{
  "abstract": "Evaluating \"neutral launches\" (e.g., infrastructure upgrades) using traditional confidence interval overlap is flawed: it is dangerously permissive with scarce data and excessively restrictive with abundant data. To resolve this, this paper introduces Expected Bayesian Loss (EBL), a continuous metric that quantifies both the probability and expected severity of metric degradation. Computable directly from standard frequentist estimates, EBL explicitly penalizes empirical noise and high-variance experiments. Validated against expert decisions, EBL provides experimentation platforms with a rigorous, tunable guardrail that aligns statistical safety with institutional risk appetite.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.10223",
  "author_total": 6,
  "authors": [
    {
      "name": "Pablo Alcain"
    },
    {
      "name": "Jason Kang"
    },
    {
      "name": "Mia Garrard"
    },
    {
      "name": "Marine Veits"
    },
    {
      "name": "Houssam Nassif"
    },
    {
      "name": "Abbas Zaidi"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/cf9qasscm6sk0.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.10223",
  "fulltext_url": null,
  "id": "cf9qasscm6sk0",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.10223",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "stat.AP"
    ],
    [
      "openalex",
      "2613|Statistical Methods and Bayesian Inference"
    ]
  ],
  "openalex": "W7220913239",
  "pdf_url": "https://arxiv.org/pdf/2610.10223",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-07",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/cf9qasscm6sk0.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:48Z",
      "native_id": "2610.10223",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.10223"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.AP",
      "subfield": "statistics"
    }
  ],
  "title": "Neutral Is Not Free: Evaluating Downside Risk in Neutral Launches",
  "title_text": "Neutral Is Not Free: Evaluating Downside Risk in Neutral Launches",
  "type": "preprint",
  "updated_at": "2026-10-30T12:27:48Z",
  "url": "https://research.draconitic.com/paper/cf9qasscm6sk0",
  "venue": "arXiv",
  "versions": []
}