The FR curve past 1khz and even more so 5khz or so is heavily dependent on the interaction between the headphones and your own anatomy (or the test rig's design) - just like the response at lower frequencies can be more or less heavily influenced by seal. I believe that this image is an interesting illustration of that phenomenon (I believe that it comes from that paper) :
https://www.aes.org/e-lib/browse.cfm?elib=16877)
:
View attachment 122152
Some of these peaks and dips might be a constant across all listeners, but some others rather are, to some degree, the product of that interaction, and above 5kz or so the delta can be quite huge.
Even the HD650 may exhibit some peaks in some measurements :
https://www.head-fi.org/threads/sennheiser-hd650-measurements-brüel-kjær-5128.951182/
That being said, if I look at the deviation in the above picture I'd probably lean towards thinking that most Focals show some oddities in their response in the 1khz-5khz range that's on the edge of what variation between individuals would suggest is acceptable (I'm being kind here), and some of the peaks past 5khz that we've seen from them are constant across various test rigs (the 6khz peak or so for example), so perhaps an intrinsic quality of the headphones rather than something that has to do with how they interact with the user's anatomy. And that conversely, given how quite impressively smooth the HD650 measures across various measurements, I'd be inclined to think that they're likely to be effectively smoother across a wider range of listeners. But it's difficult to know
for certain how they'll behave on someone's head in particular.
If I may add a personal anecdote, with the exception of their 4300-4400hz peak that shows up in all measurements of the HD560S, the HD560S and HD650 both measure reasonably smoothly across various measurements in the trebles range, but when I listen to sweeps on my own head, it's very easy for me to distinguish at which frequencies the HD560S produces peaks and dips while I'm having a much harder time doing so on the HD650, suggesting, perhaps, that the magnitude and severity of the peaks and dips of the HD560S on my own head is quite a lot more important than the HD650 (at some frequencies the HD560S is actually a little bit painful for me), despite decent measurements in both cases.