Breeding synchrony is hypothesised to influence the occurrence and frequency of extra-pair fertilisations (EPFs) in birds irrespective of the social mating system. The two proposed hypotheses make opposite predictions. (1) Synchronous breeding leads to a lower frequency of EPFs because males face a trade-off between mate guarding and obtaining additional matings via extra-pair copulations (EPCs) ('guarding constraint' hypothesis). (2) Synchronous breeding promotes EPFs because females are able to compare displaying males simultaneously, which provides them with more reliable cues for extra-pair mate choice ('mate assessment' hypothesis). In a study of great reed warblers (Acrocephalus arundillaceus) from 1987-1998, annual breeding was asynchronous and the frequency of EPFs was rather low (extra-pair young occurring in 6.4% of the broods). Within this population, however, there was no relationship between the frequency of EPFs and breeding synchrony, thus not favouring any of the two hypotheses. Contrary to assumptions of the hypotheses, mate guarding did not seem to constrain males from engaging in EPCs (disfavouring the 'guarding constraint' hypothesis), and females seem to have repeated opportunities to compare males irrespective of breeding synchrony (disfavouring the 'mate assessment' hypothesis). Our results suggest that breeding synchrony is not an important factor influencing patterns of EPFs in great reed warblers. The low frequency of EPFs may instead be explained by the socially polygynous mating system, where females are less constrained in their choice of a social male.