ELO Ratings Questions

ELO Ratings Questions

Author: Noah Gift September 18, 2025 Duration: 3:39

Key Argument

  • Thesis: Using ELO for AI agent evaluation = measuring noise
  • Problem: Wrong evaluators, wrong metrics, wrong assumptions
  • Solution: Quantitative assessment frameworks

The Comparison (00:00-02:00)

Chess ELO

  • FIDE arbiters: 120hr training
  • Binary outcome: win/loss
  • Test-retest: r=0.95
  • Cohen's κ=0.92

AI Agent ELO

  • Random users: Google engineer? CS student? 10-year-old?
  • Undefined dimensions: accuracy? style? speed?
  • Test-retest: r=0.31 (coin flip)
  • Cohen's κ=0.42

Cognitive Bias Cascade (02:00-03:30)

  • Anchoring: 34% rating variance in first 3 seconds
  • Confirmation: 78% selective attention to preferred features
  • Dunning-Kruger: d=1.24 effect size
  • Result: Circular preferences (A>B>C>A)

The Quantitative Alternative (03:30-05:00)

Objective Metrics

  • McCabe complexity ≤20
  • Test coverage ≥80%
  • Big O notation comparison
  • Self-admitted technical debt
  • Reliability: r=0.91 vs r=0.42
  • Effect size: d=2.18

Dream Scenario vs Reality (05:00-06:00)

Dream

  • World's best engineers
  • Annotated metrics
  • Standardized criteria

Reality

  • Random internet users
  • No expertise verification
  • Subjective preferences

Key Statistics

MetricChessAI Agents
Inter-rater reliabilityκ=0.92κ=0.42
Test-retestr=0.95r=0.31
Temporal drift±10 pts±150 pts
Hurst exponent0.890.31

Takeaways

  1. Stop: Using preference votes as quality metrics
  2. Start: Automated complexity analysis
  3. ROI: 4.7 months to break even

Citations Mentioned

  • Kapoor et al. (2025): "AI agents that matter" - κ=0.42 finding
  • Santos et al. (2022): Technical Debt Grading validation
  • Regan & Haworth (2011): Chess arbiter reliability κ=0.92
  • Chapman & Johnson (2002): 34% anchoring effect

Quotable Moments

"You can't rate chess with basketball fans"

"0.31 reliability? That's a coin flip with extra steps"

"Every preference vote is a data crime"

"The psychometrics are screaming"


Resources

  • Technical Debt Grading (TDG) Framework
  • PMAT (Pragmatic AI Labs MCP Agent Toolkit)
  • McCabe Complexity Calculator
  • Cohen's Kappa Calculator

🔥 Hot Course Offers:

🚀 Level Up Your Career:

Learn end-to-end ML engineering from industry veterans at PAIML.COM


Noah Gift guides you through a year-long journey with 52 Weeks of Cloud, a weekly exploration designed for anyone building, managing, or simply curious about modern cloud infrastructure. Each episode digs into a specific technical topic, moving beyond surface-level explanations to offer practical insights you can apply. You’ll hear detailed discussions on the platforms that power the industry-like AWS, Azure, and Google Cloud-and how to navigate multi-cloud strategies effectively. The conversation regularly delves into the orchestration of these systems with Kubernetes and the specialized world of machine learning operations, or MLOps, including the integration and implications of large language models. This isn't just theory; it's a focused look at the tools and methodologies shaping how software is deployed and scaled today. By committing to this podcast, you're essentially getting a structured, expert-led curriculum that breaks down complex subjects into manageable weekly segments, all aimed at building a comprehensive and practical understanding of the cloud ecosystem.
Author: Language: English Episodes: 225

52 Weeks of Cloud
Podcast Episodes
European Digital Sovereignty: Breaking Tech Dependency [not-audio_url] [/not-audio_url]

Duration: 10:38
European Digital Sovereignty: Breaking Tech DependencyEpisode NotesHeterodox Economic Foundations (00:00-02:46)Current economic context: Income inequality at historic levels (worse than pre-French Revolution)Problems wit…
What is Web Assembly? [not-audio_url] [/not-audio_url]

Duration: 7:39
WebAssembly Core Concepts - Episode NotesIntroduction [00:00-00:14]Overview of episode focus: WebAssembly core conceptsStructure: definition, purpose, implementation pathwaysFundamental Definition [00:14-00:38]Low-level…
60,000 Times Slower Python [not-audio_url] [/not-audio_url]

Duration: 10:14
The End of Moore's Law and the Future of Computing PerformanceThe Automobile Industry Parallel1960s: Focus on power over efficiency (muscle cars, gas guzzlers)Evolution through Japanese efficiency, turbocharging, to elec…
Technical Architecture for Mobile Digital Independence [not-audio_url] [/not-audio_url]

Duration: 10:12
Technical Architecture for Digital IndependenceCore ConceptSmartphones represent a monolithic architecture that needs to be broken down into microservices for better digital independence.Authentication StrategyHardware s…
What I Cannot Create, I Do Not Understand [not-audio_url] [/not-audio_url]

Duration: 5:07
Feynman's Wisdom Applied to AI LearningBackgroundFeynman helped create atomic bomb and investigated Challenger disasterChallenger investigation revealed bureaucracy prioritized power over engineering solutionsTwo key phr…
Rise of Microcontainers [not-audio_url] [/not-audio_url]

Duration: 7:23
The Rise of Micro-Containers: When Less is MorePodcast Episode NotesOpening (0:00 - 0:40)Introduction to micro-containers: containers under 100KBContrast with typical Python containers (5GB+)Languages enabling micro-cont…
Software Engineering Job Postings in 2025 And What To Do About It [not-audio_url] [/not-audio_url]

Duration: 15:11
Software Development Job Market in 2025: Challenges & OpportunitiesMarket Downturn AnalysisInterest Rate ImpactFed rates rose from ~0% to 5%, ending era of "free money" for VCsJob postings dropped to COVID-era levels (in…
Container Size Optimization in 2025 [not-audio_url] [/not-audio_url]

Duration: 8:45
# Container Size Optimization in 2025 ## Core Motivation- Container size directly impacts cost efficiency- Python containers can reach 5GB- Sub-1MB containers enable: - Incredible performance - Microservice architecture…
Tech Regulatory Entrepreneurship and Alternative Governance Systems [not-audio_url] [/not-audio_url]

Duration: 20:54
Regulatory Entrepreneurship and Alternative Governance SystemsKey ConceptsRegulatory EntrepreneurshipCompanies building businesses that require changing laws to succeedExamples: Uber, Airbnb, Tesla, DraftKings, OpenAICor…
Websockets [not-audio_url] [/not-audio_url]

Duration: 8:03
WebSockets in Rust: From Theory to ImplementationEpisode Notes for Pragmatic Labs Technical Deep DiveIntroduction [00:00-00:45]WebSockets vs HTTP request-response pattern analogyReal-time communication model comparisonRu…