Home/Soccer Analytics/GK Positioning vs. Shot-Stopping Decomposition

GK Positioning vs. Shot-Stopping Decomposition

Goalkeeper AnalysisLevel 4 — Expert

What It Is

Training two post-shot xG models — one without goalkeeper location (estimating xG against average positioning) and one with it — and using the difference to isolate the positioning contribution from the shot-stopping contribution. Enables counterfactual "what if the keeper stood here" analysis by sweeping goalkeeper position across the goal and running inference. The optimal position surface emerges as a property of the model, not a separate optimization.

Correct Execution

Model A: post-shot xG without GK position → estimates shot danger against "average" positioning. Model B: post-shot xG with GK position → estimates shot danger given actual positioning. Positioning value = Model A output - Model B output. Positive = good positioning reduced danger. Negative = bad positioning increased danger.

Diagnostic Tree

Edges

🔑 Hidden Causal Lever

GSAA Conflates Two Independent Skills That Require Opposite Training

Standard GSAA (Goals Saved Above Average) lumps positioning quality and shot-stopping reflexes into one number. A goalkeeper with elite reflexes and poor positioning can produce the same GSAA as one with elite positioning and average reflexes — but the training prescriptions are opposite, the sustainability is different (reflexes decline with age, positioning improves), and the team-building implications are different.

What most people do
Evaluate goalkeepers on aggregate GSAA and assume high GSAA means "good goalkeeper" without decomposing the source.
What the best do
Train two post-shot xG models — one with GK position, one without — and use the difference to isolate positioning value from shot-stopping value. Then match the GK's strength profile to the team's needs.
Why it's an edge: A GK whose value comes from positioning is more sustainable (positioning improves with coaching and age) than one whose value comes from reflexes (decline after ~30). Recruiting a 28-year-old reflex-dependent GK is buying a depreciating asset. Recruiting a 28-year-old positioning-elite GK is buying an appreciating one.
How to exploit: Decompose every GK target's GSAA into positioning and shot-stopping components. Favor positioning-dominant GKs for long-term contracts. Use reflex-dominant GKs as short-term solutions only.
Dr. Dinesh Vatvani, StatsBomb Conference, 2022-10-04. Two-model approach enables counterfactual "what if the keeper stood here" analysis.

Sources

  • Dr. Dinesh Vatvani, StatsBomb Conference 2022, YouTube, 2022-10-04