Cheapest P/S vs their own history (2026)

Methodology

Every US-listed large cap ($5B and up) with a full ten years of daily price-to-sales on file — at least 2,400 positive daily observations in the window, 95% of a decade of sessions — ranked by where today's P/S sits in its own ten-year distribution: the 0th percentile is the cheapest the stock has been on sales in that window, the 100th the dearest. Each company is compared with its own past, never with other companies. Price to sales rather than P/E or P/FCF because sales carry no one-off gains, are never negative, and are on file for the whole decade for far more companies. Multiples use trailing-twelve-month revenue from SEC filings over the settled close; the ten-year median and the discount to it come from the same daily series. Cheap against its own past is a screen, not a verdict — the sales- and earnings-growth columns show whether the business shrank with the multiple.

Performance columns are price returns: the change in the split-adjusted share price, the way quote screens report it, through the latest print during the session. Dividends are not reinvested. Fundamentals come from company filings via our EDGAR pipeline. Past performance does not predict future returns; this is not investment advice.

FAQ#

Which large-cap stock is the cheapest against its own price-to-sales history?

Boston Scientific Corporation (BSX) trades at 2.9× sales at the cheapest it has been in its own last ten years, 45% below its ten-year median of 5.3×.

What does the P/S percentile mean?

Where today's price-to-sales sits among every daily value the stock has recorded over the last ten years: 0 means it has never been cheaper on sales in that window, 100 never dearer, 50 is its own median. It compares a company with its own past, not with other companies — a software company at 8× sales and a grocer at 0.4× can both sit at their 5th percentile, and both belong here.

Why price to sales rather than P/E or price to free cash flow?

Sales carry no one-off gains or losses, are never negative, and are on file for a full decade for far more companies than earnings or free cash flow — so a low reading is a real de-rating, not one odd quarter. What P/S cannot see: a company whose margins fell, or one that bought a large low-margin business, looks cheaper on sales without being cheaper. The sales- and earnings-growth columns catch the first; the second needs a look at the company.

Is a stock at a ten-year-low valuation a buy?

Not by itself. A multiple at the bottom of its own range means the market now pays far less for a dollar of this company's sales than it used to — sometimes because the business changed for the worse, sometimes because sentiment did. The columns beside the percentile are there to tell the two apart: revenue and earnings growth over the last year, and how far today's multiple sits below the ten-year median. This is a screen, not advice.

Why is a well-known company missing?

The list requires a full ten years of daily price-to-sales on file (at least 2,400 observations) and a market cap of $5 billion or more. A company listed, spun off or re-keyed in our record within the last decade is left out rather than compared against a shorter history labelled as ten years; so are companies whose revenue we cannot read from their filings. Our record is not complete, and a company can be missing for no better reason than a hole in it.