Monthly momentum: hold the 15 strongest stocks until they fade

The idea

Momentum is one of the most persistent effects in markets: stocks that have risen strongly over the past several months tend to keep outperforming for a while. The classic way to use it is a rotation: own the strongest stocks, and swap them out as their strength fades.

The trick is not to swap too often. A strict “always hold the top 15” rule churns constantly as stocks hover around 15th place. Here a holding is kept until it drops below 30th, so a good stock having an ordinary month isn’t sold.

The rules

• Universe: NIFTY 500 stocks, point-in-time: on each date, only stocks in the index on that date. A holding is sold if its stock leaves the index.
• Score: the average of a stock’s 3-month and 9-month returns (60 and 190 trading days). Stocks need 9 months of history to be scored.
• Each month, on the last trading day’s close:
  1. Sell any holding ranked worse than 30th.
  2. Buy the best-ranked stocks you don’t already hold, to get back to 15 holdings, splitting the cash that’s free equally between the new buys.
  3. Leave the rest alone. Holdings that stay in the top 30 are not rebalanced, so their weights drift.
• Fills: at the next trading day’s open. Cost: 0.15% per side.
• No stop-loss, no target. The rank buffer is the only exit.

The backtest

Data: NIFTY 500 stocks, 1 January 2014 to 3 July 2026 (12½ years), about 50 buys a year.

(Calmar is CAGR divided by the worst peak-to-trough drawdown. The NIFTY 500 figure is the price index; the backtest’s stock prices exclude dividends too, so the comparison is like for like.)

Year by year, against the index:

*2026 to 3 July. Momentum beat the index in 8 of the 13 calendar years.

The pattern is typical of momentum: spectacular years when a trend runs (2017, 2020, 2021, 2023) and sharp losses when it reverses (2018, 2025).

Robustness tests

All on the same window and engine; the first row is the headline result above. The halves are the full run’s own results within each period.

• Costs matter only a little. About 50 buys a year, so doubling the cost takes off about 1% a year.
• Both halves beat the index: 18.2% against 12.3% in 2014–2019, and 25.4% against 14.1% in 2020–2026.
• No magic numbers. Every nearby combination of portfolio size and buffer gives 22–25% a year. More holdings (20) soften the drawdown slightly; fewer (10) deepen it. In hindsight 15 / 35 had the best return for its drawdown and 15 / 25 the highest return, but we kept 15 / 30, chosen before the test. Picking the best setting after seeing the results would be curve-fitting.
• Big winners matter. We measure each trade in percentage terms: its gain as a share of the portfolio when it was bought, so a late trade on a bigger portfolio doesn’t count for more. On that basis the five best of 622 trades produced about 47% of the growth. Without them the CAGR would have been roughly 11%, still profitable but behind the index. No single year contributed more than about 28% of the growth.

Risks

• Deep drawdowns. −49% at worst, deeper than the index’s −38%. Momentum portfolios fall hard when leadership reverses, and this one is fully invested in the stocks that had run up the most.
• Momentum crashes. Losing years were severe: 2018 (−27%) and 2025 (−27%), both while the index was only slightly down or up.
• Dependence on big winners. Almost half the growth came from five trades; without them the strategy would have trailed the index.
• Liquidity. The NIFTY 500 includes mid and small caps; at very large portfolio sizes some positions could be hard to buy or sell without moving the price.

Pros & cons

Pros

• Simple, fully mechanical monthly routine.
• Passed all eight checks, on survivorship-free data and net of costs.
• Robust to the exact portfolio size and buffer, to costs, and to weekly vs monthly rebalancing.

Cons

• Drawdowns deeper than the index’s.
• Very uneven year to year.

Our verdict

It passed all eight of our checks. Ranking the NIFTY 500 by 3- and 9-month strength and holding the top 15 with a generous sell buffer earned about 22% a year over 12+ years, net of costs, against about 13% for the index, and it held up in both halves of the history, at triple the costs and across nearby settings.

The price is volatility: a −49% worst drawdown and years like 2018 and 2025 when it lost over a quarter while the index barely moved. The key question is whether you can follow the rules consistently, especially through a year like that.