{
  "protocol": "experimental-support-coin-v1",
  "forecastProtocol": "experimental-support-v1",
  "inputSha256": "f1f53a813f414f6a2c30e6b75eaa96e770cf703ad93e32204b69bee8ee561039",
  "createdAt": "2026-10-22T16:34:16.758Z",
  "hypotheses": 250,
  "questions": 50,
  "heads": 117,
  "tails": 133,
  "metrics": [
    {
      "endpoint": "strict",
      "policy": "advice",
      "validHypotheses": 250,
      "validQuestions": 50,
      "unknownProbabilities": 250,
      "selected": 117,
      "referenceSupported": 2,
      "selectedWithoutReferenceSupport": 115,
      "notSelected": 133,
      "allKnownPositives": 4,
      "knownPositivesInValidQuestions": 4,
      "missedKnownPositives": 2,
      "recoveredFractionOfAllKnown": 0.5,
      "counts": {
        "yes_true": 2,
        "yes_false": 115,
        "no_true": 2,
        "no_false": 131
      },
      "rates": {
        "ppv": 0.017094017094017096,
        "npv": 0.9849624060150376,
        "sens": 0.5,
        "spec": 0.532520325203252,
        "youden": 0.03252032520325199
      },
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "tasks": 50,
        "replicates": 10000,
        "intervals": {
          "ppv": {
            "lo": 0,
            "hi": 0.04424778761061947,
            "valid_replicates": 10000
          },
          "npv": {
            "lo": 0.9621212121212122,
            "hi": 1,
            "valid_replicates": 10000
          },
          "sens": {
            "lo": 0,
            "hi": 1,
            "valid_replicates": 9824
          },
          "spec": {
            "lo": 0.476,
            "hi": 0.5870445344129555,
            "valid_replicates": 10000
          }
        },
        "degenerate": [],
        "interpretation": "Task-cluster descriptive intervals for article correspondence. Unmatched hypotheses remain biologically unlabelled; these are not intervals for future experimental success."
      }
    },
    {
      "endpoint": "core",
      "policy": "advice",
      "validHypotheses": 250,
      "validQuestions": 50,
      "unknownProbabilities": 250,
      "selected": 117,
      "referenceSupported": 17,
      "selectedWithoutReferenceSupport": 100,
      "notSelected": 133,
      "allKnownPositives": 36,
      "knownPositivesInValidQuestions": 36,
      "missedKnownPositives": 19,
      "recoveredFractionOfAllKnown": 0.4722222222222222,
      "counts": {
        "yes_true": 17,
        "yes_false": 100,
        "no_true": 19,
        "no_false": 114
      },
      "rates": {
        "ppv": 0.1452991452991453,
        "npv": 0.8571428571428571,
        "sens": 0.4722222222222222,
        "spec": 0.5327102803738317,
        "youden": 0.004932502596053823
      },
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "tasks": 50,
        "replicates": 10000,
        "intervals": {
          "ppv": {
            "lo": 0.09016393442622951,
            "hi": 0.20661869478483888,
            "valid_replicates": 10000
          },
          "npv": {
            "lo": 0.7839895522388061,
            "hi": 0.9241405433646812,
            "valid_replicates": 10000
          },
          "sens": {
            "lo": 0.34283834586466166,
            "hi": 0.6333333333333333,
            "valid_replicates": 10000
          },
          "spec": {
            "lo": 0.46918694301458086,
            "hi": 0.5954545454545455,
            "valid_replicates": 10000
          }
        },
        "degenerate": [],
        "interpretation": "Task-cluster descriptive intervals for article correspondence. Unmatched hypotheses remain biologically unlabelled; these are not intervals for future experimental success."
      }
    }
  ],
  "tosses": [
    {
      "task": "R42581103",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42581103",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42581103",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42581103",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42581103",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42601407",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42601407",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42601407",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42601407",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42601407",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42605193",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42605193",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42605193",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42605193",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42605193",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42608398",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42608398",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42608398",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42608398",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42608398",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42627902",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42627902",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42627902",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42627902",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42627902",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42630044",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42630044",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42630044",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42630044",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42630044",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42639736",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42639736",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42639736",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42639736",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42639736",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42642518",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42642518",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42642518",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42642518",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42642518",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42642521",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42642521",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42642521",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42642521",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42642521",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42647119",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42647119",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42647119",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42647119",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42647119",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42647635",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42647635",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42647635",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42647635",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42647635",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42663085",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42663085",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42663085",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42663085",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42663085",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42664348",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42664348",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42664348",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42664348",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42664348",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42665751",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42665751",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42665751",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42665751",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42665751",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42689491",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42689491",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42689491",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42689491",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42689491",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42696595",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42696595",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42696595",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42696595",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42696595",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42698274",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42698274",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42698274",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42698274",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42698274",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42717261",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42717261",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42717261",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42717261",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42717261",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42723546",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42723546",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42723546",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42723546",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42723546",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42725837",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42725837",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42725837",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42725837",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42725837",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42736449",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42736449",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42736449",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42736449",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42736449",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42744804",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42744804",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42744804",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