Methodology

Methodology and about this database:

This project is a self-maintained database of match statistics, created for personal use and public reference. The aim is to provide consistent, match-by-match indicators of team performance rather than definitive or absolute model-driven metrics.

All data is recorded manually following each match, using publicly available statistical information as guidance. 

The data provided is not official, and all data is compiled using publicly available data from numerous stats providers to create a broad representation of match statistics. 

This database exists to support analysis, discussion, and long-term tracking of team performances. It is a personal analytical project rather than an official statistical record and should be used and interpreted accordingly.

Data Sources

Match statistics such as possession and shots are recorded using publicly available Opta match reports as a reference point. Opta definitions are used as general guidance for consistency; however, the data provided is independently compiled and is not an official Opta product.

All sourced data is reviewed and compared to multiple open source data sources, with manual modifications taking place if errors are interpreted in these data sets. 

Possession and Shots

Possession values reflect the match possession percentages reported in Opta match summaries found on SofaScore. 

These figures represent the share of total ball possession over the course of the match and are recorded at full-time.

Shot totals are based on Opta’s standard definition of a shot, including:

• shots on target
  
  

• shots off target
  
  

• shots blocked
  
  

Penalty shootouts are excluded.

Official shot counts found on Opta and SofaScore may differ from this website, where manual changes sometimes occur if deemed necessary.

Big Chances

Big chances are manually classified on a shot-taken basis.

The classification follows Opta’s published definition as a guideline - generally referring to situations where a player is expected to score, such as close-range attempts with limited defensive pressure, though each chance is assessed independently and results may differ to Opta's given value for matches.

While this metric is informed by established analytics standards, it involves subjective judgement. The same internal criteria are applied consistently across all matches to allow meaningful comparison over time.

Expected Goals (xG) 

Expected goals (xG) values in this database are approximated manually through synthesis of existing open source values.

Each shot is assessed based on factors including:

• Shot location

• Angle to goal

• Defensive pressure

• Chance context (e.g. cut-backs, one-on-ones, rebounds)

These xG figures should not be directly compared with model-based xG outputs from providers such as Opta, StatsBomb, Understat or similar analytics services due to the large human-input effects of this model. While existing model-based xG values provide a baseline, human input is intentionally applied to account for contextual nuances that are often under-represented in automated models. 

Match Momentum Graphs

All match momentum graphs are sourced directly from www.sofascore.com due to their accuracy and great presentation. 

Consistency and Limitations of the data

Care is taken to apply definitions and judgement consistently across matches and seasons. However, due to the manual and interpretive nature of some metrics (particularly big chances and xG) a degree of subjectivity is unavoidable. It is important to note that this is an important characteristic of all data analysis - each model differs, there is subjectivity in all data. Analyse using multiple sources of data to guarantee a well rounded perspective.

As a result:

• xG values, as with all data, should be interpreted as indicative rather than definitive

• Trends over multiple games are more meaningful than single-match values