Composite Score Methodology

How Quintarth computes the 0-100 Composite Score on every Report Card

Inputs

The Composite Score is a weighted average of five sub-scores, each scaled to 0-100. Sub-scores are computed nightly from public-domain data sources we ingest:

Sub-score	Weight	Inputs	Source
Risk	25%	Beta, debt/equity, current ratio, interest coverage, distance to 52w-low	yfinance + EDGAR XBRL
Fundamental	30%	P/E, P/B, EV/EBITDA, ROE, gross margin, FCF margin, revenue growth 3y CAGR	yfinance + EDGAR XBRL (20+ year history)
Momentum	20%	3m and 6m total return, RSI(14), 50-day vs 200-day SMA	yfinance OHLCV
Volume	10%	Volume vs 20-day avg, On-Balance-Volume slope, anomaly score	yfinance OHLCV
Insider	15%	Net insider buys vs sells (last 90d), cluster-buy detection, officer-vs-director ratio	SEC Form 4 + SEDI

Computation

composite = 0.25 * risk + 0.30 * fundamental + 0.20 * momentum + 0.10 * volume + 0.15 * insider

Confidence band

Each Report Card shows a confidence band (e.g. ±5). This is computed from data completeness — sub-scores with missing data points get downweighted, and the band widens. Tickers with full 20-year XBRL history have a tighter band than freshly-listed micro-caps.

What it's NOT

Not a recommendation. The composite score is a non-GAAP mathematical model output. Score labels (Strong / Moderate / Neutral / Weak) describe signal strength, not investment merit. We never give buy/sell recommendations. See our full disclosures.

Reproducibility

Every Report Card lists the ingredients used. Click any sub-score in the radar chart to see the underlying data points. To independently verify a score, fetch the raw inputs from yfinance (or the SEC EDGAR XBRL endpoint at /api/v1/market/xbrl/{ticker}) and apply the formulas above.