Well, in answer to your question, when it comes to handguns, I think it is clear that whatever you can shoot most accurately is going to be most effective. I personally would not carry a pistol chambered in .380 Auto, 22 LR, .25, or .32 although any of those could potentially be effective with excellent shot placement. For a small concealed carry pistol I favor 9 mm Luger because of magazine capacity considerations and the fact that the recoil of stouter calibers can be somewhat more difficult to control in lightweight pistols with a small frame and reduced grip size. I prefer .40 S&W in medium sized pistols, especially those with all-metal frames that are a bit heavier. And I prefer .45 ACP in full-size pistols such as those that might be used for home defense, where magazine capacity limitations are not as severe. I have carried both a SIG Sauer P250 subcompact DAO in 9 mm and an M&P Shield in 9 mm inside the waistband. For OWB I tend to prefer something a bit larger like my SIG P229 in .40 S&W, but I have also carried a Glock 19 and compact SIG P320 in 9 mm if I wanted something a bit sleeker than a P229 that still had excellent capacity. For home self-defense I have a full-size SIG P320 in .45 ACP as a bedroom handgun.
Here is a somewhat rambling account of an individual who works in a Medical Examiner's office who has personally witnessed many autopsies carried out on victims of fatal gunshot wounds. Note that this was posted years ago, when 9 mm JHP ammunition performed less well than it does today:
http://www.mouseguns.com/deadmeat.htm
Now, I do not agree with everything this individual says, but there are several points on which his experience mirrors closely what I have seen. Realize that the experience of an emergency room physician, a trauma physician, or a trauma/general/vascular surgeon is not going to be the same as someone whose experience is limited to doing autopsies on fatal gunshot victims. Surgeons don't operate on dead people, and surgeons and ER providers typically do not observe autopsies on individuals who arrive DOA. On the other hand, a person doing autopsies in an ME's office will probably observe a lot fewer survivable gunshot wounds.
But here are some points upon which I agree with "deadmeat2":
The most devastating single shot wounds that I have seen resulting from close range encounters have resulted from 12 gauge shotguns, and that includes .223/5.56x45 and 7.62x39 high velocity rifle wounds. Based on my experience, I do not believe any other firearm that is commonly recommended for home self-defense is as likely to put an attacker down right now with a single, center mass hit as a 12 gauge shotgun will do. The high velocity rifle wounds I have seen have only been the result of FMJ ammunition. With these, the amount of tissue damage seems to vary dramatically depending on whether or not the projectile tumbled or fragmented within the body. I have seen a few high velocity rifle wounds to the abdomen which seemed to produce "remote" injuries like cracks in the capsule of the liver or spleen, or rupture of a loop or two of distended small bowel in which the secondary injuries appeared not to correspond to the direct path of the projectile. But I think this type of injury pattern is much less likely than some have claimed in the press and elsewhere, and in the particular instances I have witnessed none of the remote injuries would have been likely to result in immediate or early incapacitation.
As for handgun GSWs, it so happens that the most impressive results I have seen resulted from 357 Magnum revolver wounds. Back in the 1970s when I was in the trauma unit at Cook County Hospital, the majority of Chicago cops were still carrying revolvers, usually with 4" barrels loaded with 357 Magnum, and they brought the majority of police shootings to CCH. I would certainly consider 357 Magnum to be a potent self-defense round, but in truth I do not shoot it as well as I would like, and I prefer modern auto-loading pistols to revolvers.
Like deadmeat2, if I had to choose a handgun caliber for single hit effectiveness given identical shot placement, I tend to favor projectiles of higher momentum given sufficient penetration. For a given caliber, the "heavy for weight" projectiles tend to produce the highest momentum, but not always. I think projectiles that provide somewhere around 25 lb-ft/sec plus or minus 3-4 lb-ft/sec, are more likely to shatter bones or plow through them, and less likely to be deviated off course by other dense structures or intervening barriers than those with lesser momentum. Handgun calibers that commonly produce projectile momentum in this range include .40 S&W, 357 SIG, 357 Magnum, .45 ACP, and .44 Special. 38 Special standard pressure and 9 mm Luger typically fall a bit short, although with 9 mm the 147 grain projectiles can often at least exceed 20 lb-ft/sec, and some +P loads can. Calibers like 25 ACP, 32 ACP, and 380 Auto fall way short. Of course, there are pistol and revolver calibers that provide projectile momentum greater than 30 lb-ft/sec including 10 mm Auto, .41 Magnum, .44 Magnum, and .45 Long Colt, but with these over-penetration and reduced shootability concerns become an issue for some.
Here is an extract from that long, rambling discourse by deadmeat2 that also mirrors my experience:
"I can tell you that when one of the BGs comes in with multiple gunshot wounds it can be extremely difficult to determine the paths of each. We use steel probes to try to follow the path of each bullet in an attempt to determine the angle and trajectory of the wounds, and many times it's almost impossible. Unlike ballistic gelatin, the body is not translucent so the course of the bullet can't be seen. Also, unlike ballistic gelatin, which stays open allowing the damage to be analyzed, human tissue closes back up. Many times it comes down to making small scalpel slices along the wound path and trying to follow it that way. And from this I can safely say that I've never seen anything that approximates ballistic gelatin. Yes, there is damage along the course of the bullet, but usually it's due to the bullet itself, which is ripping tissue along the way and fragments of the jacket or core that are spalling off and creating their own trajectories incidental to the main path of the bullet. As I've said several time in other posts, I just don't believe that ballistic gelatin is a realistic representation of what actually happens, and I'm afraid that folks are placing their faith in a bullet that looks impressive in ballistic gelatin although the results are markedly different in the human body."
At least when it comes to handgun terminal ballistic effectiveness I don't think many people would disagree that the following factors are most important:
1. Shot placement (accuracy). 2. Adequate penetration. 3. Projectile expanded diameter (including reliability of expansion, for JHP ammunition). Those three factors are usually going to determine the likelihood that one hits a critical structure with a single shot.
After that, a lot of other considerations come into play. I do believe that projectile momentum is a factor but so too are things like magazine capacity and the ease with which the shooter can obtain accurate and rapid followup shots. And these will depend on the size and nature of the handgun used and the shooter's individual proficiency with the particular cartridge in that firearm.
