Interception of fly balls requires active locomotion toward the point where catching can take place; as a results, the visual information guiding interception is affected by the catcher's own movement. The only interception theory currently available for a catcher standing in the plane of motion of the ball is Optical Acceleration Cancellation (OAC); in this strategy, the pseudo-optical variable "optical acceleration" (OA), if nonzero, specifies how the catcher should adjust his current velocity. We formulate a precise implementation of OAC where the catcher strives to maintain OA zero at all times and analyze its implications in terms of the catcher's interception behavior for different ball trajectories under air-friction-free, low-friction, and friction-dominated conditions. We conclude that the point in the ball trajectory where first visual contact (FVC) takes place determines to a large extent the ensuing interception behavior of the catcher. Conventional trajectories (FVC slightly above eye level, ball coming toward the catcher) result in fast acceleration to a constant velocity and successful interception. Trajectories with FVC below eye level typically result in unsatisfactory behavior of the catcher, who runs away from rather than toward the point of interception. In addition, ball trajectories are identified for which the OA equals zero even though the catcher is not on an interception course. We find that two different formulations of OAC in the literature ("get rid of OA" and "choose acceleration direction based on the sign of OA") actually represent different strategies. Results of this study show that the first formulation is effective for a limited class of ball trajectories only. Regarding the second formulation, which was not analyzed in detail, we argue that it cannot result in a generally adequate strategy either and conclude that variables other than OA are indispensable for successful interception.