I'd be extremely careful about using Firefox's analytics data. I can't find Firefox's telemetry coverage numbers (they gather the info, so it's probably out the

I'd be extremely careful about using Firefox's analytics data. I can't find Firefox's telemetry coverage numbers (they gather the info, so it's probably out there somewhere), but it's not at all unreasonable to suspect that a lot of the "power user"-type interactions are correlated with an unwillingness to enable telemetry. Couple that with the fact that Firefox, as a relatively niche browser these days, has a selection bias for more techy-savvy users, and you have a recipe for extremely misleading statistics.