{
  "label": "Exploratory April-frozen vs November-4-current score revision; not isolated Bayesian-methodology effect",
  "original_h1_n": 12312,
  "missing_tested_model_papers": {
    "DeepSeek Coder 1.3B": 1,
    "GLM-4.6": 2,
    "Gemini 1.0 Pro": 374,
    "Gemini 1.5 Pro (Feb 2024)": 28,
    "Gemini 2.0 Flash (Dec 2024)": 37,
    "MiniMax-M2.5": 1,
    "Mixtral": 14,
    "Yi-34B": 1
  },
  "missing_original_seat_papers": {
    "Gemini 1.5 Pro (Feb 2024)": 1039
  },
  "original_roster_n": 164,
  "original_dated_roster_n": 160,
  "common_dated_roster_n": 147,
  "original_seats": {
    "n": 10857,
    "old": {
      "n": 10857,
      "median": 12.633479706669732,
      "mean": 11.352915247031241,
      "iqr": [
        1.121294215469078,
        19.27929806916555
      ]
    },
    "new": {
      "n": 10857,
      "median": 15.219999999999999,
      "mean": 13.53072672008842,
      "iqr": [
        1.7700000000000102,
        22.940000000000012
      ]
    },
    "median_difference": 2.586520293330267,
    "median_paired_change": 1.6997735288576052,
    "mean_paired_change": 2.1778114730571807,
    "fraction_increased": 0.7204568481164226,
    "explicit": {
      "old": {
        "n": 613,
        "median": 5.18029213073541,
        "mean": 7.615756536171349,
        "iqr": [
          0.0,
          12.633479706669732
        ]
      },
      "new": {
        "n": 613,
        "median": 6.359999999999985,
        "mean": 9.073800978792821,
        "iqr": [
          0.0,
          15.61
        ]
      }
    },
    "h2": {
      "old": {
        "n": 10475,
        "n_journals": 2136,
        "pooled_year_slope": 5.262241309387801,
        "per_domain_slopes": {
          "coding": 3.228956746831076,
          "education": 6.731673111786597,
          "law": 4.2679449758480414,
          "medicine": 5.625394766109712,
          "other": 4.770498547336242,
          "scientific_reasoning": 4.678686595854528
        }
      },
      "new": {
        "n": 10475,
        "n_journals": 2136,
        "pooled_year_slope": 5.932261078028252,
        "per_domain_slopes": {
          "coding": 3.475863295981394,
          "education": 7.796267859337192,
          "law": 4.7332554931804625,
          "medicine": 6.358942754513312,
          "other": 5.292626499780646,
          "scientific_reasoning": 5.391285419048341
        }
      }
    },
    "domains": {
      "coding": {
        "old": {
          "n": 656,
          "median": 10.389223476354168,
          "mean": 10.209964202257497,
          "iqr": [
            0.35189110909587384,
            17.389800055124095
          ]
        },
        "new": {
          "n": 656,
          "median": 11.580000000000013,
          "mean": 12.096753048780487,
          "iqr": [
            0.7599999999999909,
            18.76000000000002
          ]
        }
      },
      "education": {
        "old": {
          "n": 1393,
          "median": 14.014702408030644,
          "mean": 15.352240276832056,
          "iqr": [
            7.112565179784156,
            23.086992458601927
          ]
        },
        "new": {
          "n": 1393,
          "median": 15.61,
          "mean": 18.656927494615935,
          "iqr": [
            8.129999999999995,
            29.12999999999998
          ]
        }
      },
      "law": {
        "old": {
          "n": 149,
          "median": 10.199662686231193,
          "mean": 9.645932447601906,
          "iqr": [
            1.9423567295133353,
            18.276481497599917
          ]
        },
        "new": {
          "n": 149,
          "median": 12.700000000000017,
          "mean": 11.637785234899328,
          "iqr": [
            1.7700000000000102,
            18.76000000000002
          ]
        }
      },
      "medicine": {
        "old": {
          "n": 4261,
          "median": 10.453512751932195,
          "mean": 10.944756202181615,
          "iqr": [
            1.121294215469078,
            18.795881877364167
          ]
        },
        "new": {
          "n": 4261,
          "median": 13.519999999999982,
          "mean": 12.977559258390048,
          "iqr": [
            1.7700000000000102,
            21.77000000000001
          ]
        }
      },
      "other": {
        "old": {
          "n": 3798,
          "median": 11.33825368721321,
          "mean": 11.296939804048915,
          "iqr": [
            1.1444477981726635,
            19.27929806916555
          ]
        },
        "new": {
          "n": 3798,
          "median": 14.099999999999994,
          "mean": 13.421453396524486,
          "iqr": [
            1.7700000000000102,
            21.969999999999985
          ]
        }
      },
      "scientific_reasoning": {
        "old": {
          "n": 600,
          "median": 5.18029213073541,
          "mean": 6.994276877906454,
          "iqr": [
            0.0,
            12.633479706669732
          ]
        },
        "new": {
          "n": 600,
          "median": 6.359999999999985,
          "mean": 8.287399999999998,
          "iqr": [
            0.0,
            15.61
          ]
        }
      }
    }
  },
  "common_frontier": {
    "historical_dates": {
      "n": 11854,
      "old": {
        "n": 11854,
        "median": 10.453512751932195,
