{
  "canary_drift": {
    "endpoint_max_range": {
      "anthropic-claude45-haiku-anthropic": 0.0,
      "deepseek-v3-0324-crusoe-bf16": 0.0,
      "google-gemini25-flash-lite-vertex-eu": 0.0,
      "meta-llama31-8b-groq": 0.0,
      "meta-llama33-70b-groq": 0.0,
      "mistral-small32-24b-mistral-eu": 0.0,
      "openai-gpt41-mini-openai": 0.0,
      "openai-gpt41-nano-openai": 0.0,
      "qwen3-8b-alibaba": 0.2
    },
    "epochs": {
      "end": {
        "anthropic-claude45-haiku-anthropic": {
          "failure_rate": 0.4,
          "observed": 5
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "failure_rate": 0.2,
          "observed": 5
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "meta-llama31-8b-groq": {
          "failure_rate": 0.8,
          "observed": 5
        },
        "meta-llama33-70b-groq": {
          "failure_rate": 0.4,
          "observed": 5
        },
        "mistral-small32-24b-mistral-eu": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "openai-gpt41-mini-openai": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "openai-gpt41-nano-openai": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "qwen3-8b-alibaba": {
          "failure_rate": 0.0,
          "observed": 5
        }
      },
      "midpoint": {
        "anthropic-claude45-haiku-anthropic": {
          "failure_rate": 0.4,
          "observed": 5
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "failure_rate": 0.2,
          "observed": 5
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "meta-llama31-8b-groq": {
          "failure_rate": 0.8,
          "observed": 5
        },
        "meta-llama33-70b-groq": {
          "failure_rate": 0.4,
          "observed": 5
        },
        "mistral-small32-24b-mistral-eu": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "openai-gpt41-mini-openai": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "openai-gpt41-nano-openai": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "qwen3-8b-alibaba": {
          "failure_rate": 0.2,
          "observed": 5
        }
      },
      "start": {
        "anthropic-claude45-haiku-anthropic": {
          "failure_rate": 0.4,
          "observed": 5
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "failure_rate": 0.2,
          "observed": 5
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "meta-llama31-8b-groq": {
          "failure_rate": 0.8,
          "observed": 5
        },
        "meta-llama33-70b-groq": {
          "failure_rate": 0.4,
          "observed": 5
        },
        "mistral-small32-24b-mistral-eu": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "openai-gpt41-mini-openai": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "openai-gpt41-nano-openai": {
          "failure_rate": 0.6,
          "observed": 5
        },
        "qwen3-8b-alibaba": {
          "failure_rate": 0.2,
          "observed": 5
        }
      }
    }
  },
  "cell_states": {
    "accepted": 42813,
    "ambiguous": 334,
    "infrastructure_error": 188
  },
  "conditions": {
    "A_independent": {
      "complete_fraction": 0.90375,
      "complete_items": 723,
      "domain": {
        "arithmetic": {
          "failure_rate": 0.7408450704225352,
          "observed": 1420
        },
        "code": {
          "failure_rate": 0.48176718092566617,
          "observed": 1426
        },
        "constraint": {
          "failure_rate": 0.6208125445473984,
          "observed": 1403
        },
        "extraction": {
          "failure_rate": 0.2992957746478873,
          "observed": 1420
        },
        "tool_call": {
          "failure_rate": 0.5385684503127172,
          "observed": 1439
        }
      },
      "endpoint": {
        "anthropic-claude45-haiku-anthropic": {
          "complete_fraction": 0.98625,
          "failure_rate": 1.0,
          "failures": 789,
          "observed": 789,
          "passes": 0,
          "planned": 800
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "complete_fraction": 0.98875,
          "failure_rate": 0.24399494310998734,
          "failures": 193,
          "observed": 791,
          "passes": 598,
          "planned": 800
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "complete_fraction": 0.98,
          "failure_rate": 0.7602040816326531,
          "failures": 596,
          "observed": 784,
          "passes": 188,
          "planned": 800
        },
        "meta-llama31-8b-groq": {
          "complete_fraction": 0.98625,
          "failure_rate": 0.8643852978453739,
          "failures": 682,
          "observed": 789,
          "passes": 107,
          "planned": 800
        },
        "meta-llama33-70b-groq": {
          "complete_fraction": 0.98875,
          "failure_rate": 0.3805309734513274,
          "failures": 301,
          "observed": 791,
          "passes": 490,
          "planned": 800
        },
        "mistral-small32-24b-mistral-eu": {
          "complete_fraction": 0.98625,
          "failure_rate": 0.4068441064638783,
          "failures": 321,
          "observed": 789,
          "passes": 468,
          "planned": 800
        },
        "openai-gpt41-mini-openai": {
          "complete_fraction": 0.99,
          "failure_rate": 0.4166666666666667,
          "failures": 330,
          "observed": 792,
          "passes": 462,
          "planned": 800
        },
        "openai-gpt41-nano-openai": {
          "complete_fraction": 0.9925,
          "failure_rate": 0.6309823677581864,
          "failures": 501,
          "observed": 794,
          "passes": 293,
          "planned": 800
        },
        "qwen3-8b-alibaba": {
          "complete_fraction": 0.98625,
          "failure_rate": 0.1229404309252218,
          "failures": 97,
          "observed": 789,
          "passes": 692,
          "planned": 800
        }
      },
      "metrics": {
        "all_failure_rate": 0.017980636237897647,
        "covariance_diversification_index": 3.0124349819050766,
        "dependence_inflation": 2.6019828375825615,
        "diagonal_variance": 0.01905550208187859,
        "endpoints": 9,
        "fleet_variance": 0.04958208937856686,
        "items": 723,
        "majority_failure_rate": 0.49239280774550487,
        "mean_loss": 0.5343476256339328,
        "off_diagonal_covariance": 0.030526587296688272
      },
      "missingness_bounds": {
        "all_failure_high": 0.025,
        "all_failure_low": 0.01625,
        "items": 800,
        "majority_failure_high": 0.50625,
        "majority_failure_low": 0.4875,
        "mean_loss_high": 0.5419444444444445,
        "mean_loss_low": 0.5291666666666667,
        "missing_cells": 92
      },
      "nondegenerate_endpoint_pairs": 28,
      "nondegenerate_endpoints": 8,
      "universal_pass_rate": 0.0
    },
    "A_shared": {
      "complete_fraction": 0.91625,
      "complete_items": 733,
      "domain": {
        "arithmetic": {
          "failure_rate": 0.7469964664310954,
          "observed": 1415
        },
        "code": {
          "failure_rate": 0.47099930118798045,
          "observed": 1431
        },
        "constraint": {
          "failure_rate": 0.6311300639658849,
          "observed": 1407
        },
        "extraction": {
          "failure_rate": 0.3054187192118227,
          "observed": 1421
        },
        "tool_call": {
          "failure_rate": 0.5402777777777777,
          "observed": 1440
        }
      },
      "endpoint": {
        "anthropic-claude45-haiku-anthropic": {
          "complete_fraction": 0.98375,
          "failure_rate": 1.0,
          "failures": 787,
          "observed": 787,
          "passes": 0,
          "planned": 800
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "complete_fraction": 0.995,
          "failure_rate": 0.24120603015075376,
          "failures": 192,
          "observed": 796,
          "passes": 604,
          "planned": 800
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "complete_fraction": 0.99375,
          "failure_rate": 0.7735849056603774,
          "failures": 615,
          "observed": 795,
          "passes": 180,
          "planned": 800
        },
        "meta-llama31-8b-groq": {
          "complete_fraction": 0.99125,
          "failure_rate": 0.862547288776797,
          "failures": 684,
          "observed": 793,
          "passes": 109,
          "planned": 800
        },
        "meta-llama33-70b-groq": {
          "complete_fraction": 0.98625,
          "failure_rate": 0.37515842839036756,
          "failures": 296,
          "observed": 789,
          "passes": 493,
          "planned": 800
        },
        "mistral-small32-24b-mistral-eu": {
          "complete_fraction": 0.9925,
          "failure_rate": 0.4244332493702771,
          "failures": 337,
          "observed": 794,
          "passes": 457,
          "planned": 800
        },
        "openai-gpt41-mini-openai": {
          "complete_fraction": 0.97875,
          "failure_rate": 0.41890166028097064,
          "failures": 328,
          "observed": 783,
          "passes": 455,
          "planned": 800
        },
        "openai-gpt41-nano-openai": {
          "complete_fraction": 0.98875,
          "failure_rate": 0.629582806573957,
          "failures": 498,
          "observed": 791,
          "passes": 293,
          "planned": 800
        },
        "qwen3-8b-alibaba": {
          "complete_fraction": 0.9825,
          "failure_rate": 0.11959287531806616,
          "failures": 94,
          "observed": 786,
          "passes": 692,
          "planned": 800
        }
      },
      "metrics": {
        "all_failure_rate": 0.015006821282401092,
        "covariance_diversification_index": 3.0668976118807043,
        "dependence_inflation": 2.5565624622812404,
        "diagonal_variance": 0.019063052247534996,
        "endpoints": 9,
        "fleet_variance": 0.048735883792554,
        "items": 733,
        "majority_failure_rate": 0.49931787175989084,
        "mean_loss": 0.5361527967257845,
        "off_diagonal_covariance": 0.029672831545019006
      },
      "missingness_bounds": {
        "all_failure_high": 0.015,
        "all_failure_low": 0.01375,
        "items": 800,
        "majority_failure_high": 0.51375,
        "majority_failure_low": 0.49875,
        "mean_loss_high": 0.5440277777777778,
        "mean_loss_low": 0.5320833333333334,
        "missing_cells": 86
      },
      "nondegenerate_endpoint_pairs": 28,
      "nondegenerate_endpoints": 8,
      "universal_pass_rate": 0.0
    },
    "E_independent": {
      "complete_fraction": 0.90375,
      "complete_items": 723,
      "domain": {
        "arithmetic": {
          "failure_rate": 0.7584985835694051,
          "observed": 1412
        },
        "code": {
          "failure_rate": 0.47021723896285916,
          "observed": 1427
        },
        "constraint": {
          "failure_rate": 0.6407355021216408,
          "observed": 1414
        },
        "extraction": {
          "failure_rate": 0.3631875881523272,
          "observed": 1418
        },
