xG & Advanced Stats
How Expected Goals Models Actually Work — explained
What the model sees before it prints a number, and where that process leaks.
Core metricxG & Advanced Stats
How the expected-goals family is built: the ingredients of a shot value, the non-penalty variant, the reasons a good or bad finish is normal, regression to the mean and the sample sizes that make a number trustworthy.
xG & Advanced Stats
The Ingredients of a Shot Value: definition and limits
Distance, angle, body part, service type and pressure, one by one.
Core metric
xG & Advanced Stats
Non-Penalty xG and Why It Is Preferred — explained
Why the penalty is removed, and what changes once it is.
Applied metric
xG & Advanced Stats
Why Overperforming xG Is Completely Normal — explained
Variance, selection and skill: the three honest explanations.
Applied metric
xG & Advanced Stats
Regression to the Mean, Read Properly: the numbers
What actually regresses, what does not, and how to tell the difference.
Advanced / contested
xG & Advanced Stats
Sample Size and When xG Becomes Reliable — explained
How many shots, passes and minutes before a metric says anything.
Advanced / contested
xG & Advanced Stats
Expected Assists and Goal Involvement: the numbers
How the pass earns the shot's value, and where the pairing breaks down.
Applied metric
xG & Advanced Stats
Separating Open-Play and Set-Piece xG: the numbers
Why the phase split changes the reading of a team's attack.
Applied metric