Methodology
How Ranked Stocks builds the RS Score
A data-driven multi-factor framework designed to narrow the universe of global equities into a focused research shortlist.
Built by fundamental investors
RankedStocks was created by long-term, business-first investors. Deep fundamental research is the foundation, but filtration is always the first step.
The platform began as an internal screening framework designed to surface promising candidates efficiently before deeper analysis begins.
Phase One: Traditional Valuation Filters
- EV / EBITDA
- Price-to-Earnings (P/E)
- Price-to-Book (P/B)
- EV / Invested Capital
- Other conventional valuation ratios
Valuation screens are useful, but incomplete. Low multiples alone often identify companies that are cheap for structural or fundamental reasons.
Phase Two: Adding Business Quality
- Return on Invested Capital (ROIC)
- EBITDA margins
- Gross margins
- Related measures of operational strength
This layer helped differentiate cheap and broken businesses from cheap and structurally strong businesses.
Phase Three: Growth and Market Repricing Signals
- Lagging and leading growth metrics
- Revenue and earnings trajectory indicators
- Select sentiment and forward-looking signals
At this stage, the model became a structured multi-factor system designed to surface companies with unusually strong setups.
Deep learning and the RS Score
Rather than stopping at filters, the full dataset across the coverage universe is processed through a deep learning model that estimates the probability of outperformance over the next 12 months.
The model output is the RS Score, which compresses complex multi-dimensional information into a single interpretable metric while keeping underlying drivers transparent.
Historical observations
- From 1993 to 2018, buying the +80 stocks outperformed the S&P 500 in 84% of calendar years.
- From 2019 to 2025, the +80 RS Score paper portfolio has outperformed the S&P 500 each year.
Past performance does not guarantee future results.
A research tool, not investment advice
RankedStocks does not tell users what to buy or sell. The objective is to compress a large global universe into a focused research set.
The objective is not blind diversification. The objective is intelligent focus to support deeper business analysis, valuation work, and risk management.