Methodology: Management Quality Grade
What this grade is
The Management Quality grade is a 0–100 score (mapped to an A–F letter) that summarizes a few accounting ratios commonly used to read management effectiveness. It is a derived signal — computed from a small handful of profitability and leverage ratios — not a verdict on the people running the company.
Like every interpreted signal on Oxford Ledge, this grade is a starting point for your own work, not a substitute for it. Industry context matters; one quarter is not a trend; and a company that scores well on these inputs can still misallocate capital in ways the score does not see.
Honest framing — what this score reflects, and what it does not. The current implementation is a profitability and leverage screen, not a governance audit. It tells you whether the business is generating high returns on the capital management deploys and whether it is doing so without overstretching the balance sheet. It does not tell you about insider alignment, compensation structure, share-issuance discipline, or how management talks about capital returns in the proxy statement. Reading the proxy (DEF 14A) and the Form 4 filings is still the work.
The four components
The score is the average of up to four 0–25 component scores, drawn from ratios already on the ticker page. If fewer than two are available, the panel is hidden (the grade requires at least two inputs to be meaningful).
| Component | Max | What it measures | Code reference |
|---|---|---|---|
| ROIC vs WACC spread | 25 | Return on invested capital minus a 10% cost-of-capital default. The score is max(0, min(25, (ROIC − 10) × 2.5)). A wider positive spread is the cleanest evidence management is creating value rather than destroying it. The 10% is a placeholder until the platform carries a per-company WACC; a higher real WACC would lower this component’s score. |
management-score.ts:71–79 |
| Return on Equity | 25 | A bucketed score: ROE >20% earns 25 / >15% earns 20 / >10% earns 15 / >5% earns 10 / else 5. ROE is the cleanest test of how productively management is reinvesting retained earnings — the same denominator (shareholder equity) that the next component checks for over-leverage. | management-score.ts:81–88 |
| Debt management (Debt/Equity) | 25 | A bucketed score: D/E <0.3 earns 25 / <0.7 earns 20 / <1.5 earns 15 / <3 earns 10 / else 5. Conservative leverage is a precondition for surviving cycles; a great ROE produced by a balance sheet at 5× leverage is fragile in a way the ROE number alone does not show. | management-score.ts:90–96 |
| Net Margin | 25 | A bucketed score: net margin >20% earns 25 / >15% earns 20 / >10% earns 15 / >5% earns 10 / >0% earns 5 / else 0. Used as a proxy for pricing power and operating discipline — companies with durable competitive moats tend to sustain high net margins. | management-score.ts:98–104 |
Authoritative source: static/js/management-score.ts — calculateManagementScore(), lines 63–125. The component formulas, the bucket thresholds, and the letter-grade thresholds in the panel render are the same constants used here.
Score aggregation and the A–F letter
The 0–100 score is the unweighted average of the available component scores (each capped at 25), normalized to a 0–100 scale by the formula round(total ÷ (count × 25) × 100). There are no per-component weights beyond the equal 25-point caps — each available ratio contributes the same share.
The 0–100 score is mapped to a letter for readability:
- A — score ≥ 85
- B — score ≥ 70
- C — score ≥ 55
- D — score ≥ 40
- F — score < 40
High scores read green (A/B), middle scores amber (C), low scores red (D/F). The thresholds match common academic-style grading bands. The mapping lives at management-score.ts:113–117.
Honest disclosures (read this if you came from the panel)
- This is a profitability+leverage screen, not a governance audit. Insider ownership, compensation structure, DEF 14A analysis, share-repurchase quality, and free-cash-flow-vs-earnings divergence are not currently inputs to this grade. If you came here expecting the score to reflect those signals, it does not. Reading the proxy is still the work.
- WACC is a 10% default placeholder. The ROIC-vs-WACC spread uses a flat 10% as the cost-of-capital baseline. A real WACC computation (per-company, recomputed as the capital structure changes) is on the roadmap. For a utility (typical WACC ~6%) the score understates value creation; for a small-cap biotech (typical WACC ~14%) the score overstates it. Interpret the spread directionally.
- The panel renders with as few as 2 of the 4 metrics. If some ratios are missing from the ticker payload, the panel still renders using whatever it has — provided at least two of the four inputs are available. The “Based on N metrics” sub-label on the panel discloses this. The grade is hidden entirely when fewer than two metrics are available.
- Industry comparison is your job. A capital-light tech company will look better here than a capital-intensive industrial — that does not always mean the tech management is better. Compare within industry, not across.
- The score is not a buy/sell signal. A company with great profitability and a clean balance sheet can still trade above intrinsic value. Use this grade as one input among many.
Changelog
| Date | Change |
|---|---|
| 2026-05-27 | PROFESSOR methodology-accuracy sweep correction. The components section was fully rewritten to match the shipped algorithm. The prior stub (DESIGN+HS, 2026-05-27 AM) described a four-pillar governance framework (insider ownership, comp alignment, share repurchases, FCF-vs-earnings) that did not match the live code. The page now documents the actual inputs: ROIC vs WACC, Return on Equity, Debt/Equity, and Net Margin. Honesty disclosures added covering the WACC placeholder, the “not a governance audit” framing, and the partial-metric rendering. Source-of-truth: static/js/management-score.ts:63–125. |
| 2026-05-27 | CHAMP §5.3 (AM): initial publication. The grade panel ships with a methodology link to this page (one-click affordance for skeptical readers and pros who want to see what they’re looking at). |
Source code and corrections
This page reflects the algorithm in static/js/management-score.ts at the date shown at the top — specifically the calculateManagementScore() function at lines 63–125. The contract test tests/test_methodology_pin_contract.py pins specific claims here to specific lines in that source file; if either side drifts, CI fails.
How we keep every figure honest — the contract tests, freshness reviews, and public incident log behind the data — is documented in the Trust Dossier.
Corrections: oxfordledge@gmail.com.