How we make predictions

How match previews work

概要 + サンプル · 全文は 1,390 語

概要

Every fixture page on the site opens with a detailed match preview — predicted formations, key players, strategic factors, and expected scorelines. Here's how we generate them and what data feeds into each section.

サンプル

What the reader sees

Every fixture page now opens with a structured prose summary covering:

  1. Headline expected shape. Favoured side, Dixon-Coles λ_home and λ_away, P(H/D/A), modal scoreline.
  2. Predicted formations and key players. One paragraph per side, naming the formation string and the highest-rated player by anytime-scorer model output (Model #5), with a role descriptor derived from squad position (FW → "primary attacking outlet", MF → "creator from midfield", DF → "set-piece / late-runner threat from the back", GK → "goal-line presence").
  3. Strategic factors. Whichever of set-piece propensity, weather / altitude, referee tendency, pace differential (Elo gap), and injury notes clear the per-axis meaningfulness thresholds. Below threshold the factor is simply omitted.
  4. Set-piece outlook. Per-side set-piece xG share multiplied by the corresponding DC λ to give an expected set-piece-goals number.

…

全文

Pro Pass

全文をご覧になりますか？

How match previews work は 1,390 語です。Pass でこのドキュメントとすべてのリサーチノートの全文を解除できるほか、試合別確率、4モデル比較、試合別戦術分析も含まれます。

View pricing →

24h self-service refund·No subscription, no auto-renewal·Access through 31 Dec 2026. See refund policy.