Your explanation is a pretty good one for someone who admits to having struggled with kinematics.
The turning moment, more correctly defined as ''moment of force'', induced by the asymmetric contact projectile/target interface is dependent upon the angle of the impact face. There is also a critical angle,
ß, at which the projectile cannot overcome the dynamic strength, R
T, of the target being struck which results in the projectile being fully deflected (ricochet) from the target face, failing to defeat it.
In the AT model, Tate (of Alekseeskii and Tate) computes the deflection vector as being perpendicular to the axis of the rod multiplied by one-half the length of the rod, to obtain the moment of force about the projectile's center of gravity. This moment induces a longitudinal rotation in the projectile body—which is treated as being rigid—that results in the critical ricochet condition.
Attachment 94235
In later work, Rosenberg et al. go about the process of determining the critical deflection angle differently. In their approach, the asymmetric force assumed in Tate’s model is assumed to act only upon the mass located at the tip of the projectile which is engaged by direct contact at the nose-target interface. This alteration was made in order to account for the plastic hinge that occurs as the forward portion of the rod bends during its deflection from the target face. Rosenberg's model is also dependent upon the dynamic target strength, R
T, which is velocity-dependent in that as impact velocity increases R
T decreases, which requires iterative computational steps to correctly represent the extent of the elastic-plastic zone formed in the target ahead of the projectile's nose.
Attachment 94236
ETA: in the equations cited above, U is velocity of the projectile's nose at the point of contact with the target and V is the projectile's impact velocity.