Documentation
Research notes
Each note is shown as a summary plus a sample. Full source on GitHub.
Markets & strategy
Betting Markets and Odds — Landscape Note (Area C)
The first thing to map. Without a clear picture of what bookmakers price, how they price it, and which markets are sharp vs soft, every modeling decision downstream is mis-aimed.
markets-and-odds.md
Cross-Sport Lessons — Landscape Note (Area D)
What other betting markets and sports modelling traditions have already solved that football can copy. Most of the foundational machinery in modern sports analytics did not originate in football, and
cross-sport-lessons.md
Exploitable Markets — Phase 1 Hypothesis List
Where Phase 3 should aim first. Ranked by a rough product of *expected edge size × accessibility × validatability*. These are hypotheses derived from Area C, F, and the cross-sport notes — not validat
exploitable-markets.md
Incumbent Baselines
The strongest *public* model in each category, plus the metric it claims and where the claim comes from. The point is to set a target: if our Phase 3 baseline can't match these, we don't have an edge
incumbent-baselines.md
Quantitative Betting Framework — Landscape Note (Area E)
A great probability model is worthless without disciplined sizing and validation. This note maps the machinery that turns a calibrated probability into a sustainable edge: calibration, edge estimation
betting-framework.md
Modeling
Contextual and Structural Factors — Landscape Note (Area G)
The "polish layer." Each effect is small individually, but bookmakers price most of them crudely or with stale parameters. Aggregated, they're the difference between a model that matches the market an
contextual-factors.md
Player Expected Value in Football — Research Overview
In football analytics, "expected value" refers to a family of metrics that assign a numerical value to events, game states, or players based on the probability of contributing to a goal (or preventing
player-expected-value.md
Player Quality — Landscape Note (Area A, companion)
`player-expected-value.md` covers the action-valuation half of player quality (xG, xA, xT, VAEP, EPV, OBV, off-ball valuation, defensive metrics). This note covers the rest of section A: latent-skill
player-quality.md
Team Success and Match Outcome Modeling — Landscape Note (Area B)
This is the spine of the stack: the model whose output is what we sell against the bookmaker. Player ratings ultimately feed into here; market mechanics define the target metric. The job of this note
team-modeling.md
Data sources
Data Sources — Landscape Note (Area F)
Pinning down what's actually obtainable, cheaply, before designing models around feeds we'll never have. The bounding constraint on most amateur stacks is data, not modelling skill.
data-sources.md
Data Sources Matrix
Source × coverage × cost × access difficulty. Use this to decide what to integrate first and where the binding constraints are. Companion to `data-sources.md` (which has narrative context).
data-sources-matrix.md
Project & design
Prediction-site UI/UX — what to adopt from Polymarket and Kalshi
The two best-known prediction-market frontends — Polymarket and Kalshi — have
prediction-site-ux.md
Research Plan: Football Player Quality, Team Success, and Betting Odds
A roadmap for what to research, in what order, and why. **No findings yet** — this document only maps the territory and sequences the work.
research-plan.md
Why Each Research Area Matters
Companion to `research-plan.md`. Explains *why* the seven topic areas (A–G) were chosen and how they connect to the goal of estimating player quality, team success, and exploiting betting markets. Rea
topic-rationale.md
Other
Player Data Curation for WC2026 — Plan
How we go from "the 48 qualified nations" to "every player who plausibly
player-data-curation.md
Practitioner History — How People Actually Find and Realise Betting Edge
The other Phase 1 notes catalogue methods (Dixon-Coles, xG, Elo, Kelly) and market mechanics (CLV, sharp/soft, overround). This note covers the **practitioner layer**: who has actually made money in s
practitioner-history.md
What People Model in Soccer Betting — Targets and Variables Catalog
A cross-cutting synthesis. The other Phase 1 notes are organised by topic area (markets, teams, players, context). This note is organised by **prediction target**: for each thing a bettor or analyst t
modeling-targets-and-variables.md