        "tool_call": {
          "failure_rate": 0.4979166666666667,
          "observed": 1440
        }
      },
      "endpoint": {
        "anthropic-claude45-haiku-anthropic": {
          "complete_fraction": 0.9925,
          "failure_rate": 0.9634760705289672,
          "failures": 765,
          "observed": 794,
          "passes": 29,
          "planned": 800
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "complete_fraction": 0.99,
          "failure_rate": 0.20959595959595959,
          "failures": 166,
          "observed": 792,
          "passes": 626,
          "planned": 800
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "complete_fraction": 0.99125,
          "failure_rate": 0.9495586380832283,
          "failures": 753,
          "observed": 793,
          "passes": 40,
          "planned": 800
        },
        "meta-llama31-8b-groq": {
          "complete_fraction": 0.9925,
          "failure_rate": 0.809823677581864,
          "failures": 643,
          "observed": 794,
          "passes": 151,
          "planned": 800
        },
        "meta-llama33-70b-groq": {
          "complete_fraction": 0.98875,
          "failure_rate": 0.40075853350189633,
          "failures": 317,
          "observed": 791,
          "passes": 474,
          "planned": 800
        },
        "mistral-small32-24b-mistral-eu": {
          "complete_fraction": 0.98,
          "failure_rate": 0.3788265306122449,
          "failures": 297,
          "observed": 784,
          "passes": 487,
          "planned": 800
        },
        "openai-gpt41-mini-openai": {
          "complete_fraction": 0.9875,
          "failure_rate": 0.3911392405063291,
          "failures": 309,
          "observed": 790,
          "passes": 481,
          "planned": 800
        },
        "openai-gpt41-nano-openai": {
          "complete_fraction": 0.9875,
          "failure_rate": 0.6341772151898735,
          "failures": 501,
          "observed": 790,
          "passes": 289,
          "planned": 800
        },
        "qwen3-8b-alibaba": {
          "complete_fraction": 0.97875,
          "failure_rate": 0.16475095785440613,
          "failures": 129,
          "observed": 783,
          "passes": 654,
          "planned": 800
        }
      },
      "metrics": {
        "all_failure_rate": 0.017980636237897647,
        "covariance_diversification_index": 3.5587445402589903,
        "dependence_inflation": 2.357781831260916,
        "diagonal_variance": 0.01841140561129735,
        "endpoints": 9,
        "fleet_variance": 0.04341007763829216,
        "items": 723,
        "majority_failure_rate": 0.48409405255878285,
        "mean_loss": 0.5378822806208698,
        "off_diagonal_covariance": 0.024998672026994812
      },
      "missingness_bounds": {
        "all_failure_high": 0.02625,
        "all_failure_low": 0.01625,
        "items": 800,
        "majority_failure_high": 0.515,
        "majority_failure_low": 0.49375,
        "mean_loss_high": 0.55125,
        "mean_loss_low": 0.5388888888888889,
        "missing_cells": 89
      },
      "nondegenerate_endpoint_pairs": 36,
      "nondegenerate_endpoints": 9,
      "universal_pass_rate": 0.0
    },
    "E_shared": {
      "complete_fraction": 0.89625,
      "complete_items": 717,
      "domain": {
        "arithmetic": {
          "failure_rate": 0.7684135977337111,
          "observed": 1412
        },
        "code": {
          "failure_rate": 0.4740532959326788,
          "observed": 1426
        },
        "constraint": {
          "failure_rate": 0.6360424028268551,
          "observed": 1415
        },
        "extraction": {
          "failure_rate": 0.36325088339222616,
          "observed": 1415
        },
        "tool_call": {
          "failure_rate": 0.4822546972860125,
          "observed": 1437
        }
      },
      "endpoint": {
        "anthropic-claude45-haiku-anthropic": {
          "complete_fraction": 0.9875,
          "failure_rate": 0.9443037974683545,
          "failures": 746,
          "observed": 790,
          "passes": 44,
          "planned": 800
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "complete_fraction": 0.9875,
          "failure_rate": 0.21772151898734177,
          "failures": 172,
          "observed": 790,
          "passes": 618,
          "planned": 800
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "complete_fraction": 0.99125,
          "failure_rate": 0.9495586380832283,
          "failures": 753,
          "observed": 793,
          "passes": 40,
          "planned": 800
        },
        "meta-llama31-8b-groq": {
          "complete_fraction": 0.9875,
          "failure_rate": 0.8177215189873418,
          "failures": 646,
          "observed": 790,
          "passes": 144,
          "planned": 800
        },
        "meta-llama33-70b-groq": {
          "complete_fraction": 0.99375,
          "failure_rate": 0.3987421383647799,
          "failures": 317,
          "observed": 795,
          "passes": 478,
          "planned": 800
        },
        "mistral-small32-24b-mistral-eu": {
          "complete_fraction": 0.98375,
          "failure_rate": 0.3799237611181703,
          "failures": 299,
          "observed": 787,
          "passes": 488,
          "planned": 800
        },
        "openai-gpt41-mini-openai": {
          "complete_fraction": 0.98625,
          "failure_rate": 0.41825095057034223,
          "failures": 330,
          "observed": 789,
          "passes": 459,
          "planned": 800
        },
        "openai-gpt41-nano-openai": {
          "complete_fraction": 0.98375,
          "failure_rate": 0.5972045743329097,
          "failures": 470,
          "observed": 787,
          "passes": 317,
          "planned": 800
        },
        "qwen3-8b-alibaba": {
          "complete_fraction": 0.98,
          "failure_rate": 0.17219387755102042,
          "failures": 135,
          "observed": 784,
          "passes": 649,
          "planned": 800
        }
      },
      "metrics": {
        "all_failure_rate": 0.01813110181311018,
        "covariance_diversification_index": 3.349149332138783,
        "dependence_inflation": 2.5330299020367324,
        "diagonal_variance": 0.018936476453962146,
        "endpoints": 9,
        "fleet_variance": 0.04796666109710063,
        "items": 717,
        "majority_failure_rate": 0.49232914923291493,
        "mean_loss": 0.5378893537889353,
        "off_diagonal_covariance": 0.02903018464313848
      },
      "missingness_bounds": {
        "all_failure_high": 0.0275,
        "all_failure_low": 0.01625,
        "items": 800,
        "majority_failure_high": 0.50875,
        "majority_failure_low": 0.49125,
        "mean_loss_high": 0.5504166666666667,
        "mean_loss_low": 0.5372222222222223,
        "missing_cells": 95
      },
      "nondegenerate_endpoint_pairs": 36,
      "nondegenerate_endpoints": 9,
      "universal_pass_rate": 0.0
    },
    "N": {
      "complete_fraction": 0.91625,
      "complete_items": 733,
      "domain": {
        "arithmetic": {
          "failure_rate": 0.703886925795053,
          "observed": 1415
        },
        "code": {
          "failure_rate": 0.41596638655462187,
          "observed": 1428
        },
        "constraint": {
          "failure_rate": 0.6734982332155477,
          "observed": 1415
        },
        "extraction": {
          "failure_rate": 0.258765778401122,
          "observed": 1426
        },
        "tool_call": {
          "failure_rate": 0.4100069492703266,
          "observed": 1439
        }
      },
      "endpoint": {
        "anthropic-claude45-haiku-anthropic": {
          "complete_fraction": 0.99375,
          "failure_rate": 0.6754716981132075,
          "failures": 537,
          "observed": 795,
          "passes": 258,
          "planned": 800
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "complete_fraction": 0.98375,
          "failure_rate": 0.30241423125794153,
          "failures": 238,
          "observed": 787,
          "passes": 549,
          "planned": 800
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "complete_fraction": 0.995,
          "failure_rate": 0.41708542713567837,
          "failures": 332,
          "observed": 796,
          "passes": 464,
          "planned": 800
        },
        "meta-llama31-8b-groq": {
          "complete_fraction": 0.99125,
          "failure_rate": 0.7440100882723834,
          "failures": 590,
          "observed": 793,
          "passes": 203,
          "planned": 800
        },
        "meta-llama33-70b-groq": {
          "complete_fraction": 0.99,
          "failure_rate": 0.3888888888888889,
          "failures": 308,
          "observed": 792,
          "passes": 484,
          "planned": 800
        },
        "mistral-small32-24b-mistral-eu": {
          "complete_fraction": 0.9875,
          "failure_rate": 0.43291139240506327,
          "failures": 342,
          "observed": 790,
          "passes": 448,
          "planned": 800
        },
        "openai-gpt41-mini-openai": {
          "complete_fraction": 0.985,
          "failure_rate": 0.5532994923857868,
          "failures": 436,
          "observed": 788,
          "passes": 352,
          "planned": 800
        },
        "openai-gpt41-nano-openai": {
          "complete_fraction": 0.99125,
          "failure_rate": 0.6670870113493065,
          "failures": 529,
          "observed": 793,
          "passes": 264,
          "planned": 800
        },
        "qwen3-8b-alibaba": {
          "complete_fraction": 0.98625,
          "failure_rate": 0.24081115335868186,
          "failures": 190,
          "observed": 789,
          "passes": 599,
          "planned": 800
        }
      },
      "metrics": {
        "all_failure_rate": 0.06002728512960437,
        "covariance_diversification_index": 2.471148367351597,
        "dependence_inflation": 3.6329691430612656,
        "diagonal_variance": 0.02473240628686359,
        "endpoints": 9,
        "fleet_variance": 0.08985206887382988,
        "items": 733,
        "majority_failure_rate": 0.44747612551159616,
        "mean_loss": 0.48719114749128406,
        "off_diagonal_covariance": 0.06511966258696629
      },
      "missingness_bounds": {
        "all_failure_high": 0.06375,
        "all_failure_low": 0.055,
        "items": 800,
        "majority_failure_high": 0.46125,
        "majority_failure_low": 0.4525,
        "mean_loss_high": 0.4970833333333333,
        "mean_loss_low": 0.48638888888888887,
        "missing_cells": 77
      },
      "nondegenerate_endpoint_pairs": 36,
      "nondegenerate_endpoints": 9,
      "universal_pass_rate": 0.054570259208731244
    },
    "O": {
      "complete_fraction": 0.92,
      "complete_items": 736,
      "domain": {
        "arithmetic": {
          "failure_rate": 0.6916019760056458,
          "observed": 1417
        },
        "code": {
          "failure_rate": 0.34507042253521125,
          "observed": 1420
        },
        "constraint": {
