How to Screen for Value Stocks
To screen for value stocks, set liquidity filters first, then filter on a price ratio against earnings or book value combined with a profitability floor. The profitability requirement is what separates cheap companies from failing ones, and a screen without it returns mostly the second kind.
A value screen looks for a low price relative to something the company owns or earns. The filters are easy. What makes the exercise hard is that a low ratio is usually the market’s judgement rather than its oversight.
Before you start
A definition of cheap that you can state, because the screen encodes it whether or not you chose it. Low against earnings, against book value, or against cash generated — three different definitions with three different outputs.
Liquidity filters set before any valuation filter. Minimum price and average volume, so the list contains names that can actually be bought.
A way to record the output, since a value screen is judged over years rather than weeks. The date and the names, in a file you can return to.
The steps
1. Set liquidity floors first
The cheapest names by any ratio are disproportionately small and thinly traded. On this site’s shared series a round trip measures about 2% of the median bar range of 0.493, and in a thin name it is far worse.
2. Choose one price ratio
Price to earnings, price to book, or price to free cash flow. Stacking all three does not find better companies; it finds companies that are unusual on three measures at once.
3. Add a profitability floor
Several consecutive profitable years, or a positive return on capital. This single condition removes most of the companies whose ratio is low because the earnings are disappearing.
4. Exclude negative and near-zero denominators
A loss-making company has a meaningless price-to-earnings figure, and screeners sort them in unhelpful ways. Filter for positive earnings explicitly rather than assuming the tool handles it.
5. Check the balance sheet before the list is a list
Add a leverage limit. A company can be cheap on earnings and carrying obligations that make the equity a small residual claim, which the price ratio does not reveal.
6. Read every surviving name
Between 15 and 40. For each, answer one question: why is the market pricing this low. Sometimes there is no good answer, and those are the ones worth work.
7. Record the list and revisit it
Value screens are slow. Recording the output is the only way to find out in three years whether the approach found anything, and it costs two minutes now.
How to tell it worked
Liquidity filters were applied before any valuation filter.
Exactly 1 price ratio is in use, so the definition of cheap is single and stated.
A profitability floor is present, checked across at least 36 months.
And the list is between 15 and 40 names, small enough that every one was read.
Why the cheapest names are usually cheap
The ratio uses last year’s earnings and today’s price. When the market expects earnings to fall, the price moves first and the ratio looks attractive using a number that is about to change.
Some are cheap because nobody can buy them at size. A small, illiquid company can trade at a persistent discount for structural reasons that will never correct.
What the screen cannot see
Whether the business is deteriorating. Falling earnings produce exactly the same low ratio as a temporarily unloved company, and no filter distinguishes them.
Accounting choices. Book value depends on how assets were recorded and depreciated, so price to book compares companies using different measuring sticks.
Anything about timing. A correctly identified cheap company can stay cheap for years, which makes position size and patience more important than the screen’s accuracy.
Choosing which ratio to screen on
Price to earnings suits businesses with stable, meaningful profits. It is the most widely understood and the most easily distorted, because earnings absorb accounting choices, one-off items and financing decisions before they reach the denominator.
Price to book suits asset-heavy businesses. Banks, insurers, property companies and industrials own things whose recorded value means something. On a software company it is close to meaningless, because almost nothing of value appears on the balance sheet.
Price to free cash flow is the hardest to distort and the least available. It uses money that actually arrived rather than profit that was calculated, and many screeners do not offer it as a field.
Pick the one that matches what the business is. A single ratio applied across every industry is comparing companies that keep score in different units, and the screen will reliably surface whichever industry happens to look cheap on that particular measure.
The original data
Of the 24,971 unique videos in research/search-study-corpus.jsonl, 9 mention value stocks in the
title, at a median of 57,529 views across 8 channels, and 44% of those are instruction-shaped. Value
investing as an approach appears in 87 at 14,194 and growth stocks in 105 at 3,816. The counts come from
site/corpus_count.py and site/rank_investing.py.
9 videos at 57,529 against 87 on the approach at 14,194. A tenth of the coverage and four times the audience per video — the philosophy is heavily taught and the procedure for actually finding candidates is not.
The answer to the question on that chart is that a very low ratio is a question, not a conclusion. Something is usually expected to change in the denominator, and finding out what is the actual work — the screen has only told you where to look.
When it fails
The failure is a screen with no profitability requirement, and it systematically finds the wrong companies. Sorting by lowest price-to-earnings puts businesses whose earnings are collapsing at the top of the list, because the market has already marked the price down against results nobody has reported yet. The screen is doing exactly what it was told, and what it was told selects for the one characteristic a value investor is trying to avoid.
The second failure is stacking four ratios. That finds statistical oddities.
A third is skipping liquidity filters. The cheapest names are the thinnest.
A fourth is ignoring leverage. Cheap equity behind large debt is a residual claim.
A fifth is not reading the names. A screen produces candidates, not holdings.
And a sixth is judging it over weeks. This approach resolves over years or not at all.
Related
Value investing covers the reasoning behind these filters. Value stock is what the screen is trying to identify. And valuation is the work that starts once a name survives the list.
The filter I added that changed the output completely was a simple profitability requirement. Before it, the cheapest names by every measure were companies whose earnings were collapsing — the ratio was low because the denominator was about to be revised. Requiring several consecutive profitable years removed most of them and left a list I could actually read.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.