Stock screening and scoring

How to screen thousands of stocks down to a short list, and how a percentile-based scoring model ranks companies on growth, valuation and balance sheet.

What a screener does

A screener filters a large universe, such as the S&P 500 or Nasdaq 100, down to the handful of companies worth researching. It does not make decisions; it tells you where to look. Every name that passes still needs real research.

Filters versus scores

  • Filters are pass or fail: "PEG below 1", "debt-to-equity below 1". Simple, but a company just over a line disappears.
  • Scores rank every company on several metrics and add them up. Nothing disappears; the best rise to the top.

A four-pillar scoring model

One practical model scores six metrics in four equal pillars, 100 points in total:

Pillar Metrics Points
Trailing growth Revenue growth TTM 10, EPS growth TTM 15 25
Forward growth Revenue growth forward 10, EPS growth forward 15 25
Valuation Forward PEG 25
Balance sheet Cash versus debt 25

EPS gets more weight than revenue because earnings are what shareholders ultimately receive.

Scoring by percentile

Each metric is scored by where the company ranks within the list, not by a fixed cut-off. A percentile of 0.22 or lower earns nothing, 0.78 or higher earns full marks, and the points rise evenly in between:

points = 20 x (percentile - 0.22) / 0.56, clamped to 0 to 20, then scaled by the metric's weight.

Ranking against the list keeps the model fair in any market: in a year when every company grows slowly, the best of them still stand out. Missing data scores zero, and a negative P/E ranks worst on PEG.

For example, a PEG of 2.3 that lands at the 35th percentile of its list earns about 6 of the 25 valuation points.

Tiers

Tier Rank
S+ Perfect score
S Top 10%
A Next 10%
B 20% to 50%
C 50% to 75%
F Bottom 25%

Using a screen well

  • Screen for candidates, then research. Never buy from a screen alone.
  • Re-run it each quarter as new results arrive.
  • Look at why a company scores well: a high score from one extreme metric deserves suspicion.

For a free hands-on version, see the Finviz screener guide.

See also

Pages that link here: Building a research watchlist, Building your investing knowledge, Fundamental analysis, Growth, value and dividend investing, Systematic investing

Last updated September 30, 2026. Education only, not investment advice.