freebirdcapital
Tool Freebird RatingUpdated 10/2026

The Freebird Rating

A two-stage scoring system: sort out roughly, then score with weights — with thresholds per industry and a dividend score that knows the interest-rate level.

What the rating is for

The Freebird Rating is my filter between screening and analysis. It does not replace company analysis; it decides which stock deserves one. It works in two stages: the quick test sorts out roughly with nine metrics — quality only; the price (price against fair value, dividend yield) is shown separately and decides afterwards whether to buy —, the full rating scores eleven metrics in three categories — with industry-specific thresholds, an investment profile and a dividend score that measures yield and growth together against the current interest-rate level. The RSI is shown but not scored. The pre-filled values are a fictitious example company, not a real stock.

In daily use the rating runs with the same thresholds in the FBC WatchList, which derives the traffic light and buy price from it.

Stage 1

Quick test: nine metrics

Growth

All rates as CAGR (compound), never as the average of annual rates. EPS forecast 3 yrs = CAGR from the last full fiscal year's EPS to the estimate for the third year after it.

Quality

LT debt/capital as shown by F.A.S.T. Graphs. Above 100 % equity is negative (usually from buybacks): the metric is then not scored and the rating carries the weight.

Price and dividend

Price to fair value (P/E ÷ fair P/E) –

Fair P/E pre-filled rate-linked (100 ÷ (reference rate + premium), at most 15) — with earnings growth up to 5 % additionally at most 8.5 + 2 × growth; above 15 % enter the growth rate, or the normal P/E (average of the last 10 years) if that is your yardstick. Price to fair value and dividend yield do not count towards the points: the quick test measures quality, the price is judged separately below. Consecutive dividend increases count as quality.

Quick test points
–
Share of maximum
–
Grade
–
Metric5 P.4 P.3 P.2 P.1 P.Result
Stage 2

Full rating: eleven metrics, three categories

Settings

Sector according to GICS® (11 sectors). Finviz and Yahoo use their own classifications — e.g. “Consumer Defensive” instead of Consumer Staples; the GICS sector is shown e.g. in the company profile at S&P or MSCI. The thresholds per sector are my own values.

Financial strength

Balance-sheet quality

Valuation and price position

RSI · not scored
–
Freebird Rating total
–
Grade
–
MetricValueThresholds 5 / 4 / 3 / 2 ptsPointsWeight
Dividend relative to interest rates: yield plus growth Metric = dividend yield + dividend growth 5 yrs
5 points from reference rate + 5 pp · 4 points from reference rate + 3 pp · 3 points from reference rate + 1 pp · 2 points from reference rate − 1 pp

Why: a dividend yield is only as good as the risk-free alternative. But a fixed coupon does not grow, a dividend does: a 3.2 % yield growing at 6 % reaches 5 % on cost after about eight years. That is why yield and growth count together. For US stocks that is the 10-year US Treasury, for European stocks rather EURIBOR or the ECB deposit rate. Pre-filled values: US Treasury 10 yrs about 5.0 % (late October 2026), 3-month EURIBOR 2.67 %, ECB deposit rate 2.50 % since 16 Sep 2026, Fed funds 3.75–4.00 %. Current values: FRED (series DGS10), ECB and the European Money Markets Institute.

How the score is built

  1. Points: each metric receives 1 to 5 points. Growth, payout ratio and price-to-book are measured against sector-specific thresholds — a utility paying out 70 % is normal, a technology company is not. Negative values for P/E, PEG, price-to-book and payout ratio — that is, losses — score 1 point. Empty fields are not scored; percentages then refer to the scored metrics.
  2. Periods: growth is measured over 5 years as a CAGR (compound, not the average of annual rates); the EPS forecast over 3 years from the last full fiscal year's EPS. Five years react less to a single one-off increase than three. Watch start years with one-off effects such as 2020: the 5-year rate then sits well above the 10-year rate.
  3. Leverage: measured as long-term debt to total capital (LT debt/capital), as shown by F.A.S.T. Graphs. Converted from debt/equity, D/C = D/E ÷ (1 + D/E): the former thresholds 0.3 / 0.5 / 0.9 / 1.2 correspond to 23 / 33 / 47 / 55 %. Above 100 % equity is negative, usually from buybacks — the metric is then not scored and the credit rating carries the weight. For real-estate companies net debt to EBITDA applies.
  4. Weights: the investment profile shifts the weights. In the dividend profile, the payout ratio and the combined yield-plus-growth metric count triple; in the growth profile, earnings development does. Dividend growth on its own counts once, because it is already part of the combined metric.
  5. RSI: the 14-day RSI only shows the price position — below 30 oversold, above 70 overbought. It does not enter the score: market timing is not part of the method.
  6. Dividend safety: a dividend is paid from cash, not from earnings. The payout ratio is therefore measured twice — against earnings and against free cash flow (operating cash flow less capital expenditure; F.A.S.T. Graphs shows FCFE, which also includes net borrowing and reads somewhat friendlier for highly leveraged companies). The worse of the two values is scored, against the same sector thresholds. Real-estate companies are the exception: their earnings are distorted by depreciation, and the earnings-based ratio is often above 100 %. In the real-estate sector only the ratio against AFFO (FFO less recurring maintenance capital expenditure) counts, otherwise against FFO. FFO and free cash flow are different measures — depending on the sector, one or the other is entered. A negative free cash flow is entered as a negative number and scores 1 point; if the field is left empty, only the earnings-based ratio counts.
  7. Grade: as in the Freebird StockAnalyzer — share of maximum from 93 % exceptional, from 83 % excellent, from 67 % fine, from 50 % decent, from 33 % poor, below that inferior; from 50 % investment grade, below that speculative grade. The verdict (pass, borderline, out) follows each test's thresholds.
  8. Verdict: from 70 % of the maximum a stock is a candidate for in-depth analysis, between 55 and 70 % it goes on the watch list, below that it is out.

Then the real work follows: business model, return on capital (ROIC calculator) and a fair price (fair value calculator). How this fits the overall process is shown on the method page.

Own scoring model, developed 2017 to 2021; dividend-score thresholds aligned with the interest-rate level in September 2026; in October 2026 yield and growth were combined, the RSI was removed from the score and the payout ratio was extended by free cash flow; since v2.11 leverage as LT debt/capital and full-rating growth over 5 years; since v2.13 the quick test scores quality only (nine metrics), price to fair value and dividend yield are judged separately. Runs entirely in your browser; no inputs are stored or transmitted.

This article is for information and education only. It is not investment advice and not a recommendation to buy or sell any security. Model calculations simplify deliberately; past performance is not a reliable indicator of future results.
David Krause
David Krause
Graduate industrial engineer (Dipl.-Wirtschaftsingenieur), 15+ years of costing, cost accounting and plant controlling in manufacturing. Here he asks the same questions from the outside: what does a company earn on its capital, and what may it cost?