I dare say that a load with less penetration and more expansion/fragmentation would likely have performed better.
Far from being a bold statement, that's the general consensus in regard to .223/5.56x45mm specifically. One major problem in this case is that these loads seem to perform fine in ballistic gelatin but overpenetrate in actual targets. The other major problem is that even when they performed exactly as advertised, they failed to sufficiently alter the behavior of this particular bad guy, who continued fighting until a .40 S&W bullet disabled his arm and he was physically overpowered and restrained by LEOs.
So basically, while a number of the bullets performed fine in terms of general wounding, none exhibited any real stopping power aside from precisely where they hit. The .223 rounds may have caused his mortal wounds, but they didn't stop him or even slow him down by much in a timely manner. A .40 S&W round may have stopped him enough for the LEOs to move in, but it was a hit to his arm rather than his COM (where most of us try to aim, of course). Such is the nature of shootings.
This is why over time I've generally placed more and more emphasis on ammo capacity and being able to shoot fast (with reasonable accuracy), whatever the caliber you're using (even when using a high-powered rifle). One can stress shot placement until one is blue in the face, but the implication always seems to be that precise aim is what you need, when in reality it's awfully hard to place shots so precisely in a real gunfight. If the bad guy would hold perfectly still for a second and not shoot back, then sure, all I'd need to drop him immediately and permanently is a single shot, but things just don't work that way in the real world. Sometimes reality is easier, with the bad guys running away at the mere sight of a gun, and other times the bad guys fight back like crazy and never let you get a clean shot at them. As disappointing as this may seem, in most self-defense situations ultimately capacity and shooting speed (all aimed shots) are more relevant to effective shot placement than great accuracy.
Having said that, .32 ACP is a completely different animal from .223 Remington in terms of velocity and energy, so I'm not arguing cavitation effects in .32 or anything of that nature, just pointing out that too little upset and too much penetration can lead to a failure just like too much upset and too little penetration can.
Right, although overpenetration in handgun calibers is not quite as much of a waste as it is with small-diameter rifle calibers, it still is a waste of wounding potential. What I do is pick a level of penetration that I prefer and then find a load that expands just the right amount to achieve it. While I do err on the side of penetration more than most people, I still use JHP loads for defensive purposes whenever possible. With smaller calibers like .32 ACP and .380 ACP, I'd switch to FMJ and controlled expansion JHP (or wadcutter rounds if practical), respectively.