Scoring Methodology

1. Core Principles

• Data-driven only. All scores are computed from structured data in the database — sub-scores, event records, and objectively measurable attributes. No human "opinion" scores are used for 99.3% of figures.
• Scale matters. Commanding 100,000 troops is not the same as commanding 1,000. Governing an empire of 60 million is not the same as a city-state of 50,000. We apply era and civilization scale factors.
• Anchored to consensus. 49 of the most historically recognized figures (from MIT Pantheon rankings, Wikipedia pageviews, and Google search volume) are manually scored by historians to establish the calibration baseline.
• No moral judgment. We measure military capability, political effectiveness, and cultural influence — not whether a figure was "good" or "evil." Historical impact is independent of moral character.
• Per-dimension independence. A brilliant military strategist who was a poor politician will have a high Strategy score and a low Political score. Strengths and weaknesses are preserved.

2. The Six Dimensions

3. Scale Factors

A Roman emperor governing 60 million subjects faced fundamentally different challenges than a medieval duke ruling 50,000. A modern general commanding 500,000 troops operates at a different scale than an ancient commander with 5,000. We incorporate scale through:

Note: Scale factors are intentionally mild (0.80—1.05 range). They provide nuance, not dominance — a brilliant strategist from a small state can still outrank a mediocre general from a large empire.

4. Anchor Calibration

We use historically prominent figures as calibration anchors. These anchors are scored using the same 6-dimension rubric, then all figures are scored relative to them using LLM-based evaluation. This ensures the scoring scale is grounded in real historical consensus, not arbitrary cutoffs.

Anchors are selected from MIT Pantheon rankings, Wikipedia pageview data, and Google Trends — representing diverse eras, civilizations, and occupations. Religious figures, artists, scientists, and philosophers are included alongside rulers and generals to ensure the scale works for all types of historical impact.

5. Limitations & Transparency

• Quantitative scoring is inherently reductive. Human greatness cannot be fully captured by six numbers. Our scores are a structured comparison tool, not a definitive judgment of worth.
• Historical records are incomplete. Figures from oral traditions or non-literate societies may be underrepresented. We attempt to correct for this through era/civilization scaling, but bias cannot be eliminated entirely.
• Occupation labels can misclassify. Some figures (e.g., Deng Xiaoping — labeled "Politician" but a military commander during the Long March) may have dimension scores that require manual adjustment. We document all such adjustments publicly.
• Scores do not represent moral approval. A high Military score does not glorify violence. A high Political score does not endorse a regime. We evaluate historical impact, not moral character.
• All scores are probabilistic estimates derived from sub-indicator data with known variance. They are best interpreted as ranges, not precise values.