          "failure_rate": 0.6386732533521524,
          "observed": 1417
        },
        "extraction": {
          "failure_rate": 0.19676739283204497,
          "observed": 1423
        },
        "tool_call": {
          "failure_rate": 0.3902777777777778,
          "observed": 1440
        }
      },
      "endpoint": {
        "anthropic-claude45-haiku-anthropic": {
          "complete_fraction": 0.9875,
          "failure_rate": 0.44556962025316454,
          "failures": 352,
          "observed": 790,
          "passes": 438,
          "planned": 800
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "complete_fraction": 0.995,
          "failure_rate": 0.27889447236180903,
          "failures": 222,
          "observed": 796,
          "passes": 574,
          "planned": 800
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "complete_fraction": 0.98625,
          "failure_rate": 0.4068441064638783,
          "failures": 321,
          "observed": 789,
          "passes": 468,
          "planned": 800
        },
        "meta-llama31-8b-groq": {
          "complete_fraction": 0.99125,
          "failure_rate": 0.7187894073139974,
          "failures": 570,
          "observed": 793,
          "passes": 223,
          "planned": 800
        },
        "meta-llama33-70b-groq": {
          "complete_fraction": 0.98125,
          "failure_rate": 0.38980891719745225,
          "failures": 306,
          "observed": 785,
          "passes": 479,
          "planned": 800
        },
        "mistral-small32-24b-mistral-eu": {
          "complete_fraction": 0.99,
          "failure_rate": 0.4431818181818182,
          "failures": 351,
          "observed": 792,
          "passes": 441,
          "planned": 800
        },
        "openai-gpt41-mini-openai": {
          "complete_fraction": 0.98875,
          "failure_rate": 0.5562579013906448,
          "failures": 440,
          "observed": 791,
          "passes": 351,
          "planned": 800
        },
        "openai-gpt41-nano-openai": {
          "complete_fraction": 0.9875,
          "failure_rate": 0.6151898734177215,
          "failures": 486,
          "observed": 790,
          "passes": 304,
          "planned": 800
        },
        "qwen3-8b-alibaba": {
          "complete_fraction": 0.98875,
          "failure_rate": 0.213653603034134,
          "failures": 169,
          "observed": 791,
          "passes": 622,
          "planned": 800
        }
      },
      "metrics": {
        "all_failure_rate": 0.036684782608695655,
        "covariance_diversification_index": 2.430763178613384,
        "dependence_inflation": 3.687600457036629,
        "diagonal_variance": 0.02494716754119795,
        "endpoints": 9,
        "fleet_variance": 0.09199518642669093,
        "items": 736,
        "majority_failure_rate": 0.40896739130434784,
        "mean_loss": 0.4455012077294686,
        "off_diagonal_covariance": 0.06704801888549297
      },
      "missingness_bounds": {
        "all_failure_high": 0.0425,
        "all_failure_low": 0.03375,
        "items": 800,
        "majority_failure_high": 0.4225,
        "majority_failure_low": 0.41375,
        "mean_loss_high": 0.4583333333333333,
        "mean_loss_low": 0.44680555555555557,
        "missing_cells": 83
      },
      "nondegenerate_endpoint_pairs": 36,
      "nondegenerate_endpoints": 9,
      "universal_pass_rate": 0.06793478260869565
    }
  },
  "difficulty_residual_sensitivity": {
    "contrasts": {
      "evidence_corruption": {
        "independent": {
          "dependence_inflation": 1.051734457601712,
          "diagonal_variance": 0.017043279057587603,
          "fleet_variance": 0.017925003855386513
        },
        "log_dependence_inflation_ratio": 0.22543111689018086,
        "matched_items": 629,
        "shared": {
          "dependence_inflation": 1.3176789038521655,
          "diagonal_variance": 0.017219481681446543,
          "fleet_variance": 0.022689747746910925
        }
      },
      "instruction_anchor": {
        "independent": {
          "dependence_inflation": 1.1570304688177904,
          "diagonal_variance": 0.016358457647804935,
          "fleet_variance": 0.018927233921375713
        },
        "log_dependence_inflation_ratio": -0.041693361359673234,
        "matched_items": 646,
        "shared": {
          "dependence_inflation": 1.1097818015641485,
          "diagonal_variance": 0.016513841283532227,
          "fleet_variance": 0.018326760530382805
        }
      }
    },
    "method": "five-fold item-family cross-fit; leave-endpoint-out O difficulty plus frozen domain and predicted-band features; ridge logistic endpoint calibration",
    "prediction_support_fraction": 0.9884722222222222,
    "schema_version": "cmfc.common-shock-residual-sensitivity.v1"
  },
  "endpoints": [
    "openai-gpt41-nano-openai",
    "openai-gpt41-mini-openai",
    "google-gemini25-flash-lite-vertex-eu",
    "anthropic-claude45-haiku-anthropic",
    "meta-llama31-8b-groq",
    "meta-llama33-70b-groq",
    "qwen3-8b-alibaba",
    "mistral-small32-24b-mistral-eu",
    "deepseek-v3-0324-crusoe-bf16"
  ],
  "execution_id": "common-shock-confirmatory-v11-transport-recovery-20260827",
  "metric_boundaries": {
    "absolute_volatility_ratio": "shared fleet SD divided by independent fleet SD",
    "capital_claim": "none",
    "dependence_inflation": "fleet variance divided by condition-specific diagonal variance",
    "n_eff": "secondary benchmark-relative covariance-diversification index"
  },
  "neutral_equivalence": {
    "dependence_inflation_ratio_N_over_O": 0.9949628435283622,
    "dependence_ratio_margin": [
      0.9,
      1.1
    ],
    "dependence_ratio_within_margin": true,
    "matched_items": 684,
    "mean_difference_O_minus_N": -0.03979857050032487,
    "mean_margin": [
      -0.03,
      0.03
    ],
    "mean_within_margin": false
  },
  "planned_base_items": 800,
  "primary_contrasts": {
    "evidence_corruption": {
      "absolute_volatility_ratio": 1.0600081254773286,
      "all_failure_difference": 0.006015037593984963,
      "dependence_inflation_ratio": 1.0951136227087999,
      "independent_condition": "E_independent",
      "log_dependence_inflation_ratio": 0.09085812291611317,
      "majority_failure_difference": 0.001503759398496285,
      "matched_fraction": 0.83125,
      "matched_items": 665,
      "mean_loss_difference": 0.0018379281537176384,
      "mean_loss_difference_bounds": {
        "high": 0.011527777777777803,
        "low": -0.01402777777777775
      },
      "per_endpoint_induced_failure": {
        "anthropic-claude45-haiku-anthropic": {
          "independent_failure_given_shared_pass": 0.7674418604651163,
          "shared_failure_given_independent_pass": 0.6551724137931034
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "independent_failure_given_shared_pass": 0.04905660377358491,
          "shared_failure_given_independent_pass": 0.05970149253731343
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "independent_failure_given_shared_pass": 0.0,
          "shared_failure_given_independent_pass": 0.0
        },
        "meta-llama31-8b-groq": {
          "independent_failure_given_shared_pass": 0.6299212598425197,
          "shared_failure_given_independent_pass": 0.6492537313432836
        },
        "meta-llama33-70b-groq": {
          "independent_failure_given_shared_pass": 0.058394160583941604,
          "shared_failure_given_independent_pass": 0.06295399515738499
        },
        "mistral-small32-24b-mistral-eu": {
          "independent_failure_given_shared_pass": 0.06976744186046512,
          "shared_failure_given_independent_pass": 0.07621247113163972
        },
        "openai-gpt41-mini-openai": {
          "independent_failure_given_shared_pass": 0.04533333333333334,
          "shared_failure_given_independent_pass": 0.089058524173028
        },
        "openai-gpt41-nano-openai": {
          "independent_failure_given_shared_pass": 0.3154121863799283,
          "shared_failure_given_independent_pass": 0.2625482625482625
        },
        "qwen3-8b-alibaba": {
          "independent_failure_given_shared_pass": 0.09107468123861566,
          "shared_failure_given_independent_pass": 0.1057347670250896
        }
      },
      "planned_items": 800,
      "shared_condition": "E_shared",
      "support_by_domain": {
        "arithmetic": {
          "matched_items": 115,
          "planned_items": 160
        },
        "code": {
          "matched_items": 142,
          "planned_items": 160
        },
        "constraint": {
          "matched_items": 128,
          "planned_items": 160
        },
        "extraction": {
          "matched_items": 123,
          "planned_items": 160
        },
        "tool_call": {
          "matched_items": 157,
          "planned_items": 160
        }
      },
      "support_classification": "estimable_degraded"
    },
    "instruction_anchor": {
      "absolute_volatility_ratio": 0.9893697369608665,
      "all_failure_difference": -0.005865102639296188,
      "dependence_inflation_ratio": 0.9787596741380908,
      "independent_condition": "A_independent",
      "log_dependence_inflation_ratio": -0.02146914754912347,
      "majority_failure_difference": 0.01026392961876832,
      "matched_fraction": 0.8525,
      "matched_items": 682,
      "mean_loss_difference": 0.0029325513196480912,
      "mean_loss_difference_bounds": {
        "high": 0.014861111111111103,
        "low": -0.009861111111111098
      },
      "per_endpoint_induced_failure": {
        "anthropic-claude45-haiku-anthropic": {
          "independent_failure_given_shared_pass": null,
          "shared_failure_given_independent_pass": null
        },
        "deepseek-v3-0324-crusoe-bf16": {
          "independent_failure_given_shared_pass": 0.06395348837209303,
          "shared_failure_given_independent_pass": 0.06936416184971098
        },
        "google-gemini25-flash-lite-vertex-eu": {
          "independent_failure_given_shared_pass": 0.27972027972027974,
          "shared_failure_given_independent_pass": 0.33548387096774196
        },
        "meta-llama31-8b-groq": {
          "independent_failure_given_shared_pass": 0.3173076923076923,
          "shared_failure_given_independent_pass": 0.29
        },
        "meta-llama33-70b-groq": {
          "independent_failure_given_shared_pass": 0.04640371229698376,
          "shared_failure_given_independent_pass": 0.04640371229698376
        },
        "mistral-small32-24b-mistral-eu": {
          "independent_failure_given_shared_pass": 0.09429280397022333,
          "shared_failure_given_independent_pass": 0.11407766990291263
        },
        "openai-gpt41-mini-openai": {
          "independent_failure_given_shared_pass": 0.06753246753246753,
          "shared_failure_given_independent_pass": 0.06266318537859007
        },
        "openai-gpt41-nano-openai": {
          "independent_failure_given_shared_pass": 0.07307692307692308,
          "shared_failure_given_independent_pass": 0.0694980694980695