42744804",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42744804",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42749700",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42749700",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42749700",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42749700",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42749700",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42758848",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42758848",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42758848",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42758848",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42758848",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42759510",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42759510",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42759510",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42759510",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42759510",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42775695",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42775695",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42775695",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42775695",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42775695",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42775698",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42775698",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42775698",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42775698",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42775698",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42786310",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42786310",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42786310",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42786310",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42786310",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42805954",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42805954",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42805954",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42805954",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42805954",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42810392",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42810392",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42810392",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42810392",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42810392",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42811497",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42811497",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42811497",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42811497",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42811497",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42821663",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42821663",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42821663",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42821663",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42821663",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42822442",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42822442",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42822442",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42822442",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42822442",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42826184",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42826184",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42826184",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42826184",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42826184",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42839094",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42839094",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42839094",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42839094",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42839094",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42841227",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42841227",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42841227",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42841227",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42841227",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42842666",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42842666",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42842666",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42842666",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42842666",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42844664",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42844664",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42844664",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42844664",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42844664",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42853938",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42853938",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42853938",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42853938",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42853938",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42858183",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42858183",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42858183",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42858183",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42858183",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42806945",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42806945",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42806945",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42806945",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42806945",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42723144",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42723144",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42723144",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42723144",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42723144",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42808804",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42808804",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42808804",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42808804",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42808804",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42585340",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42585340",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42585340",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42585340",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42585340",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42669067",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42669067",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42669067",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42669067",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42669067",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42811065",
      "id": "H1",
      "recommend": false
    },
    {
      "task": "R42811065",
      "id": "H2",
      "recommend": false
    },
    {
      "task": "R42811065",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42811065",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42811065",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42841610",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42841610",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42841610",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42841610",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42841610",
      "id": "H5",
      "recommend": false
    },
    {
      "task": "R42806003",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42806003",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42806003",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42806003",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42806003",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42806572",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42806572",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42806572",
      "id": "H3",
      "recommend": true
    },
    {
      "task": "R42806572",
      "id": "H4",
      "recommend": true
    },
    {
      "task": "R42806572",
      "id": "H5",
      "recommend": true
    },
    {
      "task": "R42748238",
      "id": "H1",
      "recommend": true
    },
    {
      "task": "R42748238",
      "id": "H2",
      "recommend": true
    },
    {
      "task": "R42748238",
      "id": "H3",
      "recommend": false
    },
    {
      "task": "R42748238",
      "id": "H4",
      "recommend": false
    },
    {
      "task": "R42748238",
      "id": "H5",
      "recommend": true
    }
  ],
  "comparisons": [
    {
      "modelId": "claude-sonnet-5-5",
      "endpoint": "strict",
      "policy": "advice",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 4,
          "yes_false": 124,
          "no_true": 0,
          "no_false": 122
        },
        "rates": {
          "ppv": 0.03125,
          "npv": 1,
          "sens": 1,
          "spec": 0.4959349593495935,
          "youden": 0.4959349593495934
        },
        "selected": 128,
        "referenceSupported": 4
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": 0.014155982905982904,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.0021681070429592324,
          "hi": 0.038461538461538464,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-sonnet-5-5",
      "endpoint": "strict",
      "policy": "p50",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 17,
          "no_true": 4,
          "no_false": 229
        },
        "rates": {
          "ppv": 0,
          "npv": 0.9828326180257511,
          "sens": 0,
          "spec": 0.9308943089430894,
          "youden": -0.06910569105691056
        },
        "selected": 17,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": -0.017094017094017096,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.04424778761061947,
