The strength of weak ties in professional football (soccer): evidence from 1,823 matches across five European leagues

Abstract

Abstract Granovetter’s strength of weak ties theory predicts that infrequent connections carry disproportionately novel information, yet untested in high-performance team settings. Using StatsBomb event data from 1,823 matches across five European leagues (2015/16), we construct directed weighted passing networks for each team-match, yielding 389,943 passer-receiver edges, and measure offensive threat through progressive pass rate, expected threat, and geographic threat. OLS regressions with team and positional-pair fixed effects show that weak ties produce significantly higher threat on all three metrics (progressive rate: β = +0.015, 95 % CI [+0.009, +0.021], p < 0.001), replicating in all five leagues; positional-pair controls rule out forward-passing geometry as an explanation. A permutation test confirms the advantage exceeds what network structure alone can produce, and each successive use of a connection reduces its progressive rate ( β = −0.032, p < 10 −17 ), evidencing the novelty mechanism. Weak ties carry a modest completion cost but no residual possession-loss penalty and generate more shots for the passing team; the premium is largest when teams are trailing, survives alternative tie classifications, and is stable across 20 supplementary seasons. These findings extend weak tie theory to adversarial team domains and suggest that diversifying passing repertoires may enhance offensive threat.