The incidence of soccer (football) injuries is among the highest in sports. Despite this high rate, insufficient evidence is available on the efficacy of preventive training programmes on injury incidence. Objective To systematically study the evidence on preventive exercise-based training programmes to reduce the incidence of injuries in soccer. Data sources The databases EMBASE/MEDLINE, PubMed, CINAHL, Cochrane Central Register of controlled trials, PEDro and SPORTDiscus™ were searched for relevant articles, from inception until 20 December 2011. The methodological quality of the included studies was assessed using the PEDro scale. Study selection The inclusion criteria for this review were (1) randomized controlled trials or controlled cl inical trials; (2) primary outcome of the study is the number of soccer injuries and/or injury incidence; (3) intervention focusing on a preventive training programme, including a set of exercises aimed at improving strength, coordination, flexibility or agility; and (4) study sample of soccer players (no restrictions as to level of play, age or sex). The exclusion criteria were: (1) the article was not available as full text; (2) the article was not published in English, German or Dutch; and (3) the trial and/or training programme relates only to specific injuries and/or specific joints. To compare the effects of the different interventions, we calculated the incidence risk ratio (IRR) for each study. Results Six studies involving a total of 6,099 participants met the inclusion criteria. The results of the included studies were contradictory. Two of the six studies (one of high and one of moderate quality) reported a statistical significant reduction in terms of their primary outcome, i.e. injuries overall. Four of the six studies described an overall preventive effect (IRR<1), although the effect of one study was not statistically significant. The three studies that described a significant preventive effect were of high, moderate and low quality. Conclusions Conflicting evidence has been found for the effectiveness of exercise-based programmes to prevent soccer injuries. Some reasons for the contradictory findings could be different study samples (in terms of sex and soccer type) in the included studies, differences between the intervention programmes implemented (in terms of content, training frequency and duration) and compliance with the programme. High-quality studies investigating the best type and intensity of exercises in a generic training programme are needed to reduce the incidence of injuries in soccer effectively.