Holy schmoly, Samunition... Absolutely PROVE IT! This I gotta see.
I wasn't going to weigh in with my 2 cents but I guess maybe I'll give my opinion. Penetration is key -- you can completely destroy surface tissue and create a nasty wound that might hurt like mad... but you're trying to STOP someone. Ask a hunter what STOPS game and they'll tell you... either (1) CNS hit, (2) destruction of vital internal organs, or (3) massive blood loss. None of these things are going to be accomplished with out proper penetration. It's even more important with handgun rounds because, despite what the gun mags will tell you about "stretch cavities" and "shock power" and all that bunk, they just aren't moving at the speed necessary to do the type of collateral damage that rifle rounds will do.
That being said... you're playing a delicate game of balancing bullet mass -- heavy enough to penetrate but light enough to build adequate velocity out of your specific barrel length. When I say "adequate" velocity, I'm talking about NOT TOO SLOW but I'm also talking about NOT TOO FAST for your chosen bullet design. A rapidly expanding hollowpoint pushed at too fast a velocity will expand too quickly, dumping all its energy and probably fragmenting before reaching any sufficient penetration. This is why some hollowpoint designs will penetrate LESS when driving to higher velocities.
TCW (the original poster) has a GP100 with a 4" bbl. This is a beefy gun with a medium length barrel. There is no need to go with super-light 115/125 gr bullets, IMHO. A round like the 158gr lead hollowpoint out of a 4" bbl should be moving fast enough to deform or mushroom and still penetrate deeply enough to stand an okay chance of reaching something important.
Frangibles, IMO, are a bunch of hooey. They are extreme examples of massive but shallow wounds. If your biggest concern is not going through walls and killing people in other rooms, perhaps you need to spend some more time and money practicing hitting your target.
I have scoured the web and books looking for proof of these "goat tests" and have never seen ANYTHING. One shot stops and Fackler and Marshall and Sanow and Fuller and all these other guys who are trying hard to apply half-a$$ed statistics or mathematical formulas to something that has a nearly infinite number of variables are doing nothing but helping to mislead and confuse the general public as well as sell books, ammo, and make themselves famous.
Here is the common sense bottom line, folks. Draw your own conclusions, but don't ignore the most simple of facts.
(1) Hit what you're shooting at. Placement is absolutely vital (no pun intended).
(2) Have a bullet that is going to go deep enough to touch something important. That means even if it had to penetrate an arm, deflect from a bone, pass through a leather coat, slip through some denim, crack a rib, and enter a torso sideways. How many bad guys are going to be standing naked and facing straight at you?
(3) Anything that comes out of a handgun is going to be under powered to consistently stop a human attacker.
Hope this helps.