          "hi": 0,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-sonnet-5-5",
      "endpoint": "strict",
      "policy": "p70",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 2,
          "no_true": 4,
          "no_false": 244
        },
        "rates": {
          "ppv": 0,
          "npv": 0.9838709677419355,
          "sens": 0,
          "spec": 0.991869918699187,
          "youden": -0.008130081300813052
        },
        "selected": 2,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": -0.017094017094017096,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.04386935258500232,
          "hi": 0,
          "valid_replicates": 8760
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-sonnet-5-5",
      "endpoint": "strict",
      "policy": "p90",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 4,
          "no_false": 246
        },
        "rates": {
          "ppv": null,
          "npv": 0.984,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "claude-sonnet-5-5",
      "endpoint": "core",
      "policy": "advice",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 26,
          "yes_false": 102,
          "no_true": 10,
          "no_false": 112
        },
        "rates": {
          "ppv": 0.203125,
          "npv": 0.9180327868852459,
          "sens": 0.7222222222222222,
          "spec": 0.5233644859813084,
          "youden": 0.24558670820353057
        },
        "selected": 128,
        "referenceSupported": 26
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.05782585470085469,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.001255378931865507,
          "hi": 0.12114426312467882,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-sonnet-5-5",
      "endpoint": "core",
      "policy": "p50",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 7,
          "yes_false": 10,
          "no_true": 29,
          "no_false": 204
        },
        "rates": {
          "ppv": 0.4117647058823529,
          "npv": 0.8755364806866953,
          "sens": 0.19444444444444445,
          "spec": 0.9532710280373832,
          "youden": 0.1477154724818277
        },
        "selected": 17,
        "referenceSupported": 7
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.2664655605832076,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": 0.02484280054807578,
          "hi": 0.5133865015196498,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-sonnet-5-5",
      "endpoint": "core",
      "policy": "p70",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 1,
          "yes_false": 1,
          "no_true": 35,
          "no_false": 213
        },
        "rates": {
          "ppv": 0.5,
          "npv": 0.8588709677419355,
          "sens": 0.027777777777777776,
          "spec": 0.9953271028037384,
          "youden": 0.023104880581516163
        },
        "selected": 2,
        "referenceSupported": 1
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.35470085470085466,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.19047619047619047,
          "hi": 0.9,
          "valid_replicates": 8760
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-sonnet-5-5",
      "endpoint": "core",
      "policy": "p90",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 36,
          "no_false": 214
        },
        "rates": {
          "ppv": null,
          "npv": 0.856,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-5-5",
      "endpoint": "strict",
      "policy": "advice",
      "hypotheses": 215,
      "questions": 43,
      "model": {
        "counts": {
          "yes_true": 4,
          "yes_false": 99,
          "no_true": 0,
          "no_false": 112
        },
        "rates": {
          "ppv": 0.038834951456310676,
          "npv": 1,
          "sens": 1,
          "spec": 0.5308056872037915,
          "youden": 0.5308056872037916
        },
        "selected": 103,
        "referenceSupported": 4
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 98,
          "no_true": 2,
          "no_false": 113
        },
        "rates": {
          "ppv": 0.02,
          "npv": 0.9826086956521739,
          "sens": 0.5,
          "spec": 0.5355450236966824,
          "youden": 0.03554502369668233
        },
        "selected": 100,
        "referenceSupported": 2
      },
      "ppvDifference": 0.018834951456310676,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.0015053763440860228,
          "hi": 0.049696975522795,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-5-5",
      "endpoint": "strict",
      "policy": "p50",
      "hypotheses": 215,
      "questions": 43,
      "model": {
        "counts": {
          "yes_true": 2,
          "yes_false": 59,
          "no_true": 2,
          "no_false": 152
        },
        "rates": {
          "ppv": 0.03278688524590164,
          "npv": 0.987012987012987,
          "sens": 0.5,
          "spec": 0.7203791469194313,
          "youden": 0.22037914691943117
        },
        "selected": 61,
        "referenceSupported": 2
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 98,
          "no_true": 2,
          "no_false": 113
        },
        "rates": {
          "ppv": 0.02,
          "npv": 0.9826086956521739,
          "sens": 0.5,
          "spec": 0.5355450236966824,
          "youden": 0.03554502369668233
        },
        "selected": 100,
        "referenceSupported": 2
      },
      "ppvDifference": 0.01278688524590164,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.021052631578947368,
          "hi": 0.05882644102417082,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-5-5",
      "endpoint": "strict",
      "policy": "p70",
      "hypotheses": 215,
      "questions": 43,
      "model": {
        "counts": {
          "yes_true": 2,
          "yes_false": 11,
          "no_true": 2,
          "no_false": 200
        },
        "rates": {
          "ppv": 0.15384615384615385,
          "npv": 0.9900990099009901,
          "sens": 0.5,
          "spec": 0.9478672985781991,
          "youden": 0.44786729857819907
        },
        "selected": 13,
        "referenceSupported": 2
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 98,
          "no_true": 2,
          "no_false": 113
        },
        "rates": {
          "ppv": 0.02,
          "npv": 0.9826086956521739,
          "sens": 0.5,
          "spec": 0.5355450236966824,
          "youden": 0.03554502369668233
        },
        "selected": 100,
        "referenceSupported": 2
      },
      "ppvDifference": 0.13384615384615386,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.019417475728155338,
          "hi": 0.3485893546607831,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-5-5",
      "endpoint": "strict",
      "policy": "p90",
      "hypotheses": 215,
      "questions": 43,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 4,
          "no_false": 211
        },
        "rates": {
          "ppv": null,
          "npv": 0.9813953488372092,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 98,
          "no_true": 2,
          "no_false": 113
        },
        "rates": {
          "ppv": 0.02,
          "npv": 0.9826086956521739,
          "sens": 0.5,
          "spec": 0.5355450236966824,
          "youden": 0.03554502369668233
        },
        "selected": 100,
        "referenceSupported": 2
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-5-5",
      "endpoint": "core",
      "policy": "advice",
      "hypotheses": 215,
      "questions": 43,
      "model": {
        "counts": {
          "yes_true": 22,
          "yes_false": 81,
          "no_true": 7,
          "no_false": 105
        },
        "rates": {
          "ppv": 0.21359223300970873,
          "npv": 0.9375,
          "sens": 0.7586206896551724,
          "spec": 0.5645161290322581,
          "youden": 0.3231368186874306
        },
        "selected": 103,
        "referenceSupported": 22
      },
      "coin": {
        "counts": {
          "yes_true": 14,
          "yes_false": 86,
          "no_true": 15,
          "no_false": 100
        },
        "rates": {
          "ppv": 0.14,
          "npv": 0.8695652173913043,
          "sens": 0.4827586206896552,
          "spec": 0.5376344086021505,
          "youden": 0.020393029291805753
        },
        "selected": 100,
        "referenceSupported": 14
      },
      "ppvDifference": 0.07359223300970871,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.0011707043266008233,
          "hi": 0.15158432663654714,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-5-5",
      "endpoint": "core",
      "policy": "p50",
      "hypotheses": 215,
      "questions": 43,
      "model": {
        "counts": {
          "yes_true": 16,
          "yes_false": 45,
          "no_true": 13,
          "no_false": 141
        },
        "rates": {
          "ppv": 0.26229508196721313,
          "npv": 0.9155844155844156,
          "sens": 0.5517241379310345,
          "spec": 0.7580645161290323,
          "youden": 0.3097886540600667
        },
        "selected": 61,
        "referenceSupported": 16
      },
      "coin": {
        "counts": {
          "yes_true": 14,
          "yes_false": 86,
          "no_true": 15,
          "no_false": 100
        },
        "rates": {
          "ppv": 0.14,
          "npv": 0.8695652173913043,
          "sens": 0.4827586206896552,
          "spec": 0.5376344086021505,
          "youden": 0.020393029291805753
        },
        "selected": 100,
        "referenceSupported": 14
      },
      "ppvDifference": 0.12229508196721312,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": 0.029807126811349678,
          "hi": 0.2211545950155763,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-5-5",
      "endpoint": "core",
      "policy": "p70",
      "hypotheses": 215,
      "questions": 43,
      "model": {
        "counts": {
          "yes_true": 5,
          "yes_false": 8,
          "no_true": 24,
          "no_false": 178
        },
        "rates": {
          "ppv": 0.38461538461538464,
          "npv": 0.8811881188118812,
          "sens": 0.1724137931034483,
          "spec": 0.956989247311828,
          "youden": 0.1294030404152764
        },
        "selected": 13,
        "referenceSupported": 5
      },
      "coin": {
        "counts": {
          "yes_true": 14,
          "yes_false": 86,
          "no_true": 15,
          "no_false": 100
        },
        "rates": {
          "ppv": 0.14,
          "npv": 0.8695652173913043,
