# Metric Decomposition

> Decompose composite metrics into constituent signals, analyze polarity and ceiling effects

- Skill: `yogsoth-ai/metric-decomposition-2` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx get-skill add yogsoth-ai/metric-decomposition-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/yogsoth-ai/metric-decomposition-2/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: yogsoth-ai (https://skillmd.com/u/yogsoth-ai)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/yogsoth-ai/metric-decomposition-2

---


# Metric Decomposition SOP

Decompose a benchmark's evaluation metric into its constituent signals to understand what is actually being measured, identify ceiling effects, and detect polarity issues.

## Input

- **metric_name**: Name of the metric (e.g., "F1", "BLEU", "Pass@k", "ELO")
- **metric_definition**: Mathematical definition or description of how the metric is computed
- **score_distribution**: Known score distribution data (top scores, median, floor)

## Procedure

1. Parse metric formula into component signals
2. Identify what each component rewards/penalizes
3. Analyze ceiling and floor effects
4. Check for known pathologies (metric gaming, Goodhart's law instances)
5. Assess granularity and discriminative power at current SOTA level

## Output

Structured decomposition showing constituent signals, their weights, ceiling effects, and known pathologies.