        },
        "qwen3-8b-alibaba": {
          "independent_failure_given_shared_pass": 0.0667779632721202,
          "shared_failure_given_independent_pass": 0.06833333333333333
        }
      },
      "planned_items": 800,
      "shared_condition": "A_shared",
      "support_by_domain": {
        "arithmetic": {
          "matched_items": 127,
          "planned_items": 160
        },
        "code": {
          "matched_items": 141,
          "planned_items": 160
        },
        "constraint": {
          "matched_items": 129,
          "planned_items": 160
        },
        "extraction": {
          "matched_items": 126,
          "planned_items": 160
        },
        "tool_call": {
          "matched_items": 159,
          "planned_items": 160
        }
      },
      "support_classification": "estimable_degraded"
    }
  },
  "provider_availability": {
    "ambiguous_events": 334,
    "boundary": "retry/missing delivery diagnostics are not recoded as model failures or pooled into failure dependence",
    "endpoint": {
      "anthropic-claude45-haiku-anthropic": {
        "accepted": 4760,
        "error_events": 93,
        "missing_cells": 55,
        "planned": 4815,
        "retry_cells": 62,
        "retry_failure_phi_on_accepted": 0.019386223389020848
      },
      "deepseek-v3-0324-crusoe-bf16": {
        "accepted": 4767,
        "error_events": 151,
        "missing_cells": 48,
        "planned": 4815,
        "retry_cells": 108,
        "retry_failure_phi_on_accepted": 0.02461047556268222
      },
      "google-gemini25-flash-lite-vertex-eu": {
        "accepted": 4765,
        "error_events": 87,
        "missing_cells": 50,
        "planned": 4815,
        "retry_cells": 59,
        "retry_failure_phi_on_accepted": 0.014101413277160574
      },
      "meta-llama31-8b-groq": {
        "accepted": 4767,
        "error_events": 787,
        "missing_cells": 48,
        "planned": 4815,
        "retry_cells": 735,
        "retry_failure_phi_on_accepted": 0.02572856221954884
      },
      "meta-llama33-70b-groq": {
        "accepted": 4758,
        "error_events": 763,
        "missing_cells": 57,
        "planned": 4815,
        "retry_cells": 688,
        "retry_failure_phi_on_accepted": 0.06513677687951051
      },
      "mistral-small32-24b-mistral-eu": {
        "accepted": 4751,
        "error_events": 84,
        "missing_cells": 64,
        "planned": 4815,
        "retry_cells": 57,
        "retry_failure_phi_on_accepted": 0.045524733638789276
      },
      "openai-gpt41-mini-openai": {
        "accepted": 4748,
        "error_events": 87,
        "missing_cells": 67,
        "planned": 4815,
        "retry_cells": 59,
        "retry_failure_phi_on_accepted": -0.01070812828805044
      },
      "openai-gpt41-nano-openai": {
        "accepted": 4760,
        "error_events": 88,
        "missing_cells": 55,
        "planned": 4815,
        "retry_cells": 61,
        "retry_failure_phi_on_accepted": 0.029803061011850998
      },
      "qwen3-8b-alibaba": {
        "accepted": 4737,
        "error_events": 183,
        "missing_cells": 78,
        "planned": 4815,
        "retry_cells": 123,
        "retry_failure_phi_on_accepted": 0.003045405314212801
      }
    },
    "quarantine_events": 0,
    "schema_version": "cmfc.common-shock-availability-diagnostic.v1",
    "terminal_error_types": {
      "ConnectError": 1825,
      "HTTP_429": 312,
      "HTTP_503_error_envelope": 1,
      "ReadError": 239,
      "RemoteProtocolError": 186,
      "TimeoutError": 31,
      "orphaned_started_attempt": 63
    }
  },
  "schema_version": "cmfc.common-shock-analysis.v2",
  "simultaneous_inference": {
    "confidence": 0.95,
    "critical_value": 2.2462132172449136,
    "finite_joint_draws": 9999,
    "intervals": {
      "evidence_corruption": {
        "estimate": 0.09085812291611317,
        "finite_draws": 9999,
        "high": 0.16602846614347455,
        "low": 0.01568777968875179,
        "standard_error": 0.03346536412939523
      },
      "instruction_anchor": {
        "estimate": -0.02146914754912347,
        "finite_draws": 9999,
        "high": 0.035212698179488,
        "low": -0.07815099327773495,
        "standard_error": 0.025234401299683573
      }
    },
    "method": "studentized max-absolute-statistic item-family-by-stratum intersection cluster bootstrap",
    "resamples": 9999,
    "schema_version": "cmfc.common-shock-simultaneous-inference.v1",
    "seed": 20260823,
    "tail_probability_monte_carlo_se": 0.002179558452417543
  },
  "subgroup_sensitivity": {
    "heterogeneity": {
      "A_shared_variant": {
        "a_shared-00": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.2191820347416777,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.4399405856167675,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.3646018527483366,
            "majority_failure_difference": -0.08695652173913049,
            "matched_fraction": 0.8625,
            "matched_items": 69,
            "mean_loss_difference": -0.037037037037036924,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.913411215982746,
            "all_failure_difference": -0.028169014084507043,
            "dependence_inflation_ratio": 0.8364500470564334,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.17858847694753113,
            "majority_failure_difference": 0.07042253521126762,
            "matched_fraction": 0.8875,
            "matched_items": 71,
            "mean_loss_difference": 0.014084507042253391,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "a_shared-01": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0315663482135862,
            "all_failure_difference": 0.047619047619047616,
            "dependence_inflation_ratio": 1.1303354797969178,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.12251447343871816,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.7875,
            "matched_items": 63,
            "mean_loss_difference": 0.012345679012345845,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9734143384163312,
            "all_failure_difference": -0.014705882352941176,
            "dependence_inflation_ratio": 0.956005370342136,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.04499174843381942,
            "majority_failure_difference": 0.07352941176470584,
            "matched_fraction": 0.85,
            "matched_items": 68,
            "mean_loss_difference": 0.02941176470588258,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "a_shared-02": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.4126976484117477,
            "all_failure_difference": 0.015384615384615385,
            "dependence_inflation_ratio": 1.858101431337508,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.619555230579459,
            "majority_failure_difference": 0.015384615384615385,
            "matched_fraction": 0.8125,
            "matched_items": 65,
            "mean_loss_difference": -0.017094017094017144,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.1199234403538731,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.2052680823764963,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.18670201719996685,
            "majority_failure_difference": -0.02898550724637683,
            "matched_fraction": 0.8625,
            "matched_items": 69,
            "mean_loss_difference": -0.02576489533011256,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "a_shared-03": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0865173421150336,
            "all_failure_difference": 0.015151515151515152,
            "dependence_inflation_ratio": 1.1163891255730451,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.11009948201367657,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": -0.008417508417508324,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.073323728257061,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.1399726858181312,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.13100430231070642,
            "majority_failure_difference": -0.04477611940298509,
            "matched_fraction": 0.8375,
            "matched_items": 67,
            "mean_loss_difference": -0.016583747927031656,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "a_shared-04": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.8139659860907514,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.7005881248534772,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.35583511832889275,
            "majority_failure_difference": 0.05714285714285716,
            "matched_fraction": 0.875,
            "matched_items": 70,
            "mean_loss_difference": 0.03809523809523818,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.1214956483782883,
            "all_failure_difference": 0.015151515151515152,
            "dependence_inflation_ratio": 1.235299419173931,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.21131338537022745,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": 0.02020202020202011,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "a_shared-05": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9942498436410452,
            "all_failure_difference": -0.015151515151515152,
            "dependence_inflation_ratio": 0.8877521989211685,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.11906263020208795,
            "majority_failure_difference": -0.06060606060606055,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": 0.0016835016835016203,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.915929997853673,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.8478046600189155,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.1651050234770133,
            "majority_failure_difference": -0.030303030303030276,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": 0.011784511784511786,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "a_shared-06": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.221880429285237,
            "all_failure_difference": 0.014285714285714285,
            "dependence_inflation_ratio": 1.174586417088176,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.16091610017782967,
            "majority_failure_difference": 0.057142857142857106,
            "matched_fraction": 0.875,
            "matched_items": 70,
            "mean_loss_difference": 0.015873015873015706,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9866426339616783,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9138523926170492,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.09008621661249284,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.8875,
            "matched_items": 71,
            "mean_loss_difference": -0.0046948356807512415,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "a_shared-07": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9164455867442528,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9167943515345967,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.08687209410674451,