          "sens": 0.4827586206896552,
          "spec": 0.5376344086021505,
          "youden": 0.020393029291805753
        },
        "selected": 100,
        "referenceSupported": 14
      },
      "ppvDifference": 0.24461538461538462,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.013427715695157523,
          "hi": 0.5162713524099661,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-5-5",
      "endpoint": "core",
      "policy": "p90",
      "hypotheses": 215,
      "questions": 43,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 29,
          "no_false": 186
        },
        "rates": {
          "ppv": null,
          "npv": 0.8651162790697674,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 14,
          "yes_false": 86,
          "no_true": 15,
          "no_false": 100
        },
        "rates": {
          "ppv": 0.14,
          "npv": 0.8695652173913043,
          "sens": 0.4827586206896552,
          "spec": 0.5376344086021505,
          "youden": 0.020393029291805753
        },
        "selected": 100,
        "referenceSupported": 14
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "claude-haiku-5-5",
      "endpoint": "strict",
      "policy": "advice",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 3,
          "yes_false": 106,
          "no_true": 1,
          "no_false": 140
        },
        "rates": {
          "ppv": 0.027522935779816515,
          "npv": 0.9929078014184397,
          "sens": 0.75,
          "spec": 0.5691056910569106,
          "youden": 0.31910569105691056
        },
        "selected": 109,
        "referenceSupported": 3
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": 0.01042891868579942,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.000866039597657247,
          "hi": 0.030288025889967634,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-haiku-5-5",
      "endpoint": "strict",
      "policy": "p50",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 12,
          "no_true": 4,
          "no_false": 234
        },
        "rates": {
          "ppv": 0,
          "npv": 0.9831932773109243,
          "sens": 0,
          "spec": 0.9512195121951219,
          "youden": -0.04878048780487809
        },
        "selected": 12,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": -0.017094017094017096,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.04424778761061947,
          "hi": 0,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-haiku-5-5",
      "endpoint": "strict",
      "policy": "p70",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 4,
          "no_false": 246
        },
        "rates": {
          "ppv": null,
          "npv": 0.984,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "claude-haiku-5-5",
      "endpoint": "strict",
      "policy": "p90",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 4,
          "no_false": 246
        },
        "rates": {
          "ppv": null,
          "npv": 0.984,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "claude-haiku-5-5",
      "endpoint": "core",
      "policy": "advice",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 20,
          "yes_false": 89,
          "no_true": 16,
          "no_false": 125
        },
        "rates": {
          "ppv": 0.1834862385321101,
          "npv": 0.8865248226950354,
          "sens": 0.5555555555555556,
          "spec": 0.5841121495327103,
          "youden": 0.13966770508826576
        },
        "selected": 109,
        "referenceSupported": 20
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.03818709323296479,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.04379300458715596,
          "hi": 0.12480526224684094,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-haiku-5-5",
      "endpoint": "core",
      "policy": "p50",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 5,
          "yes_false": 7,
          "no_true": 31,
          "no_false": 207
        },
        "rates": {
          "ppv": 0.4166666666666667,
          "npv": 0.8697478991596639,
          "sens": 0.1388888888888889,
          "spec": 0.9672897196261683,
          "youden": 0.10617860851505712
        },
        "selected": 12,
        "referenceSupported": 5
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.2713675213675214,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.031316672624012075,
          "hi": 0.5823403426791274,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-haiku-5-5",
      "endpoint": "core",
      "policy": "p70",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 36,
          "no_false": 214
        },
        "rates": {
          "ppv": null,
          "npv": 0.856,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "claude-haiku-5-5",
      "endpoint": "core",
      "policy": "p90",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 36,
          "no_false": 214
        },
        "rates": {
          "ppv": null,
          "npv": 0.856,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-4-8",
      "endpoint": "strict",
      "policy": "advice",
      "hypotheses": 240,
      "questions": 48,
      "model": {
        "counts": {
          "yes_true": 3,
          "yes_false": 131,
          "no_true": 1,
          "no_false": 105
        },
        "rates": {
          "ppv": 0.022388059701492536,
          "npv": 0.9905660377358491,
          "sens": 0.75,
          "spec": 0.4449152542372881,
          "youden": 0.19491525423728806
        },
        "selected": 134,
        "referenceSupported": 3
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 110,
          "no_true": 2,
          "no_false": 126
        },
        "rates": {
          "ppv": 0.017857142857142856,
          "npv": 0.984375,
          "sens": 0.5,
          "spec": 0.5338983050847458,
          "youden": 0.03389830508474567
        },
        "selected": 112,
        "referenceSupported": 2
      },
      "ppvDifference": 0.00453091684434968,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.022431077694235586,
          "hi": 0.03091398651614443,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-4-8",
      "endpoint": "strict",
      "policy": "p50",
      "hypotheses": 240,
      "questions": 48,
      "model": {
        "counts": {
          "yes_true": 1,
          "yes_false": 77,
          "no_true": 3,
          "no_false": 159
        },
        "rates": {
          "ppv": 0.01282051282051282,
          "npv": 0.9814814814814815,
          "sens": 0.25,
          "spec": 0.673728813559322,
          "youden": -0.07627118644067798
        },
        "selected": 78,
        "referenceSupported": 1
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 110,
          "no_true": 2,
          "no_false": 126
        },
        "rates": {
          "ppv": 0.017857142857142856,
          "npv": 0.984375,
          "sens": 0.5,
          "spec": 0.5338983050847458,
          "youden": 0.03389830508474567
        },
        "selected": 112,
        "referenceSupported": 2
      },
      "ppvDifference": -0.005036630036630036,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.02727272727272727,
          "hi": 0.010466988727858294,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-4-8",
      "endpoint": "strict",
      "policy": "p70",
      "hypotheses": 240,
      "questions": 48,
      "model": {
        "counts": {
          "yes_true": 1,
          "yes_false": 4,
          "no_true": 3,
          "no_false": 232
        },
        "rates": {
          "ppv": 0.2,
          "npv": 0.9872340425531915,
          "sens": 0.25,
          "spec": 0.9830508474576272,
          "youden": 0.23305084745762716
        },
        "selected": 5,
        "referenceSupported": 1
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 110,
          "no_true": 2,
          "no_false": 126
        },
        "rates": {
          "ppv": 0.017857142857142856,
          "npv": 0.984375,
          "sens": 0.5,
          "spec": 0.5338983050847458,
          "youden": 0.03389830508474567
        },
        "selected": 112,
        "referenceSupported": 2
      },
      "ppvDifference": 0.18214285714285716,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.02586206896551724,
          "hi": 0.6361239950525666,
          "valid_replicates": 9952
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-4-8",
      "endpoint": "strict",
      "policy": "p90",
      "hypotheses": 240,
      "questions": 48,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 4,
          "no_false": 236
        },
        "rates": {
          "ppv": null,
          "npv": 0.9833333333333333,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 110,
          "no_true": 2,
          "no_false": 126
        },
        "rates": {
          "ppv": 0.017857142857142856,
          "npv": 0.984375,
          "sens": 0.5,
          "spec": 0.5338983050847458,
          "youden": 0.03389830508474567
        },
        "selected": 112,
        "referenceSupported": 2
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-4-8",
      "endpoint": "core",
      "policy": "advice",
      "hypotheses": 240,
      "questions": 48,
      "model": {
        "counts": {
          "yes_true": 28,
          "yes_false": 106,
          "no_true": 6,
          "no_false": 100
        },
        "rates": {
          "ppv": 0.208955223880597,
          "npv": 0.9433962264150944,
          "sens": 0.8235294117647058,
          "spec": 0.4854368932038835,
          "youden": 0.30896630496858934
        },
        "selected": 134,
        "referenceSupported": 28
      },
      "coin": {
        "counts": {
          "yes_true": 16,
          "yes_false": 96,
          "no_true": 18,
          "no_false": 110
        },
        "rates": {
          "ppv": 0.14285714285714285,
          "npv": 0.859375,
          "sens": 0.47058823529411764,
          "spec": 0.5339805825242718,
          "youden": 0.004568817818389359
        },
        "selected": 112,
        "referenceSupported": 16
      },
      "ppvDifference": 0.06609808102345416,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.00045132808709044594,
          "hi": 0.1326132799864956,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-4-8",
      "endpoint": "core",
      "policy": "p50",
      "hypotheses": 240,
      "questions": 48,
      "model": {
        "counts": {
          "yes_true": 18,
          "yes_false": 60,
          "no_true": 16,
