Metraipolis › archive › dc_77dd2ca3b742
FromPitch2Board: Benchmarking LLM Agents in Long-Horizon Football Management — arXiv:2609.34710
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.34710
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T20:58:05.096Z
- SHA-256 of the archived text
b439e1b44a5e7b0e8f1d57afef4f7bc12f0d0671eb64a46cbbe2b5c645756316- Stable URL of this page
https://metraipolis.com/documento/dc_77dd2ca3b742
How to cite
FromPitch2Board: Benchmarking LLM Agents in Long-Horizon Football Management — arXiv:2609.34710 — arXiv (en), https://arxiv.org/abs/2609.34710 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_77dd2ca3b742 (Metraipolis archive dc_77dd2ca3b742, sha256 b439e1b44a5e7b0e8f1d57afef4f7bc12f0d0671eb64a46cbbe2b5c645756316)
BibTeX
@misc{dc77dd2ca3b742,
title = {FromPitch2Board: Benchmarking LLM Agents in Long-Horizon Football Management — arXiv:2609.34710},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.34710},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_77dd2ca3b742; sha256 b439e1b44a5e7b0e8f1d57afef4f7bc12f0d0671eb64a46cbbe2b5c645756316},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34710. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-horizon agent benchmarks typically report how far an agent progresses, but do not identify whether its performance comes from the foundation model, scaffold, responsibility scope, match-control granularity, or horizon. We introduce FromPitch2Board, a deterministic football-management benchmark that studies five configurable factors through controlled comparisons on a single simulator, using paired seeds and a frozen calibration. We evaluate four foundation models and four agent scaffolds. In the Model Track, Coach points Z-scores span 0.19, while Manager points Z-scores span 0.68, with GPT-5.6 showing a sharp rise in passivity under responsibility expansion. Its responsibility ladder rises from 46.1 to 58.1 points with recruitment, then falls to 46.8 under full management, localizing the regression to the final responsibility boundary. Across that boundary, its skipped-decision rate rises from 1.1% to 57.9%. Within the Flash-Pro pair crossed across every scaffold, scaffold choice changes Manager points Z-scores by up to 0.48 relative to the fixed stateless scaffold. The 3Y cohort shows a directional reversal in mean ranking between years one and three, while a selected Claude Code+Pro configuration peaks in year three and remains below that peak, showing that responsibility scope and horizon expose behavior changes that a single headline score conceals. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.34710 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T20:29:01.636Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).