Home/Soccer Analytics/Pass Risk-Gain Decomposition

Pass Risk-Gain Decomposition

Passing MetricsLevel 3 — Advanced

What It Is

Decomposing every pass into two independent dimensions: risk (probability of interception, [-1, +1]) and gain (change in expected position value, ΔEPV). Plotting all passes in risk-gain 2D space reveals four distinct quadrants: (1) high risk, high gain — breakthrough passes near goal, (2) low risk, low gain — safe recycling backward, (3) high risk, low gain — bad passes that risk turnover without advancing, (4) low risk, high gain — nearly empty, because opponents prevent safe high-gain passes. The risk parameter is computed kinematically by simulating ball trajectory and defender/teammate interception probability from 360 data; the gain is ΔEPV from start to end zone.

Correct Execution

(1) Risk calculation: for each pass, simulate the ball trajectory (start → end). For each defender in 360 data, compute time to intercept based on distance to trajectory and assumed movement speed. Do the same for attacking teammates. Risk = P(defender intercepts) - P(teammate receives). Normalize to [-1, +1]: +1 = certainly intercepted, -1 = certainly received by teammate. (2) Gain calculation: compute EPV at pass destination minus EPV at pass origin. Positive = advanced toward goal, negative = retreated. (3) Plot in 2D: each pass is a point in risk × gain space. The cloud shape reveals the team's passing character. (4) Aggregate per player: average risk, average gain, risk decision parameter, gain decision parameter.

Key finding: the majority of completed passes cluster in the low-risk, slightly-positive-gain zone. The high-risk/high-gain quadrant contains the breakthrough passes that create chances. The low-risk/high-gain quadrant is nearly empty — defenses ensure safe passes don't produce high gain.

Progression Levels

Diagnostic Tree

Coaching Cues

  • "Messi takes risks that would be reckless for anyone else."
  • "The empty zone — low risk, high gain — is empty because defenses exist."
  • "High risk, low gain = bad pass. High risk, high gain = the pass that wins games."

Common Errors

  1. Combining risk and gain into a single score: Risk and gain are orthogonal. A high-risk/high-gain pass and a low-risk/low-gain pass can both be "correct" depending on context.
  2. Not simulating ball and player speeds: Without kinematic simulation, risk estimates are just distance-based. Speed determines whether an interception is physically possible.
  3. Including incomplete passes: Risk analysis on completed passes only shows execution quality. Including incomplete passes shows decision quality. Both are valid but measure different things.

Edges

🔑 Hidden Causal Lever

The "Empty Quadrant" Proves No Safe High-Gain Passes Exist

In risk × gain 2D space, the low-risk/high-gain quadrant is nearly empty. Defenses are organized specifically to ensure safe passes don't produce high gain. There is no free lunch: high gain requires high risk. Any model claiming "safe and dangerous" passes is miscalibrated.

What most people do
Search for "safe, high-value" patterns. Criticize players for not finding the "easy" dangerous pass.
What the best do
Accept the tradeoff as structural. Focus on execution quality at the risk frontier rather than trying to move it. Build the team around players who can complete at the frontier (Messi: highest risk AND highest completion).
Why it's an edge: Accepting the empty quadrant changes the framework from "find better passes" to "execute harder passes better."
How to exploit: Profile each player's execution in the high-risk/high-gain quadrant. Build the attacking plan around getting the ball to whoever can complete there.
Complex Systems Group, Madrid, StatsBomb Conference, 2021-11-04

Sources

  • Complex Systems and Sports Analytics Group, Madrid, StatsBomb Conference 2021, YouTube, 2021-11-04 — presented kinematic pass risk simulation with 360 data; 2D risk-gain decomposition; showed Barcelona 37-match analysis; identified temporal shifts (risk increases in second half)