{
  "runDate": "2026-09-05",
  "method": "Five prespecified model candidates; development-only five-fold selection; locked 20% holdout (seed 804); separate full-data leakage, ablation and grouped diagnostics. Fixed >0.5 threshold. No leaderboard queries or hidden outcomes used. Holdout rows are freshly withheld for this run, not historically untouched by the original notebook.",
  "development": {
    "Logistic regression": {
      "accuracy": 0.8314606741573034,
      "foldScores": [
        0.7902097902097902,
        0.8461538461538461,
        0.8380281690140845,
        0.8098591549295775,
        0.8732394366197183
      ],
      "correct": 592,
      "n": 712,
      "brier": 0.12403591236522209
    },
    "Random forest": {
      "accuracy": 0.848314606741573,
      "foldScores": [
        0.8111888111888111,
        0.8601398601398601,
        0.8661971830985915,
        0.8450704225352113,
        0.8591549295774648
      ],
      "correct": 604,
      "n": 712,
      "brier": 0.11632959396053467
    },
    "XGBoost": {
      "accuracy": 0.8497191011235955,
      "foldScores": [
        0.8111888111888111,
        0.8671328671328671,
        0.8732394366197183,
        0.8169014084507042,
        0.8802816901408451
      ],
      "correct": 605,
      "n": 712,
      "brier": 0.1118353721982637
    },
    "Extra trees": {
      "accuracy": 0.8497191011235955,
      "foldScores": [
        0.8111888111888111,
        0.8671328671328671,
        0.8661971830985915,
        0.8309859154929577,
        0.8732394366197183
      ],
      "correct": 605,
      "n": 712,
      "brier": 0.1150568092689115
    },
    "Simple ensemble": {
      "accuracy": 0.8370786516853933,
      "foldScores": [
        0.7972027972027972,
        0.8531468531468531,
        0.8591549295774648,
        0.823943661971831,
        0.852112676056338
      ],
      "correct": 596,
      "n": 712,
      "brier": 0.11445178852199106
    },
    "Sex-only baseline": {
      "accuracy": 0.8033707865168539,
      "foldScores": [
        0.7622377622377622,
        0.8181818181818182,
        0.8380281690140845,
        0.7746478873239436,
        0.823943661971831
      ],
      "correct": 572,
      "n": 712,
      "brier": 0.19662921348314608
    }
  },
  "winner": "XGBoost",
  "holdout": {
    "n": 179,
    "correct": 143,
    "accuracy": 0.7988826815642458,
    "brier": 0.15220314264297485,
    "confusionMatrix": [
      [
        97,
        13
      ],
      [
        23,
        46
      ]
    ],
    "baselineAccuracy": 0.7206703910614525
  },
  "safeCV": {
    "Logistic regression": {
      "accuracy": 0.8316498316498316,
      "foldScores": [
        0.8435754189944135,
        0.8314606741573034,
        0.797752808988764,
        0.848314606741573,
        0.8370786516853933
      ],
      "correct": 741,
      "n": 891,
      "brier": 0.12552826324272598
    },
    "Random forest": {
      "accuracy": 0.8451178451178452,
      "foldScores": [
        0.8435754189944135,
        0.8370786516853933,
        0.8370786516853933,
        0.8539325842696629,
        0.8539325842696629
      ],
      "correct": 753,
      "n": 891,
      "brier": 0.11860536666847146
    },
    "XGBoost": {
      "accuracy": 0.8473625140291807,
      "foldScores": [
        0.8491620111731844,
        0.8370786516853933,
        0.8370786516853933,
        0.8539325842696629,
        0.8595505617977528
      ],
      "correct": 755,
      "n": 891,
      "brier": 0.11350082204695376
    },
    "Extra trees": {
      "accuracy": 0.8473625140291807,
      "foldScores": [
        0.8435754189944135,
        0.8370786516853933,
        0.8370786516853933,
        0.8595505617977528,
        0.8595505617977528
      ],
      "correct": 755,
      "n": 891,
      "brier": 0.11968386125251396
    },
    "Simple ensemble": {
      "accuracy": 0.8383838383838383,
      "foldScores": [
        0.8268156424581006,
        0.8202247191011236,
        0.8426966292134831,
        0.8595505617977528,
        0.8426966292134831
      ],
      "correct": 747,
      "n": 891,
