exists and whether it can be systematically exploited.

FinAi lens: momentum is not a feeling. It is a measurement. FinAi helps traders define, rank, and monitor momentum so they are less likely to confuse disciplined participation with emotional chasing.

Defining Momentum: What It Is and Is Not

Momentum, in the financial sense, refers to the tendency of assets that have recently performed well to continue performing well in the near term, and assets that have recently performed poorly to continue performing poorly. It is a statistical observation about price behavior – not a theory about why companies are worth what they're worth.

This is what makes momentum fundamentally different from value or growth investing. A value investor cares deeply about a company's earnings, assets, and competitive position. A momentum investor cares primarily about relative price performance. The same stock could be a buy on momentum grounds and a sell on fundamental grounds – or vice versa. The two frameworks ask different questions and sometimes give conflicting answers.

Momentum is also not the same as chasing hot stocks or following the herd blindly. Disciplined momentum investing involves systematic rules: look back windows (typically 3-12 months), rebalancing periods, risk controls, and explicit exit criteria. Done well, it is a rules-based strategy. Done carelessly, it is a recipe for buying at the top.

Momentum is not a feeling – it is a measurement. The discipline lies in defining it precisely, applying it consistently, and knowing when to step aside.

The Academic Evidence

Momentum's strongest claim to legitimacy is not in trader lore but in peer-reviewed academic research. The evidence base is unusually robust for a market anomaly – spanning decades, geographies, and asset classes.

Jegadeesh & Titman (1993): The foundational paper on stock momentum. Studied U.S. stocks from 1965-1989 and found that buying the top decile of 6-month performers and shorting the bottom decile produced abnormal returns of approximately 1% per month – about 12% per year – over the following 3-12 months. The paper launched decades of follow-on research and remains the most cited work in momentum literature.

Fama & French (1996): Eugene Fama, the father of the Efficient Market Hypothesis, acknowledged momentum as the 'premier anomaly' in factor investing – the one unexplained pattern that his three-factor model could not account for. This was a remarkable concession from the most prominent defender of market efficiency.

Asness, Moskowitz & Pedersen (2013): Extended momentum research to 40+ years of data across stocks, bonds, currencies, and commodities in 18+ countries. Found that momentum worked in virtually every market and asset class tested. Called momentum a pervasive phenomenon that is 'everywhere and in everything.'

Israel & Moskowitz (2013): Showed that momentum returns persisted even after accounting for transaction costs in realistic implementation – a critical test for any strategy's practical viability. Found that long-only momentum (buying winners without shorting losers) captured most of the theoretical return.

The consensus from this body of research: momentum is real, persistent, and not easily explained away. But it is also imperfect, volatile, and subject to sharp, painful reversals at unpredictable intervals.

The Behavioral Finance Explanation

If momentum works, why does it work? The Efficient Market Hypothesis says it shouldn't. The answer that has accumulated the most explanatory power comes from behavioral finance – the study of how psychological biases cause investors to deviate systematically from rational behavior. Several well-documented biases conspire to create and sustain momentum.

Anchoring and Underreaction

What it is: Investors anchor to prior price levels and adjust their views too slowly when new information arrives. When a company reports better-than-expected earnings, the market initially moves prices upward – but not enough. The adjustment happens gradually over weeks and months as more investors process the information.

Effect on markets: Creates the initial momentum effect: good news gets incorporated slowly, meaning prices continue rising after the initial reaction. Buying recently strong stocks captures this gradual re-pricing.

The Disposition Effect

What it is: Investors are psychologically predisposed to sell winning positions too early (to lock in gains) and hold losing positions too long (to avoid realizing a loss). This behavior was documented by Shefrin and Statman in 1985 and has been replicated across retail investors worldwide.

Effect on markets: Suppresses the price rise of winning stocks in the short term, creating undervaluation of winners and overvaluation of losers – and setting up the subsequent momentum move as these distortions eventually correct.

Herding

What it is: Investors observe the actions of others and imitate them, especially under uncertainty. When a stock is rising, it attracts attention; attention attracts new buyers; new buyers push prices higher; higher prices attract more attention. A self-reinforcing feedback loop develops.

Effect on markets: Amplifies and extends momentum beyond what fundamentals alone would justify. The same herding dynamic operates in reverse for falling stocks. Herding is why momentum can persist for months – and why it can reverse violently.

Overconfidence and Self-Attribution Bias

What it is: After a series of winning trades, investors attribute success to their own skill rather than favorable market conditions. They become overconfident, take larger positions, and push winning sectors and stocks further upward.

Effect on markets: Pushes trends further than fundamentals justify – creating the overshoot that eventually reverses in a momentum crash. The same overconfidence that fuels late-stage momentum also makes traders particularly vulnerable to the crash.

Two Types of Momentum

Academic and practitioner research has identified two distinct forms of momentum, each with different characteristics and implementation approaches. Understanding the distinction matters for how you would actually trade each one.

Cross-Sectional Momentum (Relative Momentum)

Cross-sectional momentum compares the recent performance of different assets against each other. You rank all stocks in a universe by their trailing returns – say, the past 12 months excluding the most recent month (excluding the last month avoids short-term reversal effects, a well-known quirk of the data). You then buy the top performers (the 'winners') and either avoid or short the bottom performers (the 'losers').

This is the form of momentum studied by Jegadeesh and Titman. It is relative – a stock can have negative absolute returns and still be a momentum buy, if it fell less than everything else in its universe. Cross-sectional momentum is widely used by quantitative hedge funds and factor-based ETFs.