          "no_false": 146
        },
        "rates": {
          "ppv": 0.23076923076923078,
          "npv": 0.9012345679012346,
          "sens": 0.5294117647058824,
          "spec": 0.7087378640776699,
          "youden": 0.23814962878355228
        },
        "selected": 78,
        "referenceSupported": 18
      },
      "coin": {
        "counts": {
          "yes_true": 16,
          "yes_false": 96,
          "no_true": 18,
          "no_false": 110
        },
        "rates": {
          "ppv": 0.14285714285714285,
          "npv": 0.859375,
          "sens": 0.47058823529411764,
          "spec": 0.5339805825242718,
          "youden": 0.004568817818389359
        },
        "selected": 112,
        "referenceSupported": 16
      },
      "ppvDifference": 0.08791208791208793,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.00430750934420661,
          "hi": 0.18181954600021824,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-4-8",
      "endpoint": "core",
      "policy": "p70",
      "hypotheses": 240,
      "questions": 48,
      "model": {
        "counts": {
          "yes_true": 4,
          "yes_false": 1,
          "no_true": 30,
          "no_false": 205
        },
        "rates": {
          "ppv": 0.8,
          "npv": 0.8723404255319149,
          "sens": 0.11764705882352941,
          "spec": 0.9951456310679612,
          "youden": 0.1127926898914906
        },
        "selected": 5,
        "referenceSupported": 4
      },
      "coin": {
        "counts": {
          "yes_true": 16,
          "yes_false": 96,
          "no_true": 18,
          "no_false": 110
        },
        "rates": {
          "ppv": 0.14285714285714285,
          "npv": 0.859375,
          "sens": 0.47058823529411764,
          "spec": 0.5339805825242718,
          "youden": 0.004568817818389359
        },
        "selected": 112,
        "referenceSupported": 16
      },
      "ppvDifference": 0.6571428571428573,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": 0.17724438239595797,
          "hi": 0.905982905982906,
          "valid_replicates": 9952
        },
        "degenerate": false
      }
    },
    {
      "modelId": "claude-opus-4-8",
      "endpoint": "core",
      "policy": "p90",
      "hypotheses": 240,
      "questions": 48,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 34,
          "no_false": 206
        },
        "rates": {
          "ppv": null,
          "npv": 0.8583333333333333,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 16,
          "yes_false": 96,
          "no_true": 18,
          "no_false": 110
        },
        "rates": {
          "ppv": 0.14285714285714285,
          "npv": 0.859375,
          "sens": 0.47058823529411764,
          "spec": 0.5339805825242718,
          "youden": 0.004568817818389359
        },
        "selected": 112,
        "referenceSupported": 16
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "gpt-6-astra",
      "endpoint": "strict",
      "policy": "advice",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 3,
          "yes_false": 197,
          "no_true": 1,
          "no_false": 49
        },
        "rates": {
          "ppv": 0.015,
          "npv": 0.98,
          "sens": 0.75,
          "spec": 0.1991869918699187,
          "youden": -0.05081300813008127
        },
        "selected": 200,
        "referenceSupported": 3
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": -0.0020940170940170963,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.025510828438950986,
          "hi": 0.0196078431372549,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gpt-6-astra",
      "endpoint": "strict",
      "policy": "p50",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 3,
          "yes_false": 98,
          "no_true": 1,
          "no_false": 148
        },
        "rates": {
          "ppv": 0.0297029702970297,
          "npv": 0.9932885906040269,
          "sens": 0.75,
          "spec": 0.6016260162601627,
          "youden": 0.35162601626016254
        },
        "selected": 101,
        "referenceSupported": 3
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": 0.012608953203012606,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.01679234851835471,
          "hi": 0.0469603740923834,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gpt-6-astra",
      "endpoint": "strict",
      "policy": "p70",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 1,
          "yes_false": 18,
          "no_true": 3,
          "no_false": 228
        },
        "rates": {
          "ppv": 0.05263157894736842,
          "npv": 0.987012987012987,
          "sens": 0.25,
          "spec": 0.926829268292683,
          "youden": 0.17682926829268286
        },
        "selected": 19,
        "referenceSupported": 1
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": 0.03553756185335132,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.03636363636363636,
          "hi": 0.14989473684210525,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gpt-6-astra",
      "endpoint": "strict",
      "policy": "p90",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 4,
          "no_false": 246
        },
        "rates": {
          "ppv": null,
          "npv": 0.984,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "gpt-6-astra",
      "endpoint": "core",
      "policy": "advice",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 32,
          "yes_false": 168,
          "no_true": 4,
          "no_false": 46
        },
        "rates": {
          "ppv": 0.16,
          "npv": 0.92,
          "sens": 0.8888888888888888,
          "spec": 0.21495327102803738,
          "youden": 0.10384215991692614
        },
        "selected": 200,
        "referenceSupported": 32
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.014700854700854693,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.03789021422522496,
          "hi": 0.06573225331101824,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gpt-6-astra",
      "endpoint": "core",
      "policy": "p50",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 22,
          "yes_false": 79,
          "no_true": 14,
          "no_false": 135
        },
        "rates": {
          "ppv": 0.21782178217821782,
          "npv": 0.9060402684563759,
          "sens": 0.6111111111111112,
          "spec": 0.6308411214953271,
          "youden": 0.24195223260643828
        },
        "selected": 101,
        "referenceSupported": 22
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.07252263687907251,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.0021964968348825386,
          "hi": 0.15136966335914315,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gpt-6-astra",
      "endpoint": "core",
      "policy": "p70",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 8,
          "yes_false": 11,
          "no_true": 28,
          "no_false": 203
        },
        "rates": {
          "ppv": 0.42105263157894735,
          "npv": 0.8787878787878788,
          "sens": 0.2222222222222222,
          "spec": 0.9485981308411215,
          "youden": 0.17082035306334387
        },
        "selected": 19,
        "referenceSupported": 8
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.275753486279802,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.012196616451458642,
          "hi": 0.5612068872746838,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gpt-6-astra",
      "endpoint": "core",
      "policy": "p90",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 36,
          "no_false": 214
        },
        "rates": {
          "ppv": null,
          "npv": 0.856,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "gemini-4-argon-high",
      "endpoint": "strict",
      "policy": "advice",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 4,
          "yes_false": 148,
          "no_true": 0,
          "no_false": 98
        },
        "rates": {
          "ppv": 0.02631578947368421,
          "npv": 1,
          "sens": 1,
          "spec": 0.3983739837398374,
          "youden": 0.39837398373983746
        },
        "selected": 152,
        "referenceSupported": 4
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": 0.009221772379667113,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.006705064595447246,
          "hi": 0.03134023512382655,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gemini-4-argon-high",
      "endpoint": "strict",
      "policy": "p50",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 3,
          "yes_false": 131,
          "no_true": 1,
          "no_false": 115
        },
        "rates": {
          "ppv": 0.022388059701492536,
          "npv": 0.9913793103448276,
          "sens": 0.75,
          "spec": 0.46747967479674796,
          "youden": 0.217479674796748
        },
        "selected": 134,
        "referenceSupported": 3
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": 0.00529404260747544,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.02000561389651981,
          "hi": 0.0329265264147617,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gemini-4-argon-high",
      "endpoint": "strict",
      "policy": "p70",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 3,
          "yes_false": 37,
          "no_true": 1,
          "no_false": 209
        },
        "rates": {
          "ppv": 0.075,
          "npv": 0.9952380952380953,
          "sens": 0.75,
          "spec": 0.8495934959349594,
          "youden": 0.5995934959349594
        },
        "selected": 40,
        "referenceSupported": 3
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": 0.0579059829059829,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.008547008547008548,
          "hi": 0.14740238317031512,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gemini-4-argon-high",
      "endpoint": "strict",
      "policy": "p90",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 4,
          "no_false": 246
        },
        "rates": {
          "ppv": null,
          "npv": 0.984,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 115,
          "no_true": 2,
          "no_false": 131
        },
        "rates": {
          "ppv": 0.017094017094017096,
          "npv": 0.9849624060150376,
          "sens": 0.5,
          "spec": 0.532520325203252,
          "youden": 0.03252032520325199
        },
        "selected": 117,
        "referenceSupported": 2