      "brier": 0.11637485906324002
    },
    "Sex-only baseline": {
      "accuracy": 0.7867564534231201,
      "foldScores": [
        0.7877094972067039,
        0.7752808988764045,
        0.7865168539325843,
        0.7640449438202247,
        0.8202247191011236
      ],
      "correct": 701,
      "n": 891,
      "brier": 0.2132435465768799
    }
  },
  "leakyCV": {
    "Logistic regression": {
      "accuracy": 0.8338945005611672,
      "foldScores": [
        0.8491620111731844,
        0.8314606741573034,
        0.8089887640449438,
        0.848314606741573,
        0.8314606741573034
      ],
      "correct": 743,
      "n": 891,
      "brier": 0.12542156874540014
    },
    "Random forest": {
      "accuracy": 0.8451178451178452,
      "foldScores": [
        0.8435754189944135,
        0.8314606741573034,
        0.8426966292134831,
        0.8595505617977528,
        0.848314606741573
      ],
      "correct": 753,
      "n": 891,
      "brier": 0.11726462372396707
    },
    "XGBoost": {
      "accuracy": 0.8552188552188552,
      "foldScores": [
        0.8491620111731844,
        0.848314606741573,
        0.8426966292134831,
        0.8707865168539326,
        0.8651685393258427
      ],
      "correct": 762,
      "n": 891,
      "brier": 0.11202570210770522
    },
    "Extra trees": {
      "accuracy": 0.8462401795735129,
      "foldScores": [
        0.8491620111731844,
        0.8258426966292135,
        0.8370786516853933,
        0.8651685393258427,
        0.8539325842696629
      ],
      "correct": 754,
      "n": 891,
      "brier": 0.11807194466680902
    },
    "Simple ensemble": {
      "accuracy": 0.8451178451178452,
      "foldScores": [
        0.8379888268156425,
        0.8314606741573034,
        0.8426966292134831,
        0.8707865168539326,
        0.8426966292134831
      ],
      "correct": 753,
      "n": 891,
      "brier": 0.11533918936918676
    },
    "Sex-only baseline": {
      "accuracy": 0.7867564534231201,
      "foldScores": [
        0.7877094972067039,
        0.7752808988764045,
        0.7865168539325843,
        0.7640449438202247,
        0.8202247191011236
      ],
      "correct": 701,
      "n": 891,
      "brier": 0.2132435465768799
    }
  },
  "withoutFamilyCV": {
    "Logistic regression": {
      "accuracy": 0.8125701459034792,
      "foldScores": [
        0.8044692737430168,
        0.8146067415730337,
        0.7921348314606742,
        0.8089887640449438,
        0.8426966292134831
      ],
      "correct": 724,
      "n": 891,
      "brier": 0.1339527016353076
    },
    "Random forest": {
      "accuracy": 0.8305274971941639,
      "foldScores": [
        0.8268156424581006,
        0.8202247191011236,
        0.8314606741573034,
        0.8314606741573034,
        0.8426966292134831
      ],
      "correct": 740,
      "n": 891,
      "brier": 0.12455581938753248
    },
    "XGBoost": {
      "accuracy": 0.8462401795735129,
      "foldScores": [
        0.8491620111731844,
        0.8595505617977528,
        0.8314606741573034,
        0.848314606741573,
        0.8426966292134831
      ],
      "correct": 754,
      "n": 891,
      "brier": 0.11859141536298344
    },
    "Extra trees": {
      "accuracy": 0.8305274971941639,
      "foldScores": [
        0.8324022346368715,
        0.8089887640449438,
        0.8258426966292135,
        0.8426966292134831,
        0.8426966292134831
      ],
      "correct": 740,
      "n": 891,
      "brier": 0.12611856751567435
    },
    "Simple ensemble": {
      "accuracy": 0.8361391694725028,
      "foldScores": [
        0.8324022346368715,
        0.8258426966292135,
        0.8258426966292135,
        0.8426966292134831,
        0.8539325842696629
      ],
      "correct": 745,
      "n": 891,
      "brier": 0.12289317110455623
    },
    "Sex-only baseline": {
      "accuracy": 0.7867564534231201,
      "foldScores": [
        0.7877094972067039,
        0.7752808988764045,
        0.7865168539325843,
        0.7640449438202247,
        0.8202247191011236
      ],
      "correct": 701,
