AlphaStocks

About AlphaStocks

Mission

AlphaStocks scores every stock the same way — five models, sector-calibrated weights, updated daily. We evaluate 1,500+ companies across the S&P 1500 and top global ADRs through Graham, Buffett, Piotroski, Lynch, and Greenblatt and combine them into a single composite score from 0 to 10. No analyst opinions, no hidden methodology, no paywall on the scores themselves. Every score is transparent and fully documented.

Why We Built This

Free tools like Finviz, Yahoo Finance, and Morningstar give you raw data — dozens of ratios per stock, no framework for what they mean together. Paid services give you analyst opinions, but from a single methodology you can't verify. AlphaStocks sits between them — an algorithmic research tool that runs five independent models on 1,500+ stocks daily and explains exactly why each stock scored the way it did.

We don't sell signals or make buy/sell recommendations. We show you which companies look fundamentally strong, fairly valued, and trending upward — and let you make your own decisions.

How It Works

Our composite score weighs four dimensions (general profile shown): Quality (40%) assesses profitability, financial health and ten-year durability of returns using the Piotroski and Buffett models. Value (10%) measures price attractiveness via the Graham, Lynch and Greenblatt models. Momentum (35%) ranks each stock's 12-1 month price return against the rest of the universe. Timing (15%) is the lower of Value and Momentum, to avoid value traps — cheap stocks that keep falling.

Scores are recalculated daily from SEC EDGAR filings and Alpaca Markets price data. Seven sector-specific profiles ensure banks aren't scored like tech companies and utilities aren't penalized for slow growth. Read the full methodology →

Backtest Results

We test the scoring formula with a point-in-time backtest on the S&P 500 (January 2021 – February 2026, no look-ahead). Auditing it, we found and fixed two look-ahead biases. What remains: a top-10 basket ended slightly ahead of the market, but the excess is not statistically significant and is mostly market exposure.

We publish those results in full, including what did not work. AlphaStocks is a research tool for building a shortlist, not a return predictor. See full backtest results →

* Hypothetical backtested results. Does not reflect actual trading. No transaction costs, taxes, or slippage included. Past performance does not guarantee future results.

Founder and Author

Maksym Lytvynov

Founder & Lead Engineer

Software engineer with 10+ years of experience building products in fintech, healthcare tech, real estate automation, and blockchain. Graduate of Taras Shevchenko National University of Kyiv. Led development of 40+ projects across startups and established companies. Background in operations management, accounting fundamentals, and quantitative analysis.

Built AlphaStocks to solve a personal problem — finding fundamentally strong stocks without spending hours on financial statements. The five-model approach combines five published investment frameworks into a single, transparent score.

He writes the methodology pages and the Learn articles, and builds the scoring code they describe.

Editorial standards

  • Scores come from code, not opinion. Each of the five models is computed as its primary source defines it, and every threshold on the methodology pages is generated from the constants the scoring code uses, so the page states the rule the score was computed with. No score is set or adjusted by hand. A model that does not fit a company is reported as not applicable, never scored as zero.
  • Fair values follow published methods.One valuation method per sector profile, from Aswath Damodaran's work, with dated parameters (market inputs as of January 2026)and a low–base–high range. The market price is never an input. When a method's inputs are invalid, the fair value is withheld and the reason is shown.
  • Sources are named. Every criterion on the methodology pages names its source; the data comes from the filings and price feeds listed below.
  • Results are reported as they are. The backtest is published in full, including what did not work.

Corrections

If a score, a fair value, a figure or a sentence on this site looks wrong, email contact@alphastocks.app with the page link. A confirmed error in the data or the scoring is fixed in the code, so every affected stock is rescored, not just the one reported. The methodology pages always show the rules in force; the v23 release note explains what changed from the previous engine and why.

Data Sources

  • Financial statements: SEC EDGAR — annual and quarterly filings (10-K, 10-Q, 20-F) for all covered companies
  • Market prices: Alpaca Markets — daily OHLCV data, updated every 2 hours during US market hours
  • Company events: SEC EDGAR — each company's Form 8-K filings (results, executive changes, auditor changes, restatements)
  • Index constituents:S&P 500, MidCap 400, SmallCap 600 and top global ADRs — synced daily

Important Disclaimer

AlphaStocks is an algorithm-generated research tool, not a registered investment adviser. Scores, ratings, and fair value estimates are based on historical financial data and mathematical models. They do not constitute personalized investment advice. Always conduct your own due diligence and consult a qualified financial adviser before making investment decisions. Full terms of service →

← Back to homepage