Time-Series Momentum (Absolute Momentum)

Time-series momentum asks a simpler question: has this specific asset been going up or going down? If the asset's trailing return is positive, hold it or buy it. If it is negative, avoid it or go short. The benchmark is not other assets but zero – has this asset been rising in absolute terms?

Gary Antonacci, in his book 'Dual Momentum Investing,' popularized a practical version that combines both: first use absolute momentum to decide whether to be in equities or cash (if equities have negative 12-month returns, go to safety), then use relative momentum to select which equities to hold. This combination has shown strong risk-adjusted returns with meaningful drawdown protection.

For a beginner, the practical takeaway: cross-sectional momentum tells you what to own within a universe; time-series momentum tells you whether to be in that universe at all. Both are tools that belong in a momentum investor's toolkit.

Cross-sectional momentum says: own the strongest horses in the race. Time-series momentum says: only race when conditions are favorable.

The Dark Side: Momentum Crashes

No strategy that has worked as reliably as momentum has done so without periods of severe, stomach-churning losses. Momentum crashes are a real and documented feature of the strategy – and understanding them is as important as understanding the long-term returns.

Momentum crashes tend to occur during sharp market reversals – particularly at the end of bear markets. When markets fall sharply, the stocks with the most negative momentum (the 'losers') are often highly beaten-down, cheap, and heavily shorted. When sentiment turns and a recovery begins, these beaten-down stocks bounce dramatically – often far more than the market average. Meanwhile, the recent 'winners' – which may have been defensive or short-side positions – get sold off.

The result: at the exact moment of market recovery, momentum portfolios can suffer catastrophic losses. The research documents momentum crashes of 40-50% or more in short periods during major market transitions – 1932, 2001, 2009. These crashes are infrequent but severe enough to wipe out years of accumulated gains if not managed carefully.

Managing Momentum Risk

Experienced momentum practitioners have developed several techniques to reduce crash risk without giving up too much of the long-run return:

-> Volatility scaling: size momentum positions inversely to their recent volatility. When a stock or market becomes more volatile, reduce the position. This automatically cuts exposure heading into turbulent periods.

-> Combining with absolute momentum: if the overall market has negative trailing returns, move to cash or short-term bonds. This keeps you out of the market during the conditions most likely to generate momentum crashes.

-> Diversification across asset classes: momentum works across stocks, bonds, commodities, and currencies. Diversifying across asset classes reduces the impact of any single-market crash on the overall portfolio.

-> Sector-level momentum: trading momentum at the sector level (using ETFs rather than individual stocks) reduces the idiosyncratic risk of any single company and tends to produce smoother return streams.

Why Momentum Persists Despite Being Known

A natural question: if momentum is so well-documented, why doesn't everyone trade it, eliminating the anomaly? Several factors explain why it persists even after decades of academic publication.

First, the crashes. Many institutional investors cannot tolerate the occasional 40-50% drawdowns that momentum strategies produce. Career risk is real: a fund manager who underperforms dramatically during a momentum crash may not have a job when the strategy recovers. This limits the capital that can be deployed in pure momentum strategies.

Second, the behavioral biases that create momentum are deeply embedded in human psychology. Anchoring, the disposition effect, and herding are not bugs in human decision-making that can be patched with knowledge – they are features of how humans process uncertainty under pressure. Knowing about a bias does not automatically eliminate its influence.

Third, momentum requires discipline that most investors find difficult to maintain. Buying stocks that have already risen significantly – when everything in your gut says 'it's too late' – is psychologically hard. Selling recent losers at a loss, when you want to hold on and recover, is harder still. The strategy demands a systematic, rules-based approach that bypasses emotional judgment.

Putting It Together

Momentum is one of the most powerful and most misunderstood tools in investing. It is not about chasing performance thoughtlessly. It is about recognizing that markets underreact to information, that human psychology creates predictable patterns of behavior, and that those patterns can be exploited systematically with the right framework and discipline.

The key insights from this article: momentum works across markets and asset classes; it is explained by behavioral biases that persist because they are built into human psychology; it comes in cross-sectional and time-series forms; and it is subject to crashes that require careful risk management.

In Article 08, we move from theory to practice: the specific indicators and tools momentum traders use, how sector rotation creates momentum opportunities across the GICS framework, and a step-by-step look at how the economic cycle drives which sectors lead and which lag.

KEY TAKEAWAYS

v Momentum is the tendency of recent outperformers to keep outperforming and recent underperformers to keep underperforming – typically over a 3-12 month horizon.

v It is one of the most well-documented anomalies in finance, first rigorously studied by Jegadeesh and Titman (1993) and confirmed across markets and asset classes.

v Even Eugene Fama, the father of the EMH, acknowledged momentum as the 'premier anomaly' his models could not explain.

v Behavioral biases – anchoring, the disposition effect, herding, and overconfidence – are the leading explanations for why momentum persists.

v Cross-sectional momentum ranks assets against each other; time-series (absolute) momentum compares an asset's return against zero. Combining both improves risk management.

v Momentum crashes are real and severe – often 40-50% during sharp market reversals. They can be partially mitigated through volatility scaling and absolute momentum filters.

v Momentum persists despite being widely known because the crashes make it psychologically and institutionally difficult to implement with consistent discipline.

How FinAi Fits In

FinAi can help traders measure momentum systematically: recent performance, relative strength, trend quality, and deterioration signals.

This is the difference between chasing a stock because it feels hot and participating in a momentum move because the evidence remains aligned.