According to TP A.16, on page 7 (post speed results accounting for friction), you write, "...the CB picks up a little speed after impact for all ball-hit-fractions, except for a 0% ball-hit (90 degree cut angle)."
But the document shows that when shooting a 30° shot (halfball hit) with "perfect" balls (no friction), the post impact cb speed is 55.8% of its orginal speed (as described on page 2 of TP A.16). When accounting for friction the post impact cb speed is 53.6% (page 8). The document also shows that the ob does travel a little more than the cb when the impact friction is accounted for (except for thinner cut shots).
You show it being 33.5° for no friction and 33.2° with friction. Seems like there should be a larger difference between the two, but math is math.
According to the document, with zero friction on a 30° cut shot the difference between ob and cb post impact speeds is 6.1% (ob leaves the scene 6.1% faster than the cb leaves, both compared to the cb's orginal speed). At the same 30° shot angle, but this time accounting for friction, the ob leaves at a speed 6.5% faster than the cb. Both balls have lower post impact speeds when friction is factored in, but the ob speed is affected less. (With no friction you show ob speed as 61.9% and cb speed as 55.8%. Factoring in friction those numbers become 60.1% and 53.6%.)
Interesting stuff nonetheless.