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "gemini-4-argon-high",
      "endpoint": "core",
      "policy": "advice",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 32,
          "yes_false": 120,
          "no_true": 4,
          "no_false": 94
        },
        "rates": {
          "ppv": 0.21052631578947367,
          "npv": 0.9591836734693877,
          "sens": 0.8888888888888888,
          "spec": 0.4392523364485981,
          "youden": 0.3281412253374869
        },
        "selected": 152,
        "referenceSupported": 32
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.06522717049032836,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": 0.0010312813313839166,
          "hi": 0.13219833259393626,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gemini-4-argon-high",
      "endpoint": "core",
      "policy": "p50",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 31,
          "yes_false": 103,
          "no_true": 5,
          "no_false": 111
        },
        "rates": {
          "ppv": 0.23134328358208955,
          "npv": 0.9568965517241379,
          "sens": 0.8611111111111112,
          "spec": 0.5186915887850467,
          "youden": 0.3798026998961579
        },
        "selected": 134,
        "referenceSupported": 31
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.08604413828294424,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": 0.01443516627295785,
          "hi": 0.1639715264203931,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gemini-4-argon-high",
      "endpoint": "core",
      "policy": "p70",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 14,
          "yes_false": 26,
          "no_true": 22,
          "no_false": 188
        },
        "rates": {
          "ppv": 0.35,
          "npv": 0.8952380952380953,
          "sens": 0.3888888888888889,
          "spec": 0.8785046728971962,
          "youden": 0.2673935617860852
        },
        "selected": 40,
        "referenceSupported": 14
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": 0.20470085470085467,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": 0.05207417978614855,
          "hi": 0.3699460144927534,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "gemini-4-argon-high",
      "endpoint": "core",
      "policy": "p90",
      "hypotheses": 250,
      "questions": 50,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 36,
          "no_false": 214
        },
        "rates": {
          "ppv": null,
          "npv": 0.856,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 17,
          "yes_false": 100,
          "no_true": 19,
          "no_false": 114
        },
        "rates": {
          "ppv": 0.1452991452991453,
          "npv": 0.8571428571428571,
          "sens": 0.4722222222222222,
          "spec": 0.5327102803738317,
          "youden": 0.004932502596053823
        },
        "selected": 117,
        "referenceSupported": 17
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "grok-4.5",
      "endpoint": "strict",
      "policy": "advice",
      "hypotheses": 225,
      "questions": 45,
      "model": {
        "counts": {
          "yes_true": 3,
          "yes_false": 134,
          "no_true": 0,
          "no_false": 88
        },
        "rates": {
          "ppv": 0.021897810218978103,
          "npv": 1,
          "sens": 1,
          "spec": 0.3963963963963964,
          "youden": 0.39639639639639634
        },
        "selected": 137,
        "referenceSupported": 3
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 107,
          "no_true": 1,
          "no_false": 115
        },
        "rates": {
          "ppv": 0.01834862385321101,
          "npv": 0.9913793103448276,
          "sens": 0.6666666666666666,
          "spec": 0.5180180180180181,
          "youden": 0.1846846846846848
        },
        "selected": 109,
        "referenceSupported": 2
      },
      "ppvDifference": 0.003549186365767093,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.009137503685684797,
          "hi": 0.02054794520547945,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "grok-4.5",
      "endpoint": "strict",
      "policy": "p50",
      "hypotheses": 225,
      "questions": 45,
      "model": {
        "counts": {
          "yes_true": 2,
          "yes_false": 45,
          "no_true": 1,
          "no_false": 177
        },
        "rates": {
          "ppv": 0.0425531914893617,
          "npv": 0.9943820224719101,
          "sens": 0.6666666666666666,
          "spec": 0.7972972972972973,
          "youden": 0.4639639639639639
        },
        "selected": 47,
        "referenceSupported": 2
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 107,
          "no_true": 1,
          "no_false": 115
        },
        "rates": {
          "ppv": 0.01834862385321101,
          "npv": 0.9913793103448276,
          "sens": 0.6666666666666666,
          "spec": 0.5180180180180181,
          "youden": 0.1846846846846848
        },
        "selected": 109,
        "referenceSupported": 2
      },
      "ppvDifference": 0.02420456763615069,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.01834862385321101,
          "hi": 0.08239019660894654,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "grok-4.5",
      "endpoint": "strict",
      "policy": "p70",
      "hypotheses": 225,
      "questions": 45,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 6,
          "no_true": 3,
          "no_false": 216
        },
        "rates": {
          "ppv": 0,
          "npv": 0.9863013698630136,
          "sens": 0,
          "spec": 0.972972972972973,
          "youden": -0.027027027027026973
        },
        "selected": 6,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 107,
          "no_true": 1,
          "no_false": 115
        },
        "rates": {
          "ppv": 0.01834862385321101,
          "npv": 0.9913793103448276,
          "sens": 0.6666666666666666,
          "spec": 0.5180180180180181,
          "youden": 0.1846846846846848
        },
        "selected": 109,
        "referenceSupported": 2
      },
      "ppvDifference": -0.01834862385321101,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.047619047619047616,
          "hi": 0,
          "valid_replicates": 9983
        },
        "degenerate": false
      }
    },
    {
      "modelId": "grok-4.5",
      "endpoint": "strict",
      "policy": "p90",
      "hypotheses": 225,
      "questions": 45,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 3,
          "no_false": 222
        },
        "rates": {
          "ppv": null,
          "npv": 0.9866666666666667,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 2,
          "yes_false": 107,
          "no_true": 1,
          "no_false": 115
        },
        "rates": {
          "ppv": 0.01834862385321101,
          "npv": 0.9913793103448276,
          "sens": 0.6666666666666666,
          "spec": 0.5180180180180181,
          "youden": 0.1846846846846848
        },
        "selected": 109,
        "referenceSupported": 2
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    },
    {
      "modelId": "grok-4.5",
      "endpoint": "core",
      "policy": "advice",
      "hypotheses": 225,
      "questions": 45,
      "model": {
        "counts": {
          "yes_true": 27,
          "yes_false": 110,
          "no_true": 4,
          "no_false": 84
        },
        "rates": {
          "ppv": 0.19708029197080293,
          "npv": 0.9545454545454546,
          "sens": 0.8709677419354839,
          "spec": 0.4329896907216495,
          "youden": 0.30395743265713326
        },
        "selected": 137,
        "referenceSupported": 27
      },
      "coin": {
        "counts": {
          "yes_true": 16,
          "yes_false": 93,
          "no_true": 15,
          "no_false": 101
        },
        "rates": {
          "ppv": 0.14678899082568808,
          "npv": 0.8706896551724138,
          "sens": 0.5161290322580645,
          "spec": 0.520618556701031,
          "youden": 0.036747588959095356
        },
        "selected": 109,
        "referenceSupported": 16
      },
      "ppvDifference": 0.05029130114511485,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.016435348125055687,
          "hi": 0.11723679561914854,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "grok-4.5",
      "endpoint": "core",
      "policy": "p50",
      "hypotheses": 225,
      "questions": 45,
      "model": {
        "counts": {
          "yes_true": 12,
          "yes_false": 35,
          "no_true": 19,
          "no_false": 159
        },
        "rates": {
          "ppv": 0.2553191489361702,
          "npv": 0.8932584269662921,
          "sens": 0.3870967741935484,
          "spec": 0.8195876288659794,
          "youden": 0.20668440305952762
        },
        "selected": 47,
        "referenceSupported": 12
      },
      "coin": {
        "counts": {
          "yes_true": 16,
          "yes_false": 93,
          "no_true": 15,
          "no_false": 101
        },
        "rates": {
          "ppv": 0.14678899082568808,
          "npv": 0.8706896551724138,
          "sens": 0.5161290322580645,
          "spec": 0.520618556701031,
          "youden": 0.036747588959095356
        },
        "selected": 109,
        "referenceSupported": 16
      },
      "ppvDifference": 0.10853015811048211,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.007118122354751816,
          "hi": 0.233115871645442,
          "valid_replicates": 10000
        },
        "degenerate": false
      }
    },
    {
      "modelId": "grok-4.5",
      "endpoint": "core",
      "policy": "p70",
      "hypotheses": 225,
      "questions": 45,
      "model": {
        "counts": {
          "yes_true": 2,
          "yes_false": 4,
          "no_true": 29,
          "no_false": 190
        },
        "rates": {
          "ppv": 0.3333333333333333,
          "npv": 0.867579908675799,
          "sens": 0.06451612903225806,
          "spec": 0.979381443298969,
          "youden": 0.04389757233122715
        },
        "selected": 6,
        "referenceSupported": 2
      },
      "coin": {
        "counts": {
          "yes_true": 16,
          "yes_false": 93,
          "no_true": 15,
          "no_false": 101
        },
        "rates": {
          "ppv": 0.14678899082568808,
          "npv": 0.8706896551724138,
          "sens": 0.5161290322580645,
          "spec": 0.520618556701031,
          "youden": 0.036747588959095356
