Monthly portfolio, crash guard

Momentum with Market Filter

From NSE's Nifty200 Momentum 30 method, with Meb Faber's trend filterRead about it

Momentum, the tendency of stocks that have risen steadily to keep rising, is the strongest pattern in Indian shares, but it falls hard with the market. This holds up to 20 of the strongest stocks each month, and waits in cash whenever the Nifty 500 is below its 200-day average, the simplest trend filter there is.

Yearly return, after costs+18.2%Nifty 500 +11.5% over the same years

Worst fall−35.5%Win rate44%Holds nowCash

How it works

How stocks are chosen each month

On the first trading day of each month, checks the Nifty 500: is it above its 200-day average?
If it is, ranks the 500 most-traded NSE stocks by momentum: the 6-month and the 12-month return, each divided by the stock's 1-year volatility (NSE's Nifty200 Momentum 30 method)
Holds up to the top 20; each one joins with a twentieth of the portfolio

When a stock leaves at a monthly rebalance

A stock leaves at a monthly rebalance once it falls out of the top 40
Every stock leaves, and the money waits in cash, at a rebalance when the Nifty 500 is below its 200-day average
Or it leaves if it stops trading

Typical holdingAbout three months on average; in cash through falling markets

An example from the backtestAditya Infotech CPPLUSThe rules were met at the close before 1 Sep 2026. The trade opened that morning and closed on 1 Oct 2026, 21 trading days later, +7.2% after costs. Exit: market filter: index below its 200-day average.

The stocks it holds now Cash

In cash since the latest rebalance on 1 Oct 2026. The next check is on the first trading day of November 2026. A screen, not a recommendation.

The Nifty 500 was below its 200-day average at the latest rebalance, so the rules hold no stocks: the money waits in cash until a month starts with the index back above it.

Tested over 20 years

A separate test on NSE's end-of-day files from 2006, among the 500 most-traded stocks each month (companies later delisted included), after 0.2% costs on each side. The years 2006 to 2015 were kept aside while Swing Lab chose its strategies from 31 candidates, so they show how the rules did on data they were never fitted to. Results include the 2008 crash.

2006 to 2026+23.0%/yrNifty 500 +11.1%/yr

Compound yearly return over the whole test.

2006 to 2015, unseen+28.3%/yrNifty 500 +10.6%/yr

The decade kept aside: the fairest test.

2016 to 2026+18.3%/yrNifty 500 +11.5%/yr

The decade the live backtest below covers.

After tax+19.6%/yrIndex fund +10.2%/yr

At today's rates: 20% on gains held under a year, 12.5% after. The index fund pays 12.5% once, at the end, and 0.2% a year in fees.

Worst fall−49.9%Nifty 500 −64.3%

The largest drop from a high, 2006 to 2026 (2008 for most).

5-year stretches ahead94%

Share of all 5-year periods in which it beat the Nifty 500.

Longest behind the index36 months

The longest it went without gaining ground on the index: the patience it asks for.

How sturdy64 of 108

Nearby settings (10, 20 or 30 stocks; monthly, mid-month or quarterly; three lookbacks; top 200 or 500) that beat the index in both decades. With costs doubled, 92 still beat it over the 20 years.

Each calendar year, the strategy against the Nifty 500
Year	Strategy	Nifty 500	Difference
2006	+46.7%	+33.7%	+13.0%
2007	+216.8%	+62.5%	+154.3%
2008	−41.2%	−57.1%	+15.9%
2009	+41.0%	+88.6%	−47.6%
2010	+32.5%	+14.1%	+18.4%
2011	−5.6%	−27.2%	+21.6%
2012	+27.1%	+31.8%	−4.7%
2013	−1.5%	+3.6%	−5.1%
2014	+94.8%	+37.8%	+57.0%
2015	+3.0%	−0.7%	+3.7%
2016	−9.5%	+3.8%	−13.3%
2017	+99.5%	+35.9%	+63.6%
2018	−14.1%	−3.4%	−10.7%
2019	−2.6%	+7.7%	−10.3%
2020	+72.1%	+16.7%	+55.4%
2021	+85.4%	+30.2%	+55.2%
2022	−12.9%	+3.0%	−15.9%
2023	+45.1%	+25.8%	+19.3%
2024	+13.0%	+15.2%	−2.2%
2025	−17.0%	+6.7%	−23.7%
2026to 1 Oct 2026	+6.4%	−8.4%	+14.8%

