If you look at the number of upvotes and comments on any given post on voat.co, it's clear that very few people are actually using it. If deplatforming drove >50% of affected users there it would be troubling; my guess is that it's <5% though.