            "majority_failure_difference": 0.015873015873015928,
            "matched_fraction": 0.7875,
            "matched_items": 63,
            "mean_loss_difference": 0.022927689594356204,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9693196372994451,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.0034504908006903,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.003444551515685033,
            "majority_failure_difference": 0.05797101449275355,
            "matched_fraction": 0.8625,
            "matched_items": 69,
            "mean_loss_difference": 0.00644122383252832,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "a_shared-08": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0155688471262212,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.0459842694541617,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.04495832676689609,
            "majority_failure_difference": 0.04477611940298509,
            "matched_fraction": 0.8375,
            "matched_items": 67,
            "mean_loss_difference": 0.016583747927031545,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9042232174226651,
            "all_failure_difference": -0.029850746268656716,
            "dependence_inflation_ratio": 0.9222230268154843,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.08096819008352259,
            "majority_failure_difference": 0.029850746268656747,
            "matched_fraction": 0.8375,
            "matched_items": 67,
            "mean_loss_difference": -0.008291873963515772,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "a_shared-09": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0708804874505145,
            "all_failure_difference": -0.015151515151515152,
            "dependence_inflation_ratio": 1.1447678064085804,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.13520182727888863,
            "majority_failure_difference": -0.030303030303030276,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": -0.026936026936026924,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9800836473813466,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9218579167801193,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.08136417057603165,
            "majority_failure_difference": -0.02941176470588236,
            "matched_fraction": 0.85,
            "matched_items": 68,
            "mean_loss_difference": 0.0032679738562091387,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        }
      },
      "E_shared_variant": {
        "e_shared-00": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9942498436410452,
            "all_failure_difference": -0.015151515151515152,
            "dependence_inflation_ratio": 0.8877521989211685,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.11906263020208795,
            "majority_failure_difference": -0.06060606060606055,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": 0.0016835016835016203,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.915929997853673,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.8478046600189155,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.1651050234770133,
            "majority_failure_difference": -0.030303030303030276,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": 0.011784511784511786,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "e_shared-01": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.221880429285237,
            "all_failure_difference": 0.014285714285714285,
            "dependence_inflation_ratio": 1.174586417088176,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.16091610017782967,
            "majority_failure_difference": 0.057142857142857106,
            "matched_fraction": 0.875,
            "matched_items": 70,
            "mean_loss_difference": 0.015873015873015706,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9866426339616783,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9138523926170492,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.09008621661249284,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.8875,
            "matched_items": 71,
            "mean_loss_difference": -0.0046948356807512415,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "e_shared-02": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9164455867442528,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9167943515345967,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.08687209410674451,
            "majority_failure_difference": 0.015873015873015928,
            "matched_fraction": 0.7875,
            "matched_items": 63,
            "mean_loss_difference": 0.022927689594356204,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9693196372994451,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.0034504908006903,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.003444551515685033,
            "majority_failure_difference": 0.05797101449275355,
            "matched_fraction": 0.8625,
            "matched_items": 69,
            "mean_loss_difference": 0.00644122383252832,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "e_shared-03": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0155688471262212,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.0459842694541617,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.04495832676689609,
            "majority_failure_difference": 0.04477611940298509,
            "matched_fraction": 0.8375,
            "matched_items": 67,
            "mean_loss_difference": 0.016583747927031545,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9042232174226651,
            "all_failure_difference": -0.029850746268656716,
            "dependence_inflation_ratio": 0.9222230268154843,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.08096819008352259,
            "majority_failure_difference": 0.029850746268656747,
            "matched_fraction": 0.8375,
            "matched_items": 67,
            "mean_loss_difference": -0.008291873963515772,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "e_shared-04": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0708804874505145,
            "all_failure_difference": -0.015151515151515152,
            "dependence_inflation_ratio": 1.1447678064085804,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.13520182727888863,
            "majority_failure_difference": -0.030303030303030276,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": -0.026936026936026924,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9800836473813466,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9218579167801193,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.08136417057603165,
            "majority_failure_difference": -0.02941176470588236,
            "matched_fraction": 0.85,
            "matched_items": 68,
            "mean_loss_difference": 0.0032679738562091387,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "e_shared-05": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.2191820347416777,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.4399405856167675,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.3646018527483366,
            "majority_failure_difference": -0.08695652173913049,
            "matched_fraction": 0.8625,
            "matched_items": 69,
            "mean_loss_difference": -0.037037037037036924,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.913411215982746,
            "all_failure_difference": -0.028169014084507043,
            "dependence_inflation_ratio": 0.8364500470564334,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.17858847694753113,
            "majority_failure_difference": 0.07042253521126762,
            "matched_fraction": 0.8875,
            "matched_items": 71,
            "mean_loss_difference": 0.014084507042253391,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "e_shared-06": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0315663482135862,
            "all_failure_difference": 0.047619047619047616,
            "dependence_inflation_ratio": 1.1303354797969178,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.12251447343871816,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.7875,
            "matched_items": 63,
            "mean_loss_difference": 0.012345679012345845,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9734143384163312,
            "all_failure_difference": -0.014705882352941176,
            "dependence_inflation_ratio": 0.956005370342136,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.04499174843381942,
            "majority_failure_difference": 0.07352941176470584,
            "matched_fraction": 0.85,
            "matched_items": 68,
            "mean_loss_difference": 0.02941176470588258,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "e_shared-07": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.4126976484117477,
            "all_failure_difference": 0.015384615384615385,
            "dependence_inflation_ratio": 1.858101431337508,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.619555230579459,
            "majority_failure_difference": 0.015384615384615385,
            "matched_fraction": 0.8125,
            "matched_items": 65,
            "mean_loss_difference": -0.017094017094017144,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.1199234403538731,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.2052680823764963,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.18670201719996685,
            "majority_failure_difference": -0.02898550724637683,
            "matched_fraction": 0.8625,
            "matched_items": 69,
            "mean_loss_difference": -0.02576489533011256,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "e_shared-08": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0865173421150336,
            "all_failure_difference": 0.015151515151515152,
            "dependence_inflation_ratio": 1.1163891255730451,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.11009948201367657,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": -0.008417508417508324,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.073323728257061,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.1399726858181312,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.13100430231070642,
            "majority_failure_difference": -0.04477611940298509,
            "matched_fraction": 0.8375,
            "matched_items": 67,
            "mean_loss_difference": -0.016583747927031656,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "e_shared-09": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.8139659860907514,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.7005881248534772,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.35583511832889275,
            "majority_failure_difference": 0.05714285714285716,
            "matched_fraction": 0.875,
            "matched_items": 70,
            "mean_loss_difference": 0.03809523809523818,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.1214956483782883,
