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
The Rise of Linux Desktop for Professionals [not-audio_url] [/not-audio_url]

Duration: 1:27:28
Company: https://kfocus.org/My System: https://kfocus.org/spec/spec-m2Ecosystem: https://kfocus.org/land/business Live Stream: The Rise of Linux Desktop for Professionals Join industry experts Noah Gift and Michael Mikow…
Cattle, Not Pets: The Smartphone Trap [not-audio_url] [/not-audio_url]

Duration: 8:19
00:00:01 - Introduction to the concept of "pets vs. cattle" in the DevOps space00:00:33 - Similarities between smartphones, monolithic servers, and synchronous messaging00:01:02 - The problem with 24/7 availability and b…
Cuckoo Egg Dilemma:  Creator vs Consumer [not-audio_url] [/not-audio_url]

Duration: 5:27
00:00 - Introduction: The cuckoo egg dilemma in technology 01:00 - Explanation of the cuckoo bird's behavior and the analogy to technology 02:00 - The creator's perspective: Taylor Swift as an example 02:45 - The consume…
Why You Need A Repairable and Upgradeable Linux Laptop [not-audio_url] [/not-audio_url]

Duration: 8:08
00:00 - Introduction: Why you need a repairable Linux laptop 01:05 - The problem with non-upgradable laptops and being locked into ecosystems 02:30 - Introducing the Framework laptop as a repairable and customizable alte…
Launching World's Most Comprehensive Cloud Computing Program [not-audio_url] [/not-audio_url]

Duration: 3:33
00:00 - Introduction: Showcasing Pragmatic AI Labs and DS500's latest offering00:45 - Introducing the world's largest online cloud computing program01:30 - Host's background teaching cloud computing at top universities02…
Cloud Economics: Understanding AWS Pricing and Cost Optimization [not-audio_url] [/not-audio_url]

Duration: 15:11
edX ✨I build courses: https://insight.paiml.com/d69 📚edX Professional Certificate in Rust Programming: https://insight.paiml.com/tkg📚edX Rust Data Engineering: https://insight.paiml.com/fhd 📚edX Professional Certificate…
Year of the Dumbphone [not-audio_url] [/not-audio_url]

Duration: 9:28
00:00 - Introduction: Is 2024 the year of the dumb phone?01:23 - Host's background and relationship with media consumption 03:15 - How smartphones are like having a TV in your pocket 04:50 - The limited benefits of smart…
Leaving Apple Ecosystem in 2024 Livestream & Q/A Post WWDC [not-audio_url] [/not-audio_url]

Duration: 22:25
Noah Gift reacts to Apple's announced partnership with OpenAI, arguing it abandons Apple's core values and poses risks to user privacy and creator livelihoods. He shares his personal plan and advice for gradually transit…