I've tried to find out more about this myself - apart from anything else, it points to interesting questions about optics and trade-offs and so on. Sadly, like most camera talk, there's a lot of noise and not much data out there comparing similar lenses made for SLRs and rangefinders.
With this hefty caveat, the few bits of information that are out there point to a not-insignificant optical advantage for rf lenses.
Here is a an excellent article from Roger at lens rentals taking a detailed look at the question for a range of 50mm lenses, including the APO 'cron, Zeiss Otus and the Sigma Art. Unsurprisingly the APO 'cron is the best performing lens, but the two SLR lenses do very well, beating the everything else in the center. All three leica lenses (APO cron, cron and lux asph) are sharper than the SLR lenses at the edges. The Zeiss planar and VC nokton 1.1 do pretty well, and the sonnar has a generally poor showing. It's hard to take much from this, because the two SLR lenses are top-of-the-line, ultra corrected models (although the same is true at least for the leicas), but it does seem possible that for rf lenses (very broadly speaking) it is possible to build lenses with excellent sharpness across the frame. Even highly corrected (gigantic) lenses like the otus couldn't beat the cron' in the middle or the other leicas at the edges.
Another earlier test roger did
here uses the M9 for a regular MTF test, and gets similar results. Without the APO, the summilux is the sharpest lens at every aperture both centre and average. The canon 1.4 and 1.2 look very good wide open, but as they stop down the 'cron and lux both get better.
Roger notes that the m-mount lenses might be disadvantaged by the smaller pixel count of the m9 vs the nikon and canon cameras, in which case the rf lenses would be even better in comparison.
The last anecdote i'll share is of a guy who tested a range of medium format lenses. I can't remember the detail, but I believe it was using film and counting lines, so a real labour of love. From memory, the Mamiya 7 lenses were *by far* the sharpest lenses tested, although I think a number of the TLRs had very sharp optics, as did a few of the 'blads. This is, of course, rather unfair, given that the M7 is much more modern than the others (and worth noting too that the Fuji GW lenses made a rather poor showing).
What I have personally taken from all this rather unscientific evidence is that in sum it supports a general statement that it is easier to design optically superior (or, really, higher resolution) lenses for rangefinders, at least for normal lenses and below.
This doesn't mean that all rf lenses are superior - from Roger's tests the Zeiss and VC lenses both come out looking not so hot compared to the modern Otus and Sigma Art lenses. But it does mean that, when the stops are pulled out, as they are for the newest Leica lenses, the Otus and Arts and the Mamiya 7 lenses, rfers have a slight advantage that is most noticeable in higher corner resolution and improvement in performance as they stop down.
This is all in addition to the advantages vis-a-vis distortion at the ultra-wide levels, which is very noticeable when comparing even say an 24mm 2.8 canon lens to the 25mm vc m-mount lens.
