I'm getting very worried by the current crop of tiny semi-auto pistols, mostly chambered in .380 ACP (some in even smaller rounds). I know a lot of people who rely on them as their primary means of self-defense - in other words, they don't carry any other gun or weapon. With the greatest respect to them, I fear they haven't understood the problem.
If the object is to simply get rounds into an attacker, without worrying about their effectiveness, then I submit even the lowly .22LR is going to work. (I wrote an article about the .22's utility for self-defense; it's not a bad choice, if you're willing to put in the work necessary to use it effectively.) However, with any smaller round, you have to understand that you're giving up a lot of energy and terminal effectiveness, compared to a larger round. Yes, I know all about bullet placement being critical (it is); but if your aim isn't precise (and, under typical conditions of a crime, your heart's going to be hammering away nineteen to the dozen, your adrenaline rush is going to be at an all-time high, and you're unlikely to be steady as a rock in your aim), you need to deliver enough 'punch' to get the job done.
A lot of smaller rounds can do that under perfect conditions, but again, you've got to be able to control them well enough to put them where they'll do the most good (from your point of view) or harm (from your assailant's). A small .380 auto pistol such as the Kel-tec P-3AT, Ruger LCP, or similar, is hard to control. It's so small that you can only get a couple of fingers on the grip; its sights are minuscule; and it's so light that even a small cartridge like the .380 generates significant recoil. I'm not denying it can work, but it'll take a lot of practice to be good with it - and, in the end, you're still dealing with an inadequate round. The good people at Buffalo Bore have this to say about the .380 ACP:
The 380 auto inhabits a valuable and useful place in our society, mostly because of the easily concealable, tiny pistols chambered for it. HOWEVER, because of the very limited size of the cartridge, it is plagued with limited power and therefore most of the existing ammo in 380 auto suffers from not being reliable as a man-stopper. We've studied and played with nearly all of the existing available 380 ammo and find it wanting as a reliable means of self defense, especially against a large, insane, drugged up/pain free, determined attacker.
Here's the problem:
The current 380 auto frangible ammo delivers a large amount of surface trauma, but lacks serious penetration. For example, if you shot me or another sane man in the face with modern frangible 380 ammo, it would blow off a big portion of my cheek and send a few teeth down my throat, I would undoubtedly fall to the ground in shock and pain, but I would be very much alive and functional if I could get past the shock and pain as that frangible bullet would have stopped some where inside my face, never making it to my brain. However, if you shot a drugged up maniac in the face with that same frangible 380 ammo and blew half his cheek off, he would keep right on coming because he is insane and is not thinking like you or I. Plus, he is likely pain free and fear free and wont know that half his cheek is missing and if he did know, he would not care. So whatever 380 ammo you shoot him in the face with, had better go through his face and blow his brain stem out the back of his head, because only a CNS (central nervous system) hit with a 380 is going to stop him. Likewise, a torso hit to the sternum needs to penetrate deep enough to blow all the way through his spine in order to shut him down spontaneously. If you fail to shut him down instantly, you and your loved ones are going to have to find a way to survive while you wait for him to bleed out and pass out.
I share and endorse Buffalo Bore's opinion of the round. (I carry a small .380 pistol myself, but only as a backup gun; and when I do, I stuff it full of Buffalo Bore's 100gr. hardcast lead flat nose ammunition, which to my mind offers the best penetration in that cartridge.)
The FBI has studied gunfights in the real world for many years. A few years ago, it refocused the firearms training it provides to its own agents, in response to those studies.
The FBI has quietly broken with its long-standing firearms training regimen, putting a new emphasis on close-quarters combat to reflect the overwhelming number of incidents in which suspects are confronting their targets at point-blank range.
The new training protocols were formally implemented last January after a review of nearly 200 shootings involving FBI agents during a 17-year period. The analysis found that 75% of the incidents involved suspects who were within 3 yards of agents when shots were exchanged.
The move represents a dramatic shift for the agency, which for more than three decades has relied on long-range marksmanship training. Apart from the new shooting regimen, agents also are being exposed to technology borrowed from Hollywood in which they can apply skills acquired on the shooting range to virtual scenarios involving the pursuit of armed suspects in schools, office buildings, apartment complexes and other potential targets.