Price returns, as the Nifty 500 price index is measured; dividends would add about 1% a year to both. Past results do not promise future ones: every strategy here had years well behind the index.

Backtest

1 Jan 2016 to 9 Oct 2026, run again after every close: up to 20 stocks, chosen again on the first trading day of each month, on NSE's 500 most-traded stocks at each date, after 0.2% costs on each side, against the Nifty 500 index.

Yearly return+18.2%Index +11.5%

Compound growth a year (CAGR).

Total return+508%Index +224%

Gain or loss over the whole test.

Worst fall−35.5%Index −38.3%

Largest drop from a high (max drawdown).

Win rate44.4%

Share of trades that closed with a gain.

Profit factor2.28

Rupees gained for every rupee lost.

Average trade+9.66%

Average result per trade, after costs.

Average gain · loss+38.8% · −13.6%

A typical winning and losing trade.

Days held66.4

Average trading days per trade.

Trades554

About 51 a year.

Sharpe ratio0.93

Return for each unit of ups and downs.

Time invested73%

Average share of the money in trades.

Best · worst trade+743% · −74%

The extremes, after costs.

₹1 lakh following the rules

1 Jan 2016 to 9 Oct 2026: ₹6.08 L against ₹3.24 L in the Nifty 500.

Momentum with Market FilterNifty 500

Falls from the high

Worst in this period −35.5%

Each calendar year

Return in the year, after costs.

Momentum with Market FilterNifty 500

How the 554 trades ended

Each trade's return after costs, grouped.

GainLoss

How sturdy is it?

The same rules, tested again with one thing changed. A result worth trusting holds up across these; a big swing means the headline figure depends on luck or on the exact setup.

Beat the Nifty 500 in 5 of 5 versions and in 1 of 2 halves of the test.

Version	Yearly return	Worst fall	Win rate	Profit factor	Trades
As published20 stocks, chosen again each month, 0.2% costs on each side	+18.2%	−35.5%	44.4%	2.28	554
Costs doubled0.4% on each side, as slippage in thinner stocks can be	+17.1%	−36.7%	43.5%	2.19	554
10 stocksHalf as many, each a bigger share	+19.1%	−43.4%	43.8%	2.26	313
30 stocksHalf as many again, each a smaller share	+18.5%	−34.2%	43.5%	2.47	774
Without the market filterHolding stocks even while the index is below its 200-day average	+26.1%	−43.1%	43.5%	2.29	531

First half · 1 Jan 2016 to 25 May 2021

Strategy+28.6%/yr

Nifty 500+12.8%/yr

Ahead of the index by 15.8 points a year.

Second half · 25 May 2021 to 9 Oct 2026

Strategy+8.7%/yr

Nifty 500+10.2%/yr

Behind the index by 1.5 points a year.

Recent trades in the backtest

The last 12 stocks to leave the portfolio.

