Measuring how often specific pairs of players co-appear in passages of play leading to each outcome type (goals, shots, lost possession), then weighting by outcome importance to produce a pair synergy score. Unlike individual player metrics, pair synergy captures partnerships that are more effective than the sum of their parts — pairs that frequently appear together in goal-scoring sequences but rarely in loss-of-possession sequences. This identifies strong links (pairs that drive positive outcomes) and weak links (pairs that appear in negative outcomes) within a team.
(1) For each passage of play, identify all player pairs who participated (consecutive passers, or all players involved in the sequence). (2) For each pair, increment a counter for the outcome type of that passage. (3) After processing all passages, each pair has a frequency vector across outcome types. (4) Apply the same outcome weights learned via logistic regression (from outcome-weighted-passing-network) to compute a weighted pair value. (5) Rank pairs by their synergy score. High-ranked pairs are those who frequently co-appear in positive outcomes; low-ranked pairs frequently co-appear in negative outcomes.
Key insight: Some pairs rate highly as a pair but neither player rates highly individually. Robertson-Mane at Liverpool is the canonical example — their pair synergy score was elite, but neither dominated the individual metrics. This shows the pair metric captures something the individual metric misses: the combination is greater than the sum of its parts.
Pair synergy scoring reveals partnerships that outperform the sum of their parts. Robertson-Mane at Liverpool had elite pair synergy despite neither player dominating individual metrics. Conversely, two individually elite players can have negative synergy if they occupy the same tactical space. The pair metric captures emergent value that individual metrics structurally cannot.