Proofstock rates US stocks with a transparent, six-factor quantitative model. Every rating on this site is produced by the process described below — no discretionary overrides, no hand-picking. These are research ratings, not investment advice, and not buy or sell signals.
Each stock is scored on six independent factors. Every factor is built from the raw inputs below, drawn from daily prices and quarterly fundamentals.
Whether the stock has been trending up relative to its peers, using medium-term price performance.
How profitable and financially sound the business is — high returns on capital and stable earnings, low leverage.
How cheap the stock is relative to fundamentals — lower multiples and higher cash-flow yields score better.
How fast the business is expanding its top and bottom line over a multi-year window.
The market-capitalization profile of the company, capturing the historical tendency for size to influence returns.
How calm the stock's price has been — lower realized volatility scores better.
For each factor, a stock is measured against its own sector peers — not the whole market. A technology company’s value is judged against other technology companies, an energy company’s against other energy companies. This is expressed as a z-score: how many standard deviations above or below the sector median the stock sits. It keeps the model from simply declaring entire sectors good or bad.
Extreme outliers are capped at three standard deviations, so a single freak data point can’t dominate a stock’s score.
The six factor z-scores are averaged into a single composite z-score for each stock.
Composite z-scores are ranked across the entire universe and converted to a percentile from 0 to 100. A percentile of 90 means the stock’s composite is stronger than 90% of the universe.
The percentile is bucketed into one of five ratings (below).
| Strong Bullish | Percentile ≥ 90 | Requires positive ROE (profitability gate). |
| Bullish | Percentile 70 – 89 | |
| Neutral | Percentile 30 – 69 | |
| Bearish | Percentile 10 – 29 | |
| Strong Bearish | Percentile < 10 |
Because ratings are percentile buckets, most of the universe lands in Neutral, with progressively fewer names toward each extreme. The Strong Bullish tier is the smallest of all because of the profitability gate.
The six factors give the model a built-in point of view. It favors smaller, cheaper, steadier, higher-quality, higher-momentum companies — because those are among the most-studied return factors in finance: the size premium, the value premium, momentum, the low-volatility anomaly, the quality factor. Proofstock didn’t invent them; it measures them, the same way, for every stock. That tilt has one deliberate consequence.
Large, widely-owned mega-caps — the biggest names most investors can list from memory — tend to rate low here. They sit on the opposite side of the two factors the model weights against them most: they are, by definition, the largest companies, and they usually trade at premium valuations. A stock priced at 40, 100, or 600 times earnings is, factually, expensive — and the model treats it that way, with no exception for how famous the name is.
But the model isn’t blind to their strengths. Many of these companies score wellon the quality factor — they’re genuinely profitable, well-run businesses, and the model says so. That’s simply outweighed by enormous size and rich valuation. A great company and a low Proofstock rating are not a contradiction — they’re the model doing exactly what it’s designed to do.
The easiest thing in the world would be to quietly nudge the famous names up so the rankings “look right.” We don’t. There is no “but it’s Apple” exception, no thumb on the scale, no editorial rescue for a popular stock the model rates poorly. The same six factors, applied identically to a $2-trillion giant and a $500-million small-cap alike — that’s the entire point, and it’s what makes the record worth keeping. If you disagree with the tilt, that’s a feature: you can see every factor and exactly why any stock lands where it does. Judge the method, not just the verdict.
Nothing here is advice to buy or sell any security, or an opinion that small companies are “better” than large ones — only a transparent description of what the model measures.
The growth factor requires at least $100 million in trailing-twelve-month revenue. Very small companies can post enormous percentage growth off a tiny base, which distorts the factor; the floor keeps growth meaningful rather than an artifact of a small denominator.
A stock whose composite would otherwise reach the Strong Bullish tier (percentile ≥ 90) but has non-positive ROE is demoted to Bullish. The top rating is reserved for companies that are actually profitable.
On each run, the top and bottom of the universe are written to an append-only record and never edited afterward. Each snapshot carries a SHA-256 hash that is chained to the previous one’s hash — so the published history can’t be silently rewritten after the fact. This is the “proof” in Proofstock: the track record is verifiable, not just asserted. You can browse the locked snapshots and their hashes on the Track Record page.