• About \ Suunnilleen

Thinking Metrics

Thinking Metrics

Monthly Archives: June 2026

CHESS ANALYTICS 00.2: Methods, Part 3: How to Read the Output Files

13 Saturday Jun 2026

Posted by Ripsu-sama in chess, chess analytics

≈ Leave a comment

Tags

chess, chess analytics, chess history, engine analysis, expected score, performance metrics, Stockfish, WDL evaluation, world championship

  • CHESS ANALYTICS 00.1: Methods, Parts 1 & 2: Measuring World-Championship Roads with Stockfish 18 WDL
  • CHESS ANALYTICS 00.0: List of Other Chess Analytics Articles
  1. From Numbers to Interpretation
  2. 1. The Main File Categories
  3. 2. First Rule: Know Whether the Number Is a Mean or a Sum
  4. 3. Second Rule: Separate Faced Difficulty from Caused Difficulty
  5. 4. Third Rule: Higher Does Not Always Mean Better
  6. 5. Fourth Rule: Raw CCP-v2 and Log CCP-v2 Are Different
  7. 6. Reading article_report.md
  8. 7. Reading all_metrics_3source.csv
  9. 8. Reading metric_ratios.csv
  10. 9. Reading player_summary.csv
  11. 10. Reading game_summary_readable.csv
  12. 11. Reading critical_moves.csv
  13. 12. Reading volatility_swings.csv
  14. 13. Reading hardest_positions.csv
  15. 14. Reading difficulty_critical_moves.csv
  16. 15. Reading ccp_hardest_positions.csv
  17. 16. Reading ccp_v2_log_hardest_positions.csv
  18. 17. Reading ccp_v2_log_critical_moves.csv
  19. 18. Reading ccp_v2_root_branch_summary.csv
  20. 19. Reading ccp_v2_root_branch_ply_summary.csv
  21. 20. Reading ccp_v2_root_branch_top_moves.csv
  22. 21. Reading ccp_v2_root_branch_payoff_moves.csv
  23. 22. Reading ccp_v2_root_branch_workload_rows.csv
  24. 23. Difficulty Share
  25. 24. Difficulty Pressure Quality, DPQ
  26. 25. Pressure Yield
  27. 26. Combining the Metrics: Common Interpretation Patterns
  28. 27. Style Profiles the Files Can Reveal
  29. 28. How to Compare Two Matches
  30. 29. A Simple Expert Reading Formula
  31. 30. Final Reading Checklist
  32. 31. The Main Lesson

From Numbers to Interpretation

The first methods article explained the WDL foundation: expected score, expected-score loss, WDL Accuracy, and the main performance metri.

The second methods article introduced the post-WDL difficulty layer: all-legal difficulty, CCP Search Difficulty, CCP-v2 recursive forcing burden, probe/confirm search, root-branch diagnostics, Difficulty Pressure Quality, and Pressure Yield.

This third part explains how to read the actual output files.

Continue reading →

Subscribe

  • Entries (RSS)
  • Comments (RSS)

Archives

  • June 2026
  • May 2026
  • December 2024
  • November 2024
  • January 2024
  • December 2023
  • November 2023
  • October 2023
  • June 2020
  • September 2019
  • August 2019
  • August 2018
  • April 2018
  • March 2018

Categories

  • AI theory
    • cognitive modeling
      • imagination engineering
  • chess
    • chess analytics
    • chess theory
  • cognitive architecture
    • thought operations
  • mathematics
    • physics
      • applied physics
  • mnemonics
  • philosophy
  • Uncategorized

Meta

  • Create account
  • Log in

Blog at WordPress.com.

  • Subscribe Subscribed
    • Thinking Metrics
    • Already have a WordPress.com account? Log in now.
    • Thinking Metrics
    • Subscribe Subscribed
    • Sign up
    • Log in
    • Report this content
    • View site in Reader
    • Manage subscriptions
    • Collapse this bar