Introduction

Data displays several attributes (full documentation) :

Match Statistics (where available)

Div Date HomeTeam AwayTeam FTHG FTAG FTR HTHG HTAG HTR HS AS HST AST
D1 27/08/16 Augsburg Wolfsburg 0 2 A 0 1 A 13 12 2 6
D1 27/08/16 Dortmund Mainz 2 1 H 1 0 H 17 12 8 4
Div Date HomeTeam AwayTeam HF AF HC AC HY AY HR AR B365H B365D B365A
D1 27/08/16 Augsburg Wolfsburg 15 17 2 3 1 2 0 0 2.80 3.25 2.6
D1 27/08/16 Dortmund Mainz 3 20 7 2 0 3 0 0 1.36 5.00 8.5

Dataset gathers data from various leagues and countries

League Fullname # of matchs
D1 Bundesligua 1044
E0 Premier League 1290
F1 Ligue 1 1298
I1 Serie A 1286
SP1 LaLiga 1280

Number of goals per match

Mean winning odds

Div winning_odds A D H
E0 2.897450 0.30 0.25 0.45
D1 2.864320 0.28 0.25 0.47
F1 2.776926 0.29 0.25 0.46
SP1 2.748758 0.29 0.24 0.47
I1 2.730163 0.30 0.25 0.45

Mean Goals as a feature

Random Forest

Accuracy: 46%

Neural Network

Accuracy: 52.1%

XGBoost

## [1] "Accuracy: 50.7%"

Accuracy: 50.7%