. . .
"The thing that jumps out at you from the (shooting incident) research is that if we're not preparing agents to get off three to four rounds at a target between 0 and 3 yards, then we're not preparing them for what is likely to happen in the real world," says FBI training instructor Larry "Pogo" Akin, who helps supervise trainees on the live shooting range.
. . .
Until last January, the pistol-qualification course required agents to participate in quarterly exercises in which they fired 50 rounds, more than half of them from between 15 and 25 yards. The new course involves 60 rounds, with 40 of those fired from between 3 and 7 yards.
The new exercise also requires that agents draw their weapons from concealed positions, usually from holsters shielded by jackets or blazers, to mimic their traditional plainclothes dress in the field.
There's more at the link. Bold, underlined text is my emphasis. Highly recommended reading.
I submit that a light-caliber round, at such close range, is unlikely to achieve instant results unless it impacts the attacker's central nervous system. That being the case, I suggest that it behooves us to carry the most powerful handgun and cartridge we can control in rapid, aimed fire, in order to deliver the greatest possible impact shock, rather than a round we know to be of inadequate power.
That's why my preferred concealed carry pistol is chambered in either .45 ACP or 9mm. Parabellum, with the .45 my cartridge of choice. I know that modern 9mm. ammunition is almost as capable as .45, but the latter still makes a bigger hole and delivers more felt impact on the receiving end. The old Taylor Knockout Formula is derided these days as unscientific, but I'm an old African hand. In Africa, we know the Taylor formula works, because we've seen it at first hand, against both animal and human targets. I have no problem adding it to modern, more scientific test results, and letting it condition my cartridge selection based on that experience. (Yes, I know that in theory, applying the Taylor calculation proves that a sledgehammer is more effective against an elephant than a .470 Nitro Express round. That's using the formula inappropriately, and I won't waste my time defending it against such nonsensical comparisons.)
When I can't carry a full-power handgun in an accessible position (for example, in winter, when I'm wearing several layers that make getting at a hip holster rather difficult), I'll try to carry a suitable handgun in an overcoat pocket. This will usually be a revolver, in case I have to fire it through the coat - a semi-auto pistol's slide will get tangled up in the pocket material. The minimum I'll carry in my pocket is a .38 Special snubnose revolver, and if my pockets are big enough, I'll make that a .44 Special. In either case, I'll use effective ammunition (probably the Buffalo Bore full-wadcutter offerings in .38 or .44, to give the best penetration, and avoid clogging hollow-point ammunition with clothing fibers).
I still carry a .380 auto pistol . . . but only as a backup to a larger, more powerful handgun that's easier to control and likely to be more effective on target. I simply don't regard it as powerful enough to be my primary means of defense. Your mileage may vary; but I strongly suggest you try training with your carry gun according to the FBI protocols described above, emphasizing speed, accuracy and effectiveness. I suspect you'll find your little .380 to be a lot harder to use effectively, under those circumstances, than a larger, more hand-filling, more controllable weapon.
In particular, given that most small pistols of that sort are carried in pockets rather than in holsters, consider your speed of accessing the gun. If an attacker jumps you from 2-3 yards away, you won't have much time to react and draw your gun, particularly while you're backpedaling, ducking and maneuvering to avoid his attack. Something you can fire from inside your pocket, in a halitosis-range emergency, is likely to be more useful than something you have to take out first!
Unfortunately, it's difficult to train that way, given the expense of constantly having to buy new clothes. Nevertheless, thrift stores such as Goodwill offer cheap, used trousers and jackets. I suggest you buy a few of them, take them to the range, and expend them on pocket-shooting training with your chosen carry gun. You might be surprised at the results - and you'll learn a lot, including why we shouldn't use high-flash, high-pressure ammunition in pockets. The resulting fires - not to mention muzzle-blast and cylinder-gap burns - can be painful. Also, note the effects of muzzle blast on fabrics that can melt, such as nylon. When they do so on your skin, it's not funny!