Occasionally I re-read the old FBI Handgun Wounding Factors and Effectiveness paper from the late '80s. There was one line in particular that said that hollow point bullets only expand 60-70% of the time at best.
The paper says this:
"Damage to the hollow point by hitting bone, glass, or other intervening
obstacles can prevent expansion. Clothing fibers can wrap the nose of the bullet in a cocoon like manner
and prevent expansion. Insufficient impact velocity caused by short barrels and/or longer range will
prevent expansion, as will simple manufacturing variations. Expansion must never be the basis for bullet
selection, but considered a bonus when, and if, it occurs. Bullet selection should be determined based
on penetration first, and the unexpanded diameter of the bullet second, as that is all the shooter can
reliably expect. "
It made me think, how reliable are hollow points in expansion today? Is there a percentage number that can be ascribed? Considering that the switch back to 9mm was facilitated by the advent of well-designed expanding loads, there must have been a huge increase in the reliability of expansion. Still, everything can fail.