Incorrect bicycle configuration may predispose athletes to injury and reduce their cycling performance. There is disagreement within scientific and coaching communities regarding optimal configuration of bicycles for athletes. This review summarizes literature on methods for determining bicycle saddle height and the effects of bicycle saddle height on measures of cycling performance and lower limb injury risk. Peer-reviewed journals, books, theses and conference proceedings published since 1960 were searched using MEDLINE, Scopus, ISI Web of Knowledge, EBSCO and Google Scholar databases, resulting in 62 references being reviewed. Keywords searched included 'body positioning', 'saddle', 'posture, 'cycling' and 'injury'. The review revealed that methods for determining optimal saddle height are varied and not well established, and have been based on relationships between saddle height and lower limb length (Hamley and Thomas, trochanteric length, length from ischial tuberosity to floor, LeMond, heel methods) or a reference range of knee joint flexion. There is limited information on the effects of saddle height on lower limb injury risk (lower limb kinematics, knee joint forces and moments and muscle mechanics), but more information on the effects of saddle height on cycling performance (performance time, energy expenditure/oxygen uptake, power output, pedal force application). Increasing saddle height can cause increased shortening of the vastii muscle group, but no change in hamstring length. Length and velocity of contraction in the soleus seems to be more affected by saddle height than that in the gastrocnemius. The majority of evidence suggested that a 5% change in saddle height affected knee joint kinematics by 35% and moments by 16%. Patellofemoral compressive force seems to be inversely related to saddle height but the effects on tibiofemoral forces are uncertain. Changes of less than 4% in trochanteric length do not seem to affect injury risk or performance. The main limitations from the reported studies are that different methods have been employed for determining saddle height, small sample sizes have been used, cyclists with low levels of expertise have mostly been evaluated and different outcome variables have been measured. Given that the occurrence of overuse knee joint pain is 50% in cyclists, future studies may focus on how saddle height can be optimized to improve cycling performance and reduce knee joint forces to reduce lower limb injury risk. On the basis of the conflicting evidence on the effects of saddle height changes on performance and lower limb injury risk in cycling, we suggest the saddle height may be set using the knee flexion angle method (25-30°) to reduce the risk of knee injuries and to minimize oxygen uptake.