        },
        "selected": 109,
        "referenceSupported": 16
      },
      "ppvDifference": 0.18654434250764523,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": {
          "lo": -0.16822429906542055,
          "hi": 0.6545454545454545,
          "valid_replicates": 9983
        },
        "degenerate": false
      }
    },
    {
      "modelId": "grok-4.5",
      "endpoint": "core",
      "policy": "p90",
      "hypotheses": 225,
      "questions": 45,
      "model": {
        "counts": {
          "yes_true": 0,
          "yes_false": 0,
          "no_true": 31,
          "no_false": 194
        },
        "rates": {
          "ppv": null,
          "npv": 0.8622222222222222,
          "sens": 0,
          "spec": 1,
          "youden": 0
        },
        "selected": 0,
        "referenceSupported": 0
      },
      "coin": {
        "counts": {
          "yes_true": 16,
          "yes_false": 93,
          "no_true": 15,
          "no_false": 101
        },
        "rates": {
          "ppv": 0.14678899082568808,
          "npv": 0.8706896551724138,
          "sens": 0.5161290322580645,
          "spec": 0.520618556701031,
          "youden": 0.036747588959095356
        },
        "selected": 109,
        "referenceSupported": 16
      },
      "ppvDifference": null,
      "uncertainty": {
        "method": "task-cluster percentile bootstrap",
        "replicates": 10000,
        "interval": null,
        "degenerate": false
      }
    }
  ],
  "design": {
    "protocol": "experimental-support-coin-v1",
    "createdAt": "2026-10-22T16:32:49.172774+00:00",
    "forecastProtocol": "experimental-support-v1",
    "inputSha256": "f1f53a813f414f6a2c30e6b75eaa96e770cf703ad93e32204b69bee8ee561039",
    "questions": 50,
    "hypotheses": 250,
    "drawAlgorithm": "Node crypto.randomBytes(32); first 250 bits, most-significant bit first in frozen source task/candidate order; final six bits unused",
    "heads": "recommend testing",
    "tails": "do not recommend testing",
    "redraws": 0,
    "timing": "Coin added after model results were inspected; draw generation uses only fixed hypothesis identities and random bytes, never article labels or model decisions.",
    "pairedScope": "exact valid hypothesis identities for each model, jointly resampled by original question",
    "replicates": 10000,
    "sources": {
      "summaries/forecast-research.json": "21a22b4bfedc50059563127add79c48dc40de86663cfb44c7d81934f05f9b966",
      "summaries/experimental-support-summary.json": "e04da325083c75557bba271e897203f496d4758f078c2ba58d83f9f8c85b0cb1",
      "https://eternalsearch.net/omega/experimental-support-20261022/claude-sonnet-5-5.json": "745c4ca36b272de4f24362b7d5c1858d32bd589f3851e7fddd5177d958bf80f9",
      "https://eternalsearch.net/omega/experimental-support-20261022/claude-opus-5-5.json": "40c3b17065ab1ab5a2f808ada20f8b1fa37057872958be07125527e9456f46c5",
      "https://eternalsearch.net/omega/experimental-support-20261022/claude-haiku-5-5.json": "6a3750d3ead03a0cae4fc6bd292967586cbdf72932579b9dd70b6b7f1d98b076",
      "https://eternalsearch.net/omega/experimental-support-20261022/claude-opus-4-8.json": "f918bb5282262b13ff3f3a9491204da73c8179c3883a9507abcb1a56cf20df5c",
      "https://eternalsearch.net/omega/experimental-support-20261022/gpt-6-astra.json": "c695752c0e1b011f1d39dc11925582234176ab4c755ee7b9ca68eaf9b07f0597",
      "https://eternalsearch.net/omega/experimental-support-20261022/gemini-4-argon-high.json": "7e8f33deff79491444c3e3150c8d2f851a107b955e1aff29e485d2bdc8087000",
      "https://eternalsearch.net/omega/experimental-support-20261022/grok-4.5.json": "5625fd8a3935d5ed2bf5b9f7cdf57ccc8e9a4a8a95d816e5b27836650f518d89"
    }
  },
  "draw": {
    "protocol": "experimental-support-coin-v1",
    "protocolSha256": "a06f3be759049d9bb8b12ed75f5d08203790d4684c349708f4b410725dc913ef",
    "createdAt": "2026-10-22T16:34:16.758Z",
    "randomBytesHex": "eb54c80472ab2521e51d3a51372c28f22380cb6a9a75e354b4e644621399fe68",
    "tosses": [
      {
        "task": "R42581103",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42581103",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42581103",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42581103",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42581103",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42601407",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42601407",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42601407",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42601407",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42601407",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42605193",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42605193",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42605193",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42605193",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42605193",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42608398",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42608398",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42608398",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42608398",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42608398",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42627902",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42627902",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42627902",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42627902",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42627902",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42630044",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42630044",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42630044",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42630044",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42630044",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42639736",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42639736",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42639736",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42639736",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42639736",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42642518",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42642518",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42642518",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42642518",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42642518",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42642521",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42642521",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42642521",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42642521",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42642521",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42647119",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42647119",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42647119",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42647119",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42647119",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42647635",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42647635",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42647635",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42647635",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42647635",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42663085",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42663085",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42663085",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42663085",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42663085",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42664348",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42664348",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42664348",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42664348",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42664348",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42665751",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42665751",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42665751",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42665751",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42665751",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42689491",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42689491",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42689491",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42689491",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42689491",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42696595",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42696595",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42696595",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42696595",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42696595",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42698274",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42698274",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42698274",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42698274",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42698274",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42717261",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42717261",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42717261",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42717261",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42717261",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42723546",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42723546",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42723546",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42723546",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42723546",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42725837",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42725837",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42725837",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42725837",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42725837",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42736449",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42736449",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