I'm not sure the article's method of measuring influence is valid. - It assumes Twitter use is constant and its not losing users (I've deactivated my account,

I'm not sure the article's method of measuring influence is valid. - It assumes Twitter use is constant and its not losing users (I've deactivated my account, even though I hardly used it). - It assumes Twitter hasn't improved bot detection since before - It assumes that the issue is with Rogan rather than the guest (maybe it could be argued that Rogan is getting 'uninteresting' guests because of a waning influence)