In many equestrian pursuits such as dressage and show-jumping, it is important that the horse exhibits the same level of balance when ridden to the left as when ridden to the right in canter – that is, to show no motor bias. It is a long-held belief within such disciplines that to reduce bias that exists in horses and thus to enhance symmetry of performance to the left and right, the horse needs to be worked equally in both directions, although there is a lack of scientific evidence of this influencing bias. There also is little compelling evidence for either the existence or absence of motor bias in unridden (and therefore younger) or ridden (and therefore older) horses. In this study, we tested whether there was a difference in motor bias between unridden (n = 15) and ridden (n = 15) horses when their balance was challenged by cantering them in circles both to the left and to the right on the lunge. As indicators of a difference in balance between the left and right and thus as indicators of motor bias, we conducted three lunging tests – time spent in canter, whether the horse cantered on the correct lead and whether it became disunited. A grazing stance test, where the extended foreleg during grazing was recorded as the preferred forelimb, was also used to compare responses in a test where balance was not actively challenged, to the three lunging tests where balance was actively challenged. No bias was found in either the unridden or ridden groups when their balance was challenged, but ridden horses exhibited a motor bias in grazing stance – when their balance was not challenged. There was also a correlation between the responses in all three lunging tests, but none between the grazing stance test and any of the three lunging tests. We therefore conclude that neither ridden nor unridden horses are biased when their balance is challenged; thus it cannot be concluded that ambidextrous training affects an inherent bias, and that estimation of motor bias in horses is affected by the test conditions. Finally, if ridden horses are truly unbiased, strong human motor bias might be responsible for the common perception amongst riders that horses are biased.