Keith, i'm more familiar with my friend's Magico S5 Mk1 speakers than the newer Mk2's, but from what I understand the S5 Mk1 is perfectly phase coherent in the x-over, and very, but not perfectly time coherent. Whilst Mike commented following his trips to CES and Axpona that the newer Magico S1 Mk2 & S5 Mk2 were "completely and totally seamless". I know Alon focused a lot on driver integration with the new 'Mk2' models. From what I understand, the newer drivers are better behaved across their frequency band which allowed Magico to simplify the x-over (which in turn resulted in a 1db gain in efficiency for the S5 Mk2).
Saying phase coherent is the same as saying proper execution. If you don't have that, the drivers don't sum properly, and your FR is a roller-coaster of a mess. And there's no chance of them being time coherent. The mid and tweet might get close after some time-delay in the crossover network, but the woofers are without a doubt lagging way behind on a step response.
I really don't mean to pick on you, just gotta be careful sometimes repeating things that you might not understand. And I'm not pointing things out because I like/dislike Magico, just that I like accurate statements. But if Magico gained 1db of sensitivity by simplifying their crossover, that would mean their prior crossover was an extremely poor design. It's quite rare to have 1db total of insertion loss from a passive crossover. The speaker may be 1db more sensitive, but it likely came from new drivers with lower moving mass, more motor force, or more compliant suspensions ... not from simplifying the crossover.