World Cup Match Predictor V2 ⚽

Predict FIFA World Cup match outcomes using a comprehensive model that combines World Cup history, continental tournament performance, squad market value, and player experience data.

Data Foundation (V2 — Expanded)

| Source | Coverage | Data Points | |--------|----------|-------------| | FIFA World Cup | 22 tournaments (1930–2022) | 1,248 matches, 87 teams | | UEFA Euro | 2016, 2020/21, 2024 | 153 matches | | Copa América | 2019, 2021, 2024 | 86 matches | | Africa Cup of Nations | 2019, 2021/22, 2023/24 | 156 matches | | AFC Asian Cup | 2011, 2015, 2019, 2023/24 | 156 matches | | Transfermarkt | Current squads | 5,614 players, squad values, caps, goals |

Prediction Methodology (V2)

Six-factor weighted composite model:

| Factor | Weight | What it measures | |--------|--------|------------------| | Power V2 Score | 40% | Composite score: WC(20%) + Continental(20%) + SquadValue(25%) + ClubTier(15%) + Experience(10%) + ElitePlayers(10%) | | Continental Form | 20% | Recent continental tournament win rate gap (Euro, Copa, AFCON, Asian Cup) | | Squad Value | 20% | Total squad market value difference (Transfermarkt) | | Experience | 10% | Total international caps gap | | Elite Players | 10% | Players valued > €50M gap (game-changers) |

How to Use

Step 1: User asks about a match prediction

When user mentions a matchup, identify:

Team names (both sides)
Stage (group / round_of_16 / quarterfinal / semifinal / final) — default to "group"

Step 2: Run the prediction

cd ~/.workbuddy/skills/足球小老师/data && python predict.py <team1> <team2> [stage]

Examples:

python predict.py France Morocco semifinal
python predict.py Spain Belgium group
python predict.py Brazil Germany final

Team names: full names (Brazil, Argentina, England) or FIFA codes (BRA, ARG, ENG, ESP).

Step 3: Present results

Show:

Team profiles — Power V2, WC Win Rate, Squad Value, Elite Players
Continental records — titles and win rate in continental tournaments
Win probabilities — percentages for each team and draw
Expected score — estimated goals
V2 model contributions — which factors drove the prediction
⚠️ Disclaimer: Statistical prediction, for entertainment only

Step 4: Deeper analysis (optional)

Read team_stats_v2.json for full squad details (market values, caps, goals, club tiers)
Explain factor breakdown: "Spain's squad value (€1.51B vs €0.61B) and 3 Euro titles give them the edge"

Data Files

| File | Path | Content | |------|------|---------| | Match data | data/matches.csv | 1,248 World Cup match records | | Team stats v2 | data/team_stats_v2.json | 87 teams with comprehensive profiles | | Head-to-head | data/head_to_head.json | 736 team pair matchup records | | Continental data | data/continental/*.txt | 551 continental tournament matches | | Transfermarkt data | data/transfermarkt/*.csv.gz | 5,614 players with clubs & values | | Prediction engine | data/predict.py | V2 prediction script | | Analysis engine | data/analyze_v2.py | V2 data analysis script |

Top 20 Teams (Power V2)

| # | Team | Power V2 | Squad Value | WC WR | Cont. Titles | |---|------|----------|-------------|-------|--------------| | 1 | England | 67.6 | €1.98B | 48.0% | 0 | | 2 | France | 67.4 | €1.80B | 56.5% | 2 (Euro) | | 3 | Brazil | 67.1 | €1.57B | 67.6% | 1 (Copa) | | 4 | Argentina | 65.5 | €1.05B | 54.6% | 2 (Copa) | | 5 | Spain | 63.3 | €1.51B | 44.6% | 3 (Euro) | | 6 | Germany | 58.9 | €1.28B | 66.0% | 0 | | 7 | Portugal | 57.9 | €1.17B | 51.4% | 1 (Euro) | | 8 | Netherlands | 56.0 | €1.05B | 57.6% | 1 (Euro) | | 9 | Italy | 51.7 | €0.98B | 54.7% | 2 (Euro) | | 10 | Senegal | 43.1 | €0.53B | 30.0% | 1 (AFCON) | | 11 | Belgium | 41.0 | €0.61B | 43.1% | 0 | | 12 | Japan | 40.1 | €0.41B | 37.9% | 4 (AsianCup) | | 13 | Croatia | 38.6 | €0.42B | 56.7% | 0 | | 14 | Morocco | 36.9 | €0.62B | 26.1% | 0 | | 15 | Sweden | 36.4 | €0.51B | 47.3% | 0 | | 16 | USA | 36.2 | €0.57B | 58.6% | 0 | | 17 | Norway | 36.0 | €0.63B | 56.2% | 0 | | 18 | Nigeria | 34.9 | €0.41B | 22.7% | 0 | | 19 | Colombia | 33.8 | €0.30B | 34.2% | 0 | | 20 | Uruguay | 33.0 | €0.37B | 44.1% | 1 (Copa) |

Notes & Disclaimers

Model combines 6 data sources for the first time: WC + 4 continental tournaments + Transfermarkt squads
Squad market values reflect Transfermarkt data as of dataset snapshot date
Power V2 is recomputed from six sub-factors; see factor_breakdown in team_stats_v2.json
Club tier scores are approximate — based on predefined club ranking not league data
Always include disclaimer: "⚽ 足球比赛充满不确定性，此预测仅供娱乐参考，不构成任何投注建议。"
Data sources: Fjelstul World Cup Database (CC-BY-SA-4.0), openfootball/internationals, dcaribou/transfermarkt-datasets