A name is not a complete strategy
A relative-momentum rule needs a universe, ranking window, observation date, selection method, weighting, holding period and rebalance procedure. A trend rule needs its own signal and position rules. Changing any of these can materially change results.
A statement such as buy what is strong leaves too much unspecified to evaluate. A long-only selection process also differs from an academic long–short portfolio that buys past winners and shorts past losers. Results from one cannot be transferred directly to the other.
Rankings describe a past window
Imagine three fictional assets with trailing six-period returns of +12%, +5% and −3%. A simple relative ranking places them in that order. It says nothing yet about their next-period returns. If the first subsequently loses 8% while the others gain, the original ranking was still correctly calculated.
This illustrates the difference between a signal and its outcome. A method can be precisely specified and still lose money. Choosing the window after seeing which ranking worked would also contaminate the evaluation with hindsight.
Evidence needs its implementation attached
Jegadeesh and Titman’s original research studied specific winner/loser portfolio constructions and holding periods. It is evidence about those tests, not a guarantee that a modern product or a newly invented ranking rule will earn the same result.
Rapid reversals, crowded positions, turnover, trading costs and the availability of short positions can affect implementation. A sharp change in leadership can hurt a process that relies on persistence. A strategy should explain those failure conditions alongside its rationale.
Use market leadership language carefully
A momentum label may refer to a factor index, a fund, an oscillator or a qualitative description of a price move. Those are not interchangeable. A fund with momentum in its name follows its actual portfolio rules rather than every casual use of the word.
When a market report identifies momentum leadership, find the instrument, return window and benchmark. Treat that as a dated observation. Turning it into a personal trading process requires additional rules, risk decisions and evidence that the article does not supply.
A security ranks first over the past six months. Does that prove it will rank first next month?
Show the explanation
No. The ranking measures past performance. A predictive or strategy claim requires separate evidence and remains uncertain.
Sources and editorial notes
Original explanations and hypothetical examples prepared with AI assistance. Primary references checked October 16, 2026. Numerical examples are hypothetical unless explicitly identified otherwise.
- Jegadeesh and Titman: Returns to Buying Winners and Selling Losers (1993)
Original evidence for specified historical momentum portfolio constructions, not a future guarantee.