It threw away half of the Magic Formula and did better. One number, no quality checks.

Ten years ago you put ₹10 lakh into the cheapest slice of the market, ranked by nothing but operating earnings yield, and rebuilt the list once a year. Today the top decile is worth about ₹59 lakh. The same money in a plain Nifty 500 index fund is around ₹38 lakh. That is roughly 20% a year against the market's 13.9%, a six point edge, from a screen so simple it fits on one line: buy what is cheap.

Greenblatt's Magic Formula famously used two numbers, cheapness and quality. Carlisle's bet was that the quality half is a distraction, and sometimes a tax. Strip it out, he argued, and you do just as well. In India, over this decade, he did better. But the return is the least interesting part. The interesting part is that the ranking actually worked.

Did buying cheap actually sort the winners?

This is the test most people skip. A ranking screen only earns its keep if the stocks it rates highly really do beat the ones it rates poorly. Sort every company into ten buckets by cheapness, from the dearest to the cheapest, and the returns should step up as you go. Here, remarkably, they do.

That number matters more than it looks. When we ran the two factor Magic Formula through the same test, its ranking was essentially random, a rank correlation near zero. Carlisle's single cheapness factor scored −0.88, a genuinely strong sort, with only three of nine steps out of order. Dropping quality did not just simplify the formula. It sharpened it. Cheapness, on its own, carried real information about what came next.

First you had to survive losing two thirds

So why isn't everyone rich off one line of arithmetic? Because the screen buys the unloved, and the unloved get more unloved before they turn. To keep that 20% you had to sit through a peak to trough fall of about 72%, nearly twice the index's, watching two rupees in three disappear and doing nothing. That is not a statistic on a chart. That is years of wondering whether you were a genius or a fool.

The 20% was really ten quite different years

A single headline number flatters a strategy this streaky. The honest way to read it is one bar for every month you might have begun, each one held the same length. Do that and the tidy 20% fans out into a spread wide enough to change your life, or end your patience, depending on nothing but timing.

Over one year the outcomes ran from losing half your money to more than doubling it, and it beat the index in 57 of 100 starts. But stretch the hold and the picture steadies in a way the Magic Formula's never did. Over five years every single start date made money, the worst still up 4% a year, and it beat the index in 62% of runs. This is the profile of a real edge that needs time, not a lucky number that needs a good entry.

A basket built from the unwanted

Cheapness leads you to unglamorous places. The screen lived in commodities, energy and heavy industry, holding the same deep value names for years: Bombay Burmah, GHCL, ONGC, NMDC. Around seven in ten rupees sat in materials, energy and industrials, the cyclical, out of favour corners where cheap stocks gather. This is not a portfolio of household compounders. It is a basket of things nobody wanted, bought because nobody wanted them.

Right idea. Iron stomach required.

This is not a case against the Acquirer's Multiple. It is close to the opposite. Carlisle's heresy held up: on this decade of Indian data the one number screen beat the market, and unlike its two factor cousin its ranking genuinely sorted good from bad. The catch is not the edge, it is the endurance. A 72% drawdown ends most people's discipline long before the strategy ends its recovery. The point of checking an idea like this is not to prove it wrong. It is to learn what holding it actually demands of you, so you find out before your money does.

So we took it to Krest, and ran it through the whole test.