            "all_failure_difference": 0.015151515151515152,
            "dependence_inflation_ratio": 1.235299419173931,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.21131338537022745,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.825,
            "matched_items": 66,
            "mean_loss_difference": 0.02020202020202011,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        }
      },
      "domain": {
        "arithmetic": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9744128568874008,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9827849902097596,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.017364910928580522,
            "majority_failure_difference": 0.017391304347826098,
            "matched_fraction": 0.71875,
            "matched_items": 115,
            "mean_loss_difference": 0.01545893719806768,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9806028137765634,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.969379972431682,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.031098615541155693,
            "majority_failure_difference": 0.03149606299212593,
            "matched_fraction": 0.79375,
            "matched_items": 127,
            "mean_loss_difference": 0.006999125109361293,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        },
        "code": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.1131968676267892,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.1407034714956352,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.13164515234999938,
            "majority_failure_difference": 0.08450704225352113,
            "matched_fraction": 0.8875,
            "matched_items": 142,
            "mean_loss_difference": 0.007042253521126696,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9416318919923704,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.908643243841306,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.09580273286323235,
            "majority_failure_difference": -0.02836879432624112,
            "matched_fraction": 0.88125,
            "matched_items": 141,
            "mean_loss_difference": -0.010244286840031536,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        },
        "constraint": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.1935265328381204,
            "all_failure_difference": 0.03125,
            "dependence_inflation_ratio": 1.3604674671030026,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.3078283664988087,
            "majority_failure_difference": 0.015625,
            "matched_fraction": 0.8,
            "matched_items": 128,
            "mean_loss_difference": 0.0008680555555555802,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.0351652390299302,
            "all_failure_difference": -0.007751937984496124,
            "dependence_inflation_ratio": 1.0656384934059797,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.06357414384334058,
            "majority_failure_difference": 0.06201550387596899,
            "matched_fraction": 0.80625,
            "matched_items": 129,
            "mean_loss_difference": 0.011197243755383446,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        },
        "extraction": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9649779046925492,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.984954451644897,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.015159880864117823,
            "majority_failure_difference": -0.016260162601626015,
            "matched_fraction": 0.76875,
            "matched_items": 123,
            "mean_loss_difference": 0.004516711833784992,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9611268939652227,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9243954625532734,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.07861530912568325,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.7875,
            "matched_items": 126,
            "mean_loss_difference": 0.007054673721340388,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        },
        "tool_call": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0586523196745383,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.098810158444929,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.09422792023167534,
            "majority_failure_difference": -0.08280254777070062,
            "matched_fraction": 0.98125,
            "matched_items": 157,
            "mean_loss_difference": -0.014154281670205271,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9436523923040985,
            "all_failure_difference": -0.018867924528301886,
            "dependence_inflation_ratio": 0.9715589069927781,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.028853376909948672,
            "majority_failure_difference": -0.0062893081761006275,
            "matched_fraction": 0.99375,
            "matched_items": 159,
            "mean_loss_difference": 0.0013976240391335715,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        }
      },
      "predicted_difficulty_band": {
        "tier-1": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.1004693244832684,
            "all_failure_difference": -0.007407407407407408,
            "dependence_inflation_ratio": 1.1382512907171907,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.12949312918088432,
            "majority_failure_difference": -0.07407407407407407,
            "matched_fraction": 0.84375,
            "matched_items": 135,
            "mean_loss_difference": -0.018106995884773602,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9146121754411178,
            "all_failure_difference": -0.014598540145985403,
            "dependence_inflation_ratio": 0.835305958575266,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.17995720379035487,
            "majority_failure_difference": 0.02189781021897813,
            "matched_fraction": 0.85625,
            "matched_items": 137,
            "mean_loss_difference": 0.012976480129764933,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        },
        "tier-2": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.099275927487106,
            "all_failure_difference": 0.03007518796992481,
            "dependence_inflation_ratio": 1.1281423304418934,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.12057232458567108,
            "majority_failure_difference": 0.030075187969924866,
            "matched_fraction": 0.83125,
            "matched_items": 133,
            "mean_loss_difference": 0.014202172096908883,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9821949033517849,
            "all_failure_difference": -0.007194244604316547,
            "dependence_inflation_ratio": 0.9305083690818214,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.07202420876314684,
            "majority_failure_difference": 0.03597122302158273,
            "matched_fraction": 0.86875,
            "matched_items": 139,
            "mean_loss_difference": 0.011990407673860948,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        },
        "tier-3": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0913611318986005,
            "all_failure_difference": 0.0078125,
            "dependence_inflation_ratio": 1.1990609222943938,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.18153868567564177,
            "majority_failure_difference": 0.015625,
            "matched_fraction": 0.8,
            "matched_items": 128,
            "mean_loss_difference": 0.0026041666666665186,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.0317928472202602,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.0723292658492172,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.06983316646615256,
            "majority_failure_difference": 0.014492753623188415,
            "matched_fraction": 0.8625,
            "matched_items": 138,
            "mean_loss_difference": -0.009661835748792424,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        },
        "tier-4": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0484896646535022,
            "all_failure_difference": 0.007518796992481203,
            "dependence_inflation_ratio": 1.0774986417880492,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.07464228247343929,
            "majority_failure_difference": 0.022556390977443608,
            "matched_fraction": 0.83125,
            "matched_items": 133,
            "mean_loss_difference": 0.004177109440267279,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9791011608159185,
            "all_failure_difference": -0.014925373134328356,
            "dependence_inflation_ratio": 1.0121483036802161,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.012075105267137403,
            "majority_failure_difference": -0.007462686567164201,
            "matched_fraction": 0.8375,
            "matched_items": 134,
            "mean_loss_difference": -0.01243781094527352,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        },
        "tier-5": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9653531789174336,
            "all_failure_difference": -0.00735294117647059,
            "dependence_inflation_ratio": 0.9360265058417545,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.06611148469849518,
            "majority_failure_difference": 0.014705882352941235,
            "matched_fraction": 0.85,
            "matched_items": 136,
            "mean_loss_difference": 0.0065359477124183885,
            "planned_items": 160,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.0429633807159673,
            "all_failure_difference": 0.007462686567164176,
            "dependence_inflation_ratio": 1.0620610508690527,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.06021140786147365,
            "majority_failure_difference": -0.01492537313432829,
            "matched_fraction": 0.8375,
            "matched_items": 134,
            "mean_loss_difference": 0.011608623548922004,
            "planned_items": 160,
            "shared_condition": "A_shared"
          }
        }
      },
      "template_family_id": {
        "cs-v7-arithmetic-lower-tier-batch-reconciliation": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.1117665423642216,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.182284634080475,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.16744869716919525,
            "majority_failure_difference": 0.06896551724137934,
            "matched_fraction": 0.725,
            "matched_items": 29,
            "mean_loss_difference": 0.02298850574712652,
            "planned_items": 40,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.0543121331342413,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.1264417052772677,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.11906373101788732,
            "majority_failure_difference": 0.08823529411764702,
            "matched_fraction": 0.85,
            "matched_items": 34,
