I’m not sure that the methodology here makes any sense. Much less that it should be put into a title that some people are just going to read and scroll past and accept uncritically.

I don’t know how accurate Statcounter is as a snapshot of actual usage. It might be good, but it might be bad. The slice of websites that actually use stat counter and how representative they are of global traffic, is probably a complex question.

Then, supposing it can be taken as representative, it’s treating market share like it’s absolute numbers, and like a percentage going down from one year to the next means a loss of users.

But it doesn’t necessarily mean that. It can mean that global users are growing, and that Firefox user share is not growing as fast as global growth, meaning that is remaining stagnant. Or it might lose a small portion of its user base while the total number of global users is increasing. A year to year percentage loss in market share doesn’t necessarily mean a loss of 12% of users.