Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note2 min read

From Score Matrices to Football-Aware Match-State Simulation: An Auditable LLM Harness for Exact-Score Reranking

Source

arXiv

arxiv.org/abs/2608.05030v1

Read the source

Source

Plain-Language Brief

The paper presents an auditable information harness that combines dynamic Poisson-family models and large language models (LLMs) to forecast football scores. The system goes through four iterations, each improving on the previous one in terms of exact-score accuracy. The best iteration (V4) achieved 14.7% Top-1 and 30.7% Top-3 exact-score accuracy for the first 150 matches of the 2025-26 English Premier League.

What the Document Says

  • The system combines dynamic Poisson-family models and LLMs through an auditable information harness.
  • There are four iterations of the system: V1, V2, V3, and V4.
  • V1 achieved 10.0% Top-1 and 26.7% Top-3 exact-score accuracy.
  • V3 reached 12.0% and 30.0% in Top-1 and Top-3 exact-score accuracy respectively.
  • V4 reached 14.7% Top-1 and 30.7% Top-3 exact-score accuracy.
  • V4 increased candidate coverage from 77.3% to 84.7%.
  • V1's native 1X2 distribution achieved certain accuracy, log loss, Brier score, and ranked probability score values.

Key Concepts Mentioned

  • Football score forecasting
  • Dynamic Poisson-family models
  • Large language models (LLMs)
  • Iterations
  • Exact-score accuracy

Claims Made by the Document

  • The combination of dynamic Poisson-family models and LLMs improves football score forecasting.
  • V4 achieved better exact-score accuracy than previous iterations.

Evidence Used

  • The results presented in the paper
  • The reported Top-1 and Top-3 exact-score accuracies for each iteration

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Football score forecasting combines a strong statistical core with a difficult contextual edge. Dynamic Poisson-family models estimate team strength, expected goals, and coherent score probabilities, but do not directly understand roles, tactical matchups, motivation, or how a first goal changes behaviour. Large language models (LLMs) can reason about such concepts, yet are not calibrated probability engines. We combine both components through an auditable information harness. This paper documents four iterations: V1, a dynamic score-driven Dixon-Coles baseline; V2, which maps LLM contextual ratings back into expected-goal parameters; V3, which replaces scalar correction with goal-by-goal simulations over a frozen score-candidate set; and V4, which adds shared first-breakthrough and post-goal cascade judgments, time-aware stopping, and deterministic tail candidates. The harness defines input semantics, supplies pre-match evidence, and constrains the LLM to an inspectable reasoning route. On a chronological replay of the first 150 matches of the 2025-26 English Premier League, V1 achieved 10.0% Top-1 and 26.7% Top-3 exact-score accuracy. V3 reached 12.0% and 30.0%, while V4 reached 14.7% and 30.7%. V4 increased candidate coverage from 77.3% to 84.7%, although no added tail candidate became a Top-3 exact hit. V1's native 1X2 distribution achieved 53.3% argmax accuracy, 0.9878 log loss, 0.5870 Brier score, and 0.2095 ranked probability score. These results are exploratory: the development slice is not an untouched benchmark, and temporal input isolation cannot exclude outcome memory in a closed LLM. The contribution is an auditable hybrid architecture, a clear design evolution, and negative findings showing where football-aware simulation does and does not improve score selection.