Obviously there have been no "perfect" loudspeakers measured(so you got me there
), but there have been a few really good ones that didn't get recommended, as well as a few great ones that didn't get the top rating.
SVS Ultra: measurements were almost identical to Revel M106(side by side you could almost say it's a different sample of the same speaker). The biggest difference is maybe the distortion profile? They weren't recommended, while the M106 received the highest score.
Genelec 8341b: measured noticeably better than the M105/M106, but still received a lower score. BTW, I agree with Amir's criticism there. Running full range, even the 8351b runs out of steam well before I'm satisfied. That said, I also own the M105, and it's far more limited on that front, so who knows.
I'm not saying there weren't measurement differences in those examples, but looking at measurements alone, I think you'd have a hard time justifying a huge score difference between the SVS Ultra and Revel M106. Likewise, by measurements alone, I think you'd have a hard time justifying a higher score for the M105/M106 than the 8341a.
I don't think there's anything we can hear that we can't measure, but I do think we have room to go when it comes to interpreting what we can measure.