I strongly doubt that Perl 5 would be any more popular in 2018 with or without P6.

It's an unprovable concept, sure, but P6 was intended as a replacement from the start. That's why it conforms to the version numbering scheme of almost every other programming language ever.

P6's "sister languages" concept fails to account for the difficulty of explaining that the thing that looks like a version number isn't, as evidenced by the fact that not even Perl folks understand it.