42736449",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42736449",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42736449",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42744804",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42744804",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42744804",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42744804",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42744804",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42749700",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42749700",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42749700",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42749700",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42749700",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42758848",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42758848",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42758848",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42758848",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42758848",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42759510",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42759510",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42759510",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42759510",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42759510",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42775695",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42775695",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42775695",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42775695",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42775695",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42775698",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42775698",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42775698",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42775698",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42775698",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42786310",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42786310",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42786310",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42786310",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42786310",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42805954",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42805954",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42805954",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42805954",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42805954",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42810392",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42810392",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42810392",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42810392",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42810392",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42811497",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42811497",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42811497",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42811497",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42811497",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42821663",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42821663",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42821663",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42821663",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42821663",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42822442",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42822442",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42822442",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42822442",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42822442",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42826184",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42826184",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42826184",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42826184",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42826184",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42839094",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42839094",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42839094",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42839094",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42839094",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42841227",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42841227",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42841227",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42841227",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42841227",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42842666",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42842666",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42842666",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42842666",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42842666",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42844664",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42844664",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42844664",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42844664",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42844664",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42853938",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42853938",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42853938",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42853938",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42853938",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42858183",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42858183",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42858183",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42858183",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42858183",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42806945",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42806945",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42806945",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42806945",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42806945",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42723144",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42723144",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42723144",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42723144",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42723144",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42808804",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42808804",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42808804",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42808804",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42808804",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42585340",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42585340",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42585340",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42585340",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42585340",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42669067",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42669067",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42669067",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42669067",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42669067",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42811065",
        "id": "H1",
        "recommend": false
      },
      {
        "task": "R42811065",
        "id": "H2",
        "recommend": false
      },
      {
        "task": "R42811065",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42811065",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42811065",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42841610",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42841610",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42841610",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42841610",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42841610",
        "id": "H5",
        "recommend": false
      },
      {
        "task": "R42806003",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42806003",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42806003",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42806003",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42806003",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42806572",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42806572",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42806572",
        "id": "H3",
        "recommend": true
      },
      {
        "task": "R42806572",
        "id": "H4",
        "recommend": true
      },
      {
        "task": "R42806572",
        "id": "H5",
        "recommend": true
      },
      {
        "task": "R42748238",
        "id": "H1",
        "recommend": true
      },
      {
        "task": "R42748238",
        "id": "H2",
        "recommend": true
      },
      {
        "task": "R42748238",
        "id": "H3",
        "recommend": false
      },
      {
        "task": "R42748238",
        "id": "H4",
        "recommend": false
      },
      {
        "task": "R42748238",
        "id": "H5",
        "recommend": true
      }
    ]
  },
  "drawSha256": "63ccb838ab31e1e560712e072c8840a6818a646dc4c4e580a4024f2e8d041983",
  "interpretation": "One fair binary recommendation per hypothesis; no biological probability assigned. Labels measure correspondence to a saved article, not future experimental truth. Same recorded coin on both endpoints and all policies. Paired PPV comparisons exclude technical missing model forecasts from both sides, with original questions jointly resampled 10000 times. Descriptive intervals condition on this one recorded coin; they do not integrate uncertainty across new toss sequences or test adjusted multiple-comparison significance.",
  "reproduction": "Saved model outputs are in reconstruction.zip; coin-reconstruction.zip adds this design, the recorded draw, the addendum and the frozen public inputs. Analysis code is available on request."
}
