Transparency

Methodology & Limitations

This page documents exactly how the screen works — where data comes from, how each score is computed, what is approximated, and what the tool cannot do. It is deliberately detailed. A ranking you cannot interrogate is not worth much, and an automated screen that does not disclose its approximations is asking for more trust than it has earned.

What the screen does

An automated scan runs on a schedule during U.S. market hours. Each run evaluates every constituent of the S&P 500 and S&P 400 MidCap indices — roughly 900 stocks — against several published trading frameworks, then publishes two ranked lists of ten stocks each.

The two lists draw from the same evaluated pool but rank by different weightings. Momentum Leaders emphasizes intraday price and volume signals. Trend Quality & Value emphasizes fundamental and valuation factors, deliberately excluding the intraday momentum triggers. The two lists never overlap: names appearing in the first are removed before the second is ranked.

Data sources and refresh cadence

Filtering

Before scoring, stocks are excluded if they fail basic liquidity and price thresholds — a minimum share price and a minimum 20-day average dollar volume. The purpose is to remove names where the screen's own signals would be unreliable and where a position would be difficult to exit.

Stocks with insufficient price history (roughly under 210 trading days) are also excluded, since the 200-day moving average and 52-week range cannot be computed meaningfully without it. This means recent IPOs will not appear regardless of how they are performing.

How the Trend Template is scored

All eight of Minervini's published criteria are evaluated, and the count of criteria met is the primary sort key for both lists — not one input among several. A stock meeting seven criteria ranks above one meeting five regardless of any other factor. The composite score only breaks ties within the same count.

Two implementation details worth stating explicitly:

• The "200-day average trending up" test compares the current 200-day average against its value approximately one month (22 trading sessions) earlier.
• The Relative Strength Rating is an approximation — see the next section.

The Relative Strength approximation

RS Rating is computed as a weighted blend of trailing returns (weighted most heavily toward the trailing three months, with additional weight on six, nine, and twelve months), then percentile-ranked from 1 to 99.

The important caveat: that percentile is calculated within the scanned universe only — roughly 900 large and mid-cap stocks. It is not ranked against the entire market. A stock showing RS 90 here has outperformed about 90% of the S&P 500 and 400, which is not the same as outperforming 90% of all publicly traded stocks. Commercial providers such as Investor's Business Daily use proprietary calculations against a broader universe; our numbers are not equivalent and should not be compared directly.

How the other frameworks are scored

CANSLIM factors

Current and annual earnings growth, proximity to new highs, volume-based demand, relative strength leadership, and institutional ownership are scored where data is available. Missing fields are skipped rather than counted against a stock — an unavailable data point is not evidence of a weak fundamental. Because free fundamental data coverage is inconsistent, a given stock's CANSLIM score reflects only the factors that could actually be measured. Each stock's detail view lists which signals were found.

The M factor (market direction) is evaluated once per scan at the market level and displayed as the status indicator in the site header, not as a per-stock score.

Breakout and momentum triggers

Zanger-style rules look for a move above a recent high combined with meaningfully elevated volume and a significant daily gain. Qullamaggie-style rules look for top-decile relative strength, sufficient average daily range, a tight consolidation, and either a breakout or a large gap on heavy volume.

These are simplified approximations built from publicly described criteria. Neither trader has reviewed, endorsed, or is affiliated with this implementation.

Value metrics

PEG ratio, trailing P/E, profit margin, and debt-to-equity are evaluated against conventional thresholds where the data is published. These are deliberately simple screening thresholds, not a valuation model, and they make no attempt to estimate intrinsic value.

Why non-qualifying stocks still appear

Both lists always show ten names. When fewer than ten stocks fully clear the Trend Template threshold, the remaining slots are filled with the closest available candidates, each flagged as not qualifying.

This is a deliberate design decision with a tradeoff worth naming. Showing the closest candidates gives a continuous view of relative strength even in poor conditions. The risk is that a list looks like ten opportunities when it may contain none. That is why every non-qualifying name is explicitly marked, and why the header reports how many actually qualified. On a weak day, that number may be zero — and that is meaningful information rather than a malfunction.

Known limitations

• Delayed data. Roughly 15–20 minutes behind live prices. Not suitable for execution.
• Approximated RS Rating. Ranked within the scanned universe only, not the full market.
• Inconsistent fundamentals. Free-tier coverage varies; missing fields are skipped, so scores are based on partial data for some stocks.
• No earnings calendar. The screen detects the price-and-volume signature of a large gap but cannot confirm what caused it. A gap on a positive earnings surprise and one on a disastrous announcement can look identical to the scanner.
• No news or event awareness. Litigation, regulatory action, management changes, accounting concerns — none of this is visible in price data.
• Manually maintained index membership. The constituent list is fixed in code and updated periodically. Between updates, recently added index members may be missing and removed ones may linger.
• No survivorship-bias correction and no backtesting. This screen reports current conditions. It makes no historical performance claims, because none have been rigorously tested here.
• Large-cap and mid-cap only. Small-caps, micro-caps, ADRs, and recent IPOs are outside the scanned universe.
• Company names and sectors come from the same third-party feed as the fundamentals, and are occasionally missing or out of date. They are shown for identification convenience, not as authoritative classification.
• No charts are drawn here. On delayed data a rendered chart would look more authoritative than it deserves to, so each detail view links out to a live chart instead.
• No position sizing, stops, or risk management. The screen identifies candidates only. The practitioners whose frameworks it borrows from all emphasize that risk management — not stock selection — determines outcomes.

Why the methodology is published at all

Automated financial screens are easy to build and easy to dress up as more authoritative than they are. Publishing the approximations, the data limitations, and the specific things the tool cannot do is the only honest way to offer one. If a ranking here looks wrong to you, this page should give you enough detail to work out why — and if the reason is a flaw in the method, that is worth knowing.