      "n": 891,
      "brier": 0.2132435465768799
    }
  },
  "groupedCV": {
    "Logistic regression": {
      "accuracy": 0.8024691358024691,
      "foldScores": [
        0.8324022346368715,
        0.8314606741573034,
        0.8033707865168539,
        0.7808988764044944,
        0.7640449438202247
      ],
      "correct": 715,
      "n": 891,
      "brier": 0.13731279569806612
    },
    "Random forest": {
      "accuracy": 0.8103254769921436,
      "foldScores": [
        0.8268156424581006,
        0.8146067415730337,
        0.8258426966292135,
        0.8033707865168539,
        0.7808988764044944
      ],
      "correct": 722,
      "n": 891,
      "brier": 0.1364159199435859
    },
    "XGBoost": {
      "accuracy": 0.8035914702581369,
      "foldScores": [
        0.8044692737430168,
        0.8089887640449438,
        0.8258426966292135,
        0.797752808988764,
        0.7808988764044944
      ],
      "correct": 716,
      "n": 891,
      "brier": 0.14005146017756265
    },
    "Extra trees": {
      "accuracy": 0.8035914702581369,
      "foldScores": [
        0.8268156424581006,
        0.8033707865168539,
        0.8146067415730337,
        0.797752808988764,
        0.7752808988764045
      ],
      "correct": 716,
      "n": 891,
      "brier": 0.13945385297541313
    },
    "Simple ensemble": {
      "accuracy": 0.8148148148148148,
      "foldScores": [
        0.8435754189944135,
        0.8314606741573034,
        0.8146067415730337,
        0.797752808988764,
        0.7865168539325843
      ],
      "correct": 726,
      "n": 891,
      "brier": 0.134928147336856
    },
    "Sex-only baseline": {
      "accuracy": 0.7867564534231201,
      "foldScores": [
        0.7988826815642458,
        0.7921348314606742,
        0.8146067415730337,
        0.7696629213483146,
        0.7584269662921348
      ],
      "correct": 701,
      "n": 891,
      "brier": 0.2132435465768799
    }
  },
  "stats": {
    "all23Correlation": 0.07015226333794626,
    "all23PearsonP": 0.7504329486475416,
    "excludingFirstCorrelation": -0.49110636715837713,
    "excludingFirstPearsonP": 0.020288586930695303,
    "wilson95": [
      0.7707043487241715,
      0.845640743841379
    ],
    "onePredictionPercentagePoints": 0.23923444976076555,
    "currentCorrect": 339,
    "currentErrors": 79
  },
  "submission": {
    "file": "contender-submission.csv",
    "rows": 418,
    "survivors": 150,
    "publicScore": null,
    "status": "Prepared locally; not submitted to Kaggle"
  },
  "versions": {
    "python": "3.12.14",
    "sklearn": "1.9.0",
    "xgboost": "3.4.1",
    "numpy": "2.5.2",
    "pandas": "3.0.1"
  },
  "data": {
    "source": "https://github.com/agconti/kaggle-titanic/tree/master/data",
    "boundary": "Public mirror of competition train/test files; official direct download requires authentication. Dimensions, ID ranges, training outcome total and missing-value counts verified; official byte parity unverified.",
    "trainSHA256": "4a437fde05fe5264e1701a7387ac6fb75393772ba38bb2c9c566405af5af4bd7",
    "testSHA256": "5c78320a80dc159b35c408363be46a5d93f078183d539db75ee42a796d6a4e95",
    "missingTrain": {
      "PassengerId": 0,
      "Survived": 0,
      "Pclass": 0,
      "Name": 0,
      "Sex": 0,
      "Age": 177,
      "SibSp": 0,
      "Parch": 0,
      "Ticket": 0,
      "Fare": 0,
      "Cabin": 687,
      "Embarked": 2
    },
    "missingTest": {
      "PassengerId": 0,
      "Pclass": 0,
      "Name": 0,
      "Sex": 0,
      "Age": 86,
      "SibSp": 0,
      "Parch": 0,
      "Ticket": 0,
      "Fare": 1,
      "Cabin": 327,
      "Embarked": 0
    }
  },
  "limitations": [
    "One holdout split; no claim of a statistical maximum or superiority on Kaggle.",
    "Family/ticket grouping is a conservative proxy and can group unrelated surnames.",
    "Leakage experiment isolates reference-label contamination in a new controlled implementation; it does not reproduce the original dependency environment.",
    "Repeated diagnostic folds overlap; fold scores are not independent replications."
  ]
}
