# Stock screening strategies: filters, criteria, and what screens miss > Stock screening strategies turn a market of thousands of names into a shortlist you can research. The classic screens, the best screening criteria, the traps, and the one criterion no screener shows you. By Maya Koeva · 2026-09-01  There are several thousand listed stocks in the US alone. Nobody researches them one by one, and nobody needs to: a stock screen cuts the list down by rule. You set the criteria, the screener returns the names that pass, and your research starts from dozens instead of thousands. The strategy is in the criteria, and criteria are where most screens go wrong. The scope first: a screen produces a research list, not a buy list. Passing a filter means a stock matched some numbers on some date. Deciding whether to buy it is a different job, covered in [how to know what stock to invest in](/learn/how-to-know-what-stock-to-invest-in). ## Four screening strategies that have held up Most working screens are a variation on one of these. **Value screens** look for stocks priced low relative to what the business earns or owns: low price to earnings, low price to free cash flow, low enterprise value to EBITDA. The catch is that the market prices most cheap stocks cheaply for a reason, so bare value screens fill up with melting businesses. Pair the cheapness criterion with a quality floor (positive free cash flow, manageable debt) to filter the traps. **Growth screens** look for expanding businesses: revenue growth above some rate, earnings growing faster than revenue, estimates rising. Their failure mode is the price tag; a great business at any price is how investors overpaid for the 2021 growth cohort. A valuation ceiling, even a loose one, keeps a growth screen honest. **Quality screens** ignore price and hunt for durable economics: high return on invested capital, stable margins, low debt, consistent cash generation. They surface companies you would want to own; they say nothing about whether now is a reasonable time to buy them, so pair a quality screen with a valuation or timing criterion. **Momentum screens** buy strength: stocks near 52-week highs or outperforming the market over recent months. Momentum is one of the most studied effects in markets, and also the most uncomfortable to hold, because it reverses hard at turns. Momentum screens need an exit rule far more than the other three. None of these is secret, and a screen everyone can run surfaces names everyone has already seen. ## The best stock screening criteria are few, and boring Screeners let you stack twenty filters. Resist it. Every criterion you add tunes the screen more tightly to the past, and a screen fitted perfectly to the last two years is a backward-looking portrait, not a forward-looking filter. A workable pattern: - **One liquidity floor.** Minimum market cap or dollar volume, so the list holds stocks you can trade at close to the quoted price. - **One strategy criterion.** The value, growth, quality, or momentum rule doing the real work. This is the screen's thesis; know which it is. - **One guard against the strategy's failure mode.** The quality floor on a value screen, the valuation ceiling on a growth screen, the exit rule on momentum. - **Relative beats absolute where cycles bite.** "Cheapest quarter of its sector" survives regime changes better than "P/E under 15", which empties out in expensive markets and floods in crashes. Then leave it alone. A screen you rewrite weekly is not a strategy, it is a mirror for your mood. Run the same screen on a schedule in whatever tool you use (Finviz, TradingView, and most brokers ship one) and let the list change while the rules stay put. ## The traps **Cheap for a reason.** The most common screening casualty: a value screen full of businesses in decline. Guards help; reading the names still beats trusting the list. **Overfitting.** If you tweaked criteria until the backtested list looked great, you built a description of the past. Fewer, rounder numbers are a feature. **Snapshot data.** Screens run on reported numbers, which lag. A stock can pass a quality screen on figures from a quarter that ended before its business turned. **Passing means matched, not endorsed.** A screen has no idea a lawsuit landed yesterday, or that the sector's biggest customer just cut guidance. Every screened name still needs the news read before it needs your money. ## The criterion no screener shows you Every screener on the market filters the same public numbers, which is why ten value investors running ten screens surface the same forty names. The input screens do not carry is attention: whether anyone credible is talking about a stock, what they are saying, and whether the people who have been right before agree with the people shouting. That is the layer Quantral covers, and to be plain about the category: Quantral is not a stock screener. You cannot filter the whole market by P/E in it, and the tools above already do that well. What it does is track the finance conversation across a curated set of accounts on X and Reddit, grade each account's calls into a [credibility score](/learn/what-is-a-credibility-score), and rank the names it covers by what the credible side of the room is saying. Used after a screen, that answers questions the screen cannot: is this cheap stock cheap and ignored, or cheap and being argued about by people with real records? Is the momentum name still being bought by accounts that were early, or only by the crowd that arrives late? [Volume alone will not tell you](/learn/volume-vs-signal); a loud room is not a right room. And a name your screen loves that nobody we track discusses is a coverage note rather than a warning: the crowd layer has nothing to add, and the numbers carry the whole case. ## The bottom line Pick one screening strategy and know its failure mode. Use few criteria: a liquidity floor, the thesis rule, a guard. Rerun it on a schedule instead of rewriting it. Treat the output as a reading list, check what the credible crowd thinks about the survivors, and then do the research the screen was never going to do for you.