Just to reiterate on the IBIS front:
My own personal experience (focal lengths from 17mm to 55mm, all primes, some full-alt) is that IBIS works quite well. Let me illustrate with a relative comparison from general use, not controlled testing-- but this includes shooting on assignment, so it follows an internal belief of what I can actually, reliably pull off.
My benchmark is the 5D2. If I'm careful, I know I can shoot at 1/FL and get pretty sharp results most of the time. If the shot is absolutely critical, I try to get up to 1/(2 x FL). (Note: this doesn't mean critical-- aka tripod- sharpness, but good enough sharpness to deliver to, say, a big circulation glossy mag for a single-photo, 2-page spread.)
When I added my Fuji X-E1, I found that 1/FL wasn't quite as reliable. It was doable, but I had to work much more slowly. So my standard was to try to get to 1/(2 x FL) for attaining "pretty sharp results most of the time." For those critical things I tried for more. Keep in mind that's AFTER doing the APS-C to FF equiv. to get focal length. So with the 18mm native lens mounted, 1/60 was my mental baseline, and 1/30 was very much a push scenario where I'd want to have multiple shots to make sure I got one that really worked for me. I suspect this had a lot to do with: i) how light the system was, and ii) a tendency to either shoot using the LCD or not quite being able to lock the camera into my body as well when shooting through the EVF.
On the Sony (A7rII), I very comfortably can shoot at 1/(1/2 x FL) or 1/(1/3 x FL) for my baseline "pretty sharp results most of the time." Not only this, but I think I get better, sharper, pixel-level results at these shutter speeds than I do with 1/FL on my 5D2. When you consider what pixel-level sharpness translates to when you are comparing 22mp to 42mp-- the A7rII wipes the floor with my 5D2 from a stabilization standpoint. Being able to reliably shoot my 35L at 1/10 or 1/15 is something I would never, ever have been comfortable doing with my 5D2. (Note: as mentioned earlier, this introduces a greater risk of subject motion becoming an issue.)
So, if you're like me, going from no stabilization at all to IBIS-- even a couple stops is pretty remarkable for what it opens up. The other day I was shooting 0.5s exposures on a 17mm lens, while perched in an awkward position over a river on rocks (to blur the water motion)... and if there was softness I think it was from the diffraction of shooting at f/22! I might have been able to do that on my 5D2 after 1 of 10 shots compared to the pixel level sharpness I get on the 42mp of the A7rII in 1 of 2 shots in that situation.
I very much doubt the IBIS is very effective at >100-200mm focal lengths, particularly compared to lens-based stabilization designed for a single lens.
IBIS is just another nice thing to have, but it's definitely not magic. And if it can buy you a couple stops of exposure, aperture, or lower ISO... there are times that makes all the difference.