I don't know how one can realistically compare cartridges since each shooting has so many different variables, but here are some things I've seen in cases I've worked.

From what I've seen, 9x19 hollowpoints will penetrate enough to be effective, and I've seen that .380 hollowpoints often won't. From their ballistic similarities, I would assume - without having seen it - that 9x18 hollowpoints would behave like .380 hollowpoints.

I've never seen a failure to adequately penetrate with .380 or 9x18 ball. 9x19 ball is damned menace and is practically guaranteed to overpenetrate a human being and still be dangerous on the other side. I've worked on a case in which 9x19 ball penetrated its intended target (a human head) and badly injured a non-intended person. I've worked on one case where a .380 ball round penetrated a limb of its intended target and went on to injure a non-intended target. But I've worked on a slew of cases in which .380 and 9x18 ball hit their intended targets (generally human torsos, sometimes heads) and did not overpenetrate.

The only time I'd consider carrying 9x19 ball would be if I were hiking in the boonies and this was going to be my defensive gun against big critters (certainly not my first choice, but remember the grizzly that ranger killed a few years back by draining his Beretta 92 into it). I carry 9x19 pistols frequently, and they're always loaded with hollowpoints (which vary depending on the gun - I carry what my chrono, test targets and reliability functioning shows works best in a given gun). I occasionally carry a tiny LCP .380 and a PPK-sized Hungarian R61 9x18, and they're always loaded with the fastest ball rounds from that gun.

A person should, of course, test rounds from his gun and think about the conditions under which he will be carrying that gun before deciding what round is best for him. What works for me and my guns in my situations, very well might not be the best thing for another person and his guns in his situations.