        "mean": 10.701086043400668,
        "iqr": [
          0.6741662432047733,
          18.276481497599917
        ]
      },
      "new": {
        "n": 11854,
        "median": 13.519999999999982,
        "mean": 12.83778555761768,
        "iqr": [
          1.329999999999984,
          21.22
        ]
      },
      "median_difference": 3.066487248067787,
      "median_paired_change": 1.6997735288576052,
      "mean_paired_change": 2.136699514217013,
      "fraction_increased": 0.7126708284123503,
      "explicit": {
        "old": {
          "n": 689,
          "median": 3.831283040994961,
          "mean": 6.962460837789461,
          "iqr": [
            0.0,
            12.633479706669732
          ]
        },
        "new": {
          "n": 689,
          "median": 5.950000000000003,
          "mean": 8.391059506531205,
          "iqr": [
            0.0,
            15.61
          ]
        }
      },
      "h2": {
        "old": {
          "n": 11459,
          "n_journals": 2256,
          "pooled_year_slope": 5.48707895884091,
          "per_domain_slopes": {
            "coding": 3.4326290944443016,
            "education": 6.91246589105797,
            "law": 4.591220960224214,
            "medicine": 5.800679454274352,
            "other": 5.099159447490752,
            "scientific_reasoning": 4.715123321066888
          }
        },
        "new": {
          "n": 11459,
          "n_journals": 2256,
          "pooled_year_slope": 6.157628308689246,
          "per_domain_slopes": {
            "coding": 3.672777248571264,
            "education": 7.97498617966555,
            "law": 5.102125850009097,
            "medicine": 6.503211708606898,
            "other": 5.662581639929382,
            "scientific_reasoning": 5.4118516273804955
          }
        }
      },
      "domains": {
        "coding": {
          "old": {
            "n": 706,
            "median": 9.616777194910242,
            "mean": 9.783477377737572,
            "iqr": [
              0.0,
              16.464762747664693
            ]
          },
          "new": {
            "n": 706,
            "median": 11.090000000000003,
            "mean": 11.669858356940507,
            "iqr": [
              0.0,
              18.76000000000002
            ]
          }
        },
        "education": {
          "old": {
            "n": 1502,
            "median": 12.633479706669732,
            "mean": 14.67520998086582,
            "iqr": [
              4.43022647114239,
              21.989286894691432
            ]
          },
          "new": {
            "n": 1502,
            "median": 15.61,
            "mean": 17.9197603195739,
            "iqr": [
              7.159999999999997,
              25.539999999999992
            ]
          }
        },
        "law": {
          "old": {
            "n": 163,
            "median": 9.007578580949968,
            "mean": 9.006670902719048,
            "iqr": [
              0.8341810245478101,
              14.029615913156036
            ]
          },
          "new": {
            "n": 163,
            "median": 10.530000000000001,
            "mean": 10.891226993865029,
            "iqr": [
              0.8850000000000051,
              18.710000000000008
            ]
          }
        },
        "medicine": {
          "old": {
            "n": 4678,
            "median": 10.453512751932195,
            "mean": 10.274911494606279,
            "iqr": [
              0.44165004474854186,
              18.276481497599917
            ]
          },
          "new": {
            "n": 4678,
            "median": 13.519999999999982,
            "mean": 12.293101752885846,
            "iqr": [
              0.28000000000000114,
              19.36999999999999
            ]
          }
        },
        "other": {
          "old": {
            "n": 4160,
            "median": 10.453512751932195,
            "mean": 10.598458883190117,
            "iqr": [
              0.7896078289981574,
              18.276481497599917
            ]
          },
          "new": {
            "n": 4160,
            "median": 13.519999999999982,
            "mean": 12.65223798076923,
            "iqr": [
              1.75,
              20.0
            ]
          }
        },
        "scientific_reasoning": {
          "old": {
            "n": 645,
            "median": 3.831283040994961,
            "mean": 6.632032954334979,
            "iqr": [
              0.0,
              12.633479706669732
            ]
          },
          "new": {
            "n": 645,
            "median": 6.1299999999999955,
            "mean": 7.920930232558138,
            "iqr": [
              0.0,
              15.61
            ]
          }
        }
      }
    },
    "corrected_o3": {
      "n": 11854,
      "old": {
        "n": 11854,
        "median": 10.453512751932195,
        "mean": 10.152249779972209,
        "iqr": [
          0.44165004474854186,
          16.718717861210806
        ]
      },
      "new": {
        "n": 11854,
        "median": 12.930000000000007,
        "mean": 12.072864012147795,
        "iqr": [
          0.1699999999999875,
          19.289999999999992
        ]
      },
      "median_difference": 2.476487248067812,
      "median_paired_change": 1.4852379906156727,