            "mean_loss_difference": 0.01633986928104575,
            "planned_items": 40,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-arithmetic-semantic-crate-reconciliation": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9863939238321437,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.0884697647287573,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.08477282430048244,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.875,
            "matched_items": 35,
            "mean_loss_difference": 0.015873015873015817,
            "planned_items": 40,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.929029191196508,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.819796145883102,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.1986995722121654,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.85,
            "matched_items": 34,
            "mean_loss_difference": -0.006535947712418277,
            "planned_items": 40,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-arithmetic-semantic-shift-loss-recovery": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.8390684563487422,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.7050882449543189,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.3494323138550277,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.6375,
            "matched_items": 51,
            "mean_loss_difference": 0.010893246187363648,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9645873064788374,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9467821822406182,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.0546862204456452,
            "majority_failure_difference": 0.016949152542372836,
            "matched_fraction": 0.7375,
            "matched_items": 59,
            "mean_loss_difference": 0.00941619585687381,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-code-fresh-layered-config-precedence": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.0778964320248612,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.035975866203041,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.035343848395144706,
            "majority_failure_difference": 0.04477611940298509,
            "matched_fraction": 0.8375,
            "matched_items": 67,
            "mean_loss_difference": 0.008291873963515495,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.922515845415673,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.8851776518220942,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.12196691759694706,
            "majority_failure_difference": -0.029850746268656803,
            "matched_fraction": 0.8375,
            "matched_items": 67,
            "mean_loss_difference": -0.01492537313432818,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-code-fresh-versioned-transition-policy": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.3236588084310064,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.754974624807613,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.5624543980142188,
            "majority_failure_difference": 0.12,
            "matched_fraction": 0.9375,
            "matched_items": 75,
            "mean_loss_difference": 0.005925925925926001,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9823221989970311,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9465194340405005,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.05496377601264514,
            "majority_failure_difference": -0.02702702702702703,
            "matched_fraction": 0.925,
            "matched_items": 74,
            "mean_loss_difference": -0.006006006006006037,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-constraint-fresh-multi-adjustment-band-policy": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.1313418538961841,
            "all_failure_difference": 0.03076923076923077,
            "dependence_inflation_ratio": 1.1419470125799447,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.1327347113657191,
            "majority_failure_difference": -0.01538461538461533,
            "matched_fraction": 0.8125,
            "matched_items": 65,
            "mean_loss_difference": -0.010256410256410331,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.0095695960312836,
            "all_failure_difference": -0.015384615384615385,
            "dependence_inflation_ratio": 0.9879825889477672,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.012090203912050497,
            "majority_failure_difference": 0.0461538461538461,
            "matched_fraction": 0.8125,
            "matched_items": 65,
            "mean_loss_difference": 0.017094017094017033,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-constraint-fresh-ordered-policy-exceptions": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.3227951666115594,
            "all_failure_difference": 0.031746031746031744,
            "dependence_inflation_ratio": 1.7526424494375523,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.5611246201825096,
            "majority_failure_difference": 0.04761904761904756,
            "matched_fraction": 0.7875,
            "matched_items": 63,
            "mean_loss_difference": 0.012345679012345734,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.0400659648881279,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.1047084648408836,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.09958146743655814,
            "majority_failure_difference": 0.078125,
            "matched_fraction": 0.8,
            "matched_items": 64,
            "mean_loss_difference": 0.00520833333333337,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-extraction-fresh-authoritative-audit-reduction": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.060568240237133,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.2153065432962151,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.19499634397969964,
            "majority_failure_difference": -0.032786885245901634,
            "matched_fraction": 0.7625,
            "matched_items": 61,
            "mean_loss_difference": -0.0018214936247722413,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9698891926133639,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.9179527260806407,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.08560938633182652,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.7375,
            "matched_items": 59,
            "mean_loss_difference": -0.003766478342749513,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-extraction-fresh-cross-record-final-join": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9899897379644396,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.152644501002761,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.14205886850881028,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.775,
            "matched_items": 62,
            "mean_loss_difference": 0.010752688172042946,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 1.0062112398448113,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.0506247798246882,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": 0.04938501561468955,
            "majority_failure_difference": 0.0,
            "matched_fraction": 0.8375,
            "matched_items": 67,
            "mean_loss_difference": 0.016583747927031545,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-tool_call-conditional-incident-response-dag": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 0.9223894630856704,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.8607176870137102,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": -0.14998871796922894,
            "majority_failure_difference": -0.08860759493670889,
            "matched_fraction": 0.9875,
            "matched_items": 79,
            "mean_loss_difference": -0.01406469760900153,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.8794743143506991,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 0.8691601315076091,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.14022789968265528,
            "majority_failure_difference": 0.0,
            "matched_fraction": 1.0,
            "matched_items": 80,
            "mean_loss_difference": 0.006944444444444475,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        },
        "cs-v7-tool_call-multi-region-release-dag": {
          "evidence_corruption": {
            "absolute_volatility_ratio": 1.1774473989074907,
            "all_failure_difference": 0.0,
            "dependence_inflation_ratio": 1.2965961705004834,
            "independent_condition": "E_independent",
            "log_dependence_inflation_ratio": 0.2597425002610166,
            "majority_failure_difference": -0.07692307692307687,
            "matched_fraction": 0.975,
            "matched_items": 78,
            "mean_loss_difference": -0.014245014245014231,
            "planned_items": 80,
            "shared_condition": "E_shared"
          },
          "instruction_anchor": {
            "absolute_volatility_ratio": 0.9535424136493694,
            "all_failure_difference": -0.0379746835443038,
            "dependence_inflation_ratio": 0.9837347469100033,
            "independent_condition": "A_independent",
            "log_dependence_inflation_ratio": -0.01639898441773919,
            "majority_failure_difference": -0.012658227848101333,
            "matched_fraction": 0.9875,
            "matched_items": 79,
            "mean_loss_difference": -0.0042194092827004814,
            "planned_items": 80,
            "shared_condition": "A_shared"
          }
        }
      }
    },
    "inference": "exploratory point estimates; no multiplicity-adjusted claims",
    "leave_one_template_family_out": {
      "cs-v7-arithmetic-lower-tier-batch-reconciliation": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.060593616985864,
          "all_failure_difference": 0.0062893081761006275,
          "dependence_inflation_ratio": 1.097159798994405,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.09272483977681971,
          "majority_failure_difference": -0.0015723270440251569,
          "matched_fraction": 0.8368421052631579,
          "matched_items": 636,
          "mean_loss_difference": 0.0008735150244583156,
          "planned_items": 760,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.9887281125291806,
          "all_failure_difference": -0.006172839506172839,
          "dependence_inflation_ratio": 0.9789586467645018,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.021265877625180202,
          "majority_failure_difference": 0.006172839506172867,
          "matched_fraction": 0.8526315789473684,
          "matched_items": 648,
          "mean_loss_difference": 0.0022290809327846217,
          "planned_items": 760,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-arithmetic-semantic-crate-reconciliation": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.061798724484891,
          "all_failure_difference": 0.006349206349206351,
