I am working in a group project on how to model ball/racket interaction in the game of Pong. In our model, we consider the ball as a particle unable to spin and are interested in how the friction between the particle and the moving racket can make the ball change direction in the direction tangential to the movement of the racket. I have found some useful insights but we noticed a weird thing. Consider a ball and a racket in a fixed frame independent of the ball and racket. Define tangential direction as being parallel to the trajectory of the racket. Assume that the ball hits the racket and that their relative speeds are such that the ball bounces off with opposite tangential speed. In other words, the ball hits the racket with an incidence angle alpha and bounces off following its trajectory backwards with the same reversed speed.
If we formulate the law of kinetic energy on the ball, we find that the work on the ball is zero, since the squared speeds before and after are the same. But how does it makes sense that the ball changed direction without any work on it?
EDIT More specifically, if I consider the ball in positions 1 and 2, respectively before and after the bounce, and write the total work on the ball accordingly to the law of kinetic energy, I get: $U_{1-2} = \frac{1}{2}mv_2^2 - \frac{1}{2}mv_1^2 = 0$, under the assumption that $v_2 = - v_1$. Where did all the work go? How can the total work sum to zero and the ball change its course?