Stock	Joined	Left	Days	Return	Why it left
Avalon TechnologiesAVALON	1 Sep 2026	1 Oct 2026	21	−3.9%	Market filter: index below its 200-day average
R R KabelRRKABEL	1 Sep 2026	1 Oct 2026	21	−14.6%	Market filter: index below its 200-day average
Aditya InfotechCPPLUS	1 Sep 2026	1 Oct 2026	21	+7.2%	Market filter: index below its 200-day average
Shilpa MedicareSHILPAMED	1 Sep 2026	1 Oct 2026	21	+9.5%	Market filter: index below its 200-day average
Syrma SGS TechnologySYRMA	3 Aug 2026	1 Oct 2026	42	+28.7%	Market filter: index below its 200-day average
Aeroflex IndustriesAEROFLEX	3 Aug 2026	1 Oct 2026	42	+23.4%	Market filter: index below its 200-day average
Diamond Power InfrastructureDIACABS	3 Aug 2026	1 Oct 2026	42	+9.8%	Market filter: index below its 200-day average
Welspun CorpWELCORP	3 Aug 2026	1 Oct 2026	42	+61.3%	Market filter: index below its 200-day average
Sansera EngineeringSANSERA	3 Aug 2026	1 Oct 2026	42	+31.7%	Market filter: index below its 200-day average
Bliss GVS PharmaBLISSGVS	3 Aug 2026	1 Oct 2026	42	+43.3%	Market filter: index below its 200-day average
Laurus LabsLAURUSLABS	3 Aug 2026	1 Oct 2026	42	+8.0%	Market filter: index below its 200-day average
Mtar TechnologiesMTARTECH	3 Aug 2026	1 Oct 2026	42	+35.6%	Market filter: index below its 200-day average

How the strategies were chosen and tested

Twenty years, with a decade kept aside. 31 published strategies were tested on NSE's end-of-day files from 2006: the 24 that Swing Lab used to show, six from academic research and NSE's factor indices, and one variant. The years 2006 to 2015 were not used to pick anything, so they show how each did on data it was never fitted to. Only three cleared every test below.
The bar they had to clear. Beat the index in 2006 to 2015 and again in 2016 to 2026; still beat it with costs doubled; give similar results with nearby settings (more or fewer stocks, other rebalance days, other lookbacks); beat buying 20 stocks at random; and stay statistically significant after allowing for all 31 strategies tried, since some will look good by luck alone.
What did not make it. Every short-term pullback strategy (RSI-2, cumulative RSI, stochastic), every quick breakout (NR7, inside bar, Bollinger squeeze, Turtle, Ichimoku, Keltner, breakaway gap) and the futures long build-up trailed the index in both decades once costs were paid. Darvas Box, EMA 20/50, Pocket Pivot, 52-Week High, RS-Line High and Golden Cross did beat it, but not by enough to be sure it was more than luck, or only with one setup. Delivery Accumulation came closest: far ahead from 2006 to 2015, it only matched the index from 2016, and fell behind with costs doubled. In Indian shares, trends that last months have paid; short swings have not.
Universe. The 500 most-traded NSE stocks at each date, by six-month average traded value, re-ranked every month, including companies that later stopped trading (131 of the 1,353 since 2016). Testing only today's survivors would flatter every result.
Trading. Rules are read at a day's close and the portfolio changes at the next day's open. The momentum strategies hold up to 20 stocks, chosen again on the first trading day of each month; Trend Template holds up to 10 trades at a time, each a tenth of the money.
Costs. 0.2% on each side of every trade (STT, stamp duty, exchange charges and slippage). Every strategy is also run with costs doubled.
Tax. Each strategy's 20-year record is also shown after capital-gains tax at today's rates (20% on gains held under a year, 12.5% after), against an index fund that pays tax only once.
Live backtest. 1 Jan 2016 to 9 Oct 2026, rerun on NSE's files after every close. Splits, bonus issues, demergers and symbol changes are joined up, so none shows up as a gain or a loss.
What is left out. Dividends (as in the Nifty 500 price index it is measured against) and the chance that a large order moves a thinly traded price.
What to expect. All three are forms of momentum, so they tend to do well and badly at the same times: they have led in trending markets and lagged in choppy ones and in sharp rebounds (2009, 2016, 2018, 2022, 2025). Each has had stretches of two to five years behind the index.

Swing Lab is a screen and a history lesson, not advice. RupeeVerse is not registered with SEBI as an investment adviser or research analyst. A stock meeting a strategy's rules, or held by one, is not a recommendation to trade it, and past results do not predict future ones. Do your own research.