{"id":"W2982119900","doi":"10.48550/arxiv.1910.12622","title":"Trouble with the Curve: Predicting Future MLB Players Using Scouting Reports","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Sports Analytics and Performance","field":"Economics, Econometrics and Finance","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"League; Minor (academic); Valuation (finance); Computer science; Data science; Artificial intelligence; Operations research; Machine learning; World Wide Web; Engineering; Political science; Business; Accounting","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0007154264,0.0003283909,0.0005277189,0.0002289277,0.0003358524,0.0001606463,0.0004603004,0.0002784225,0.0002111898],"category_scores_gemma":[0.00001387331,0.0003038056,0.0002169526,0.0003991833,0.00009187865,0.0002633327,0.0004025522,0.0007869377,0.00004236423],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002508847,"about_ca_system_score_gemma":0.0001332345,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001243961,"about_ca_topic_score_gemma":0.0001226341,"domain_scores_codex":[0.9980907,0.00001347166,0.0004410458,0.0009903017,0.00004283565,0.0004216014],"domain_scores_gemma":[0.9973936,0.00003373394,0.001340673,0.001057897,0.00008185939,0.00009221657],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002155883,0.00001735792,0.4185462,0.00005907883,0.0001309419,0.000112813,0.0001864835,0.5563715,5.291665e-7,0.02434678,0.0001922301,0.000014532],"study_design_scores_gemma":[0.0004205377,0.00004522589,0.0134907,0.0001676241,0.0001153235,0.00003174343,0.0008473276,0.9701169,0.000007140115,0.00251419,0.01157873,0.0006646279],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9691104,0.0005254151,0.01674905,0.0001129228,0.001256307,0.0003557173,0.0000611233,0.00005946051,0.01176963],"genre_scores_gemma":[0.9947333,0.0002387634,0.0001046649,0.00009905385,0.0004446504,6.595889e-7,0.00003110716,0.00004671033,0.00430111],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4137453,"threshold_uncertainty_score":0.9999414,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07420720122022083,"score_gpt":0.1721069129775375,"score_spread":0.09789971175731667,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}