          "dependence_inflation_ratio": 1.0997082350564604,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.09504490376398392,
          "majority_failure_difference": 0.0015873015873015817,
          "matched_fraction": 0.8289473684210527,
          "matched_items": 630,
          "mean_loss_difference": 0.0010582010582010914,
          "planned_items": 760,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.9919538915176777,
          "all_failure_difference": -0.006172839506172839,
          "dependence_inflation_ratio": 0.9817147149787535,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.018454527113373954,
          "majority_failure_difference": 0.010802469135802462,
          "matched_fraction": 0.8526315789473684,
          "matched_items": 648,
          "mean_loss_difference": 0.003429355281207247,
          "planned_items": 760,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-arithmetic-semantic-shift-loss-recovery": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.0720445206538018,
          "all_failure_difference": 0.0065146579804560255,
          "dependence_inflation_ratio": 1.1174639391918992,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.11106177780541716,
          "majority_failure_difference": 0.0016286644951140072,
          "matched_fraction": 0.8527777777777777,
          "matched_items": 614,
          "mean_loss_difference": 0.0010857763300760048,
          "planned_items": 720,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.9824893381024717,
          "all_failure_difference": -0.006420545746388442,
          "dependence_inflation_ratio": 0.9696467537096188,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.03082344524461094,
          "majority_failure_difference": 0.009630818619582648,
          "matched_fraction": 0.8652777777777778,
          "matched_items": 623,
          "mean_loss_difference": 0.002318530408418118,
          "planned_items": 720,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-code-fresh-layered-config-precedence": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.0578081908324204,
          "all_failure_difference": 0.006688963210702342,
          "dependence_inflation_ratio": 1.1024789687548493,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.09756125219759522,
          "majority_failure_difference": -0.0033444816053511683,
          "matched_fraction": 0.8305555555555556,
          "matched_items": 598,
          "mean_loss_difference": 0.0011148272017837968,
          "planned_items": 720,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.9972486686785976,
          "all_failure_difference": -0.006504065040650405,
          "dependence_inflation_ratio": 0.9897616294436247,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.010291143185032773,
          "majority_failure_difference": 0.014634146341463372,
          "matched_fraction": 0.8541666666666666,
          "matched_items": 615,
          "mean_loss_difference": 0.00487804878048792,
          "planned_items": 720,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-code-fresh-versioned-transition-policy": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.0570907111018513,
          "all_failure_difference": 0.0067796610169491515,
          "dependence_inflation_ratio": 1.0893965967431152,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.08562396198891613,
          "majority_failure_difference": -0.013559322033898313,
          "matched_fraction": 0.8194444444444444,
          "matched_items": 590,
          "mean_loss_difference": 0.0013182674199623268,
          "planned_items": 720,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.985782879847381,
          "all_failure_difference": -0.006578947368421054,
          "dependence_inflation_ratio": 0.9766627502020251,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.02361387567923018,
          "majority_failure_difference": 0.014802631578947345,
          "matched_fraction": 0.8444444444444444,
          "matched_items": 608,
          "mean_loss_difference": 0.004020467836257424,
          "planned_items": 720,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-constraint-fresh-multi-adjustment-band-policy": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.066393324850808,
          "all_failure_difference": 0.003333333333333334,
          "dependence_inflation_ratio": 1.1021522630208036,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.09726487089651559,
          "majority_failure_difference": 0.003333333333333355,
          "matched_fraction": 0.8333333333333334,
          "matched_items": 600,
          "mean_loss_difference": 0.003148148148148122,
          "planned_items": 720,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.9831996000390938,
          "all_failure_difference": -0.004862236628849272,
          "dependence_inflation_ratio": 0.9711584131566775,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.029265679654789874,
          "majority_failure_difference": 0.0064829821717990455,
          "matched_fraction": 0.8569444444444444,
          "matched_items": 617,
          "mean_loss_difference": 0.00144066270484422,
          "planned_items": 720,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-constraint-fresh-ordered-policy-exceptions": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.0378307434915655,
          "all_failure_difference": 0.0033222591362126255,
          "dependence_inflation_ratio": 1.0532105600518058,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.05184317522458362,
          "majority_failure_difference": -0.0033222591362126463,
          "matched_fraction": 0.8361111111111111,
          "matched_items": 602,
          "mean_loss_difference": 0.000738279808047082,
          "planned_items": 720,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.985057293621414,
          "all_failure_difference": -0.006472491909385111,
          "dependence_inflation_ratio": 0.9716714292440747,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.02873756742578818,
          "majority_failure_difference": 0.003236245954692518,
          "matched_fraction": 0.8583333333333333,
          "matched_items": 618,
          "mean_loss_difference": 0.0026968716289105243,
          "planned_items": 720,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-extraction-fresh-authoritative-audit-reduction": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.0607730909210171,
          "all_failure_difference": 0.006622516556291392,
          "dependence_inflation_ratio": 1.0989699086282025,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.09437329436255242,
          "majority_failure_difference": 0.004966887417218513,
          "matched_fraction": 0.8388888888888889,
          "matched_items": 604,
          "mean_loss_difference": 0.002207505518763697,
          "planned_items": 720,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.9855835624438627,
          "all_failure_difference": -0.006420545746388444,
          "dependence_inflation_ratio": 0.9744333729892409,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.025899132829311177,
          "majority_failure_difference": 0.011235955056179803,
          "matched_fraction": 0.8652777777777778,
          "matched_items": 623,
          "mean_loss_difference": 0.0035669698591047627,
          "planned_items": 720,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-extraction-fresh-cross-record-final-join": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.075739519512652,
          "all_failure_difference": 0.006633499170812603,
          "dependence_inflation_ratio": 1.1213217626755696,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.1145081347290417,
          "majority_failure_difference": 0.0016583747927031434,
          "matched_fraction": 0.8375,
          "matched_items": 603,
          "mean_loss_difference": 0.0009213193292795241,
          "planned_items": 720,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.9962884501219049,
          "all_failure_difference": -0.006504065040650405,
          "dependence_inflation_ratio": 0.9866955742178458,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.01339372256756173,
          "majority_failure_difference": 0.011382113821138296,
          "matched_fraction": 0.8541666666666666,
          "matched_items": 615,
          "mean_loss_difference": 0.0014453477868112685,
          "planned_items": 720,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-tool_call-conditional-incident-response-dag": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.0608749478633517,
          "all_failure_difference": 0.0068259385665529,
          "dependence_inflation_ratio": 1.0858095300829669,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.08232581946576946,
          "majority_failure_difference": 0.013651877133105894,
          "matched_fraction": 0.8138888888888889,
          "matched_items": 586,
          "mean_loss_difference": 0.00398179749715577,
          "planned_items": 720,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.9922486508740522,
          "all_failure_difference": -0.006644518272425248,
          "dependence_inflation_ratio": 0.9854086134140404,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.014698887877352938,
          "majority_failure_difference": 0.011627906976744207,
          "matched_fraction": 0.8361111111111111,
          "matched_items": 602,
          "mean_loss_difference": 0.0023994093761534607,
          "planned_items": 720,
          "shared_condition": "A_shared"
        }
      },
      "cs-v7-tool_call-multi-region-release-dag": {
        "evidence_corruption": {
          "absolute_volatility_ratio": 1.0556864502770813,
          "all_failure_difference": 0.006814310051107324,
          "dependence_inflation_ratio": 1.0914861104776996,
          "independent_condition": "E_independent",
          "log_dependence_inflation_ratio": 0.08754017175758723,
          "majority_failure_difference": 0.011925042589437773,
          "matched_fraction": 0.8152777777777778,
          "matched_items": 587,
          "mean_loss_difference": 0.003975014196479165,
          "planned_items": 720,
          "shared_condition": "E_shared"
        },
        "instruction_anchor": {
          "absolute_volatility_ratio": 0.9935552410254501,
          "all_failure_difference": -0.001658374792703152,
          "dependence_inflation_ratio": 0.9799141280547047,
          "independent_condition": "A_independent",
          "log_dependence_inflation_ratio": -0.020290335590739256,
          "majority_failure_difference": 0.013266998341625202,
          "matched_fraction": 0.8375,
          "matched_items": 603,
          "mean_loss_difference": 0.003869541182974112,
          "planned_items": 720,
          "shared_condition": "A_shared"
        }
      }
    },
    "schema_version": "cmfc.common-shock-subgroups.v1"
  },
  "verdict_outcomes": {
    "model_failure": 22173,
    "nan": 522,
    "pass": 20640
  }
}