      "mean_paired_change": 1.920614232175587,
      "fraction_increased": 0.6546313480681627,
      "explicit": {
        "old": {
          "n": 689,
          "median": 3.0900586829194197,
          "mean": 6.3186748471120096,
          "iqr": [
            0.0,
            12.633479706669732
          ]
        },
        "new": {
          "n": 689,
          "median": 4.579999999999984,
          "mean": 7.5064876632801125,
          "iqr": [
            0.0,
            15.61
          ]
        }
      },
      "h2": {
        "old": {
          "n": 11459,
          "n_journals": 2256,
          "pooled_year_slope": 5.253404676427834,
          "per_domain_slopes": {
            "coding": 3.2844006460566253,
            "education": 6.626795465763033,
            "law": 4.24380860217005,
            "medicine": 5.533875772894525,
            "other": 4.9070077086236505,
            "scientific_reasoning": 4.521530329518523
          }
        },
        "new": {
          "n": 11459,
          "n_journals": 2256,
          "pooled_year_slope": 5.832983109511605,
          "per_domain_slopes": {
            "coding": 3.4662585013266396,
            "education": 7.580253879335631,
            "law": 4.609366496414493,
            "medicine": 6.133373567130247,
            "other": 5.395612212396,
            "scientific_reasoning": 5.134863863112363
          }
        }
      },
      "domains": {
        "coding": {
          "old": {
            "n": 706,
            "median": 8.957432472689732,
            "mean": 9.357170871983515,
            "iqr": [
              0.0,
              14.11051447763715
            ]
          },
          "new": {
            "n": 706,
            "median": 9.264999999999986,
            "mean": 11.057195467422094,
            "iqr": [
              0.0,
              15.940000000000005
            ]
          }
        },
        "education": {
          "old": {
            "n": 1502,
            "median": 12.633479706669732,
            "mean": 14.082060084938856,
            "iqr": [
              3.831283040994961,
              21.989286894691432
            ]
          },
          "new": {
            "n": 1502,
            "median": 15.61,
            "mean": 17.100605858854856,
            "iqr": [
              7.159999999999997,
              25.539999999999992
            ]
          }
        },
        "law": {
          "old": {
            "n": 163,
            "median": 8.957432472689732,
            "mean": 8.391491001127095,
            "iqr": [
              0.8341810245478101,
              14.008729035684922
            ]
          },
          "new": {
            "n": 163,
            "median": 9.389999999999986,
            "mean": 10.02171779141104,
            "iqr": [
              0.08499999999999375,
              15.61
            ]
          }
        },
        "medicine": {
          "old": {
            "n": 4678,
            "median": 10.453512751932195,
            "mean": 9.683449612618768,
            "iqr": [
              0.0,
              16.718717861210806
            ]
          },
          "new": {
            "n": 4678,
            "median": 11.509999999999991,
            "mean": 11.472383497221031,
            "iqr": [
              0.0,
              19.180000000000007
            ]
          }
        },
        "other": {
          "old": {
            "n": 4160,
            "median": 10.453512751932195,
            "mean": 10.092843594802284,
            "iqr": [
              0.5496043957310803,
              16.718717861210806
            ]
          },
          "new": {
            "n": 4160,
            "median": 11.89500000000001,
            "mean": 11.946533653846153,
            "iqr": [
              0.28000000000000114,
              19.289999999999992
            ]
          }
        },
        "scientific_reasoning": {
          "old": {
            "n": 645,
            "median": 3.0900586829194197,
            "mean": 6.0994299739538445,
            "iqr": [
              0.0,
              12.633479706669732
            ]
          },
          "new": {
            "n": 645,
            "median": 3.969999999999999,
            "mean": 7.164821705426353,
            "iqr": [
              0.0,
              15.61
            ]
          }
        }
      }
    }
  },
  "limitations": [
    "Exact-name model identity only; no substituted snapshots or new aliases.",
    "No new papers or post-freeze model entrants.",
    "Scores are current rounded CSV point estimates; no Bayesian posterior propagated.",
    "Common-roster comparison excludes missing original models symmetrically from old/new frontier selection.",
    "Original first-of-month frontier convention retained, including potential negative gaps within release months.",
    "Historical-date arm retains documented o3 Dec20 announcement date; corrected arm uses Apr16 public release for both fits and imputed-date clipping."
  ],
  "analysis_date": "2026-11-04",
  "sources": {
    "eci": "https://epoch.ai/data/eci_scores.csv",
    "methodology": "https://epoch.ai/publications/eci-bayesian-methodology",
    "original_paper": "https://arxiv.org/abs/2605.04135",
    "o3_release": "https://openai.com/index/introducing-o3-and-o4-mini/"
  }
}
