It's not an anecdote. It's a basic understanding of traffic in Atlanta, the geographic limits of the ban, and all the Covid related fall out happening at the same time. It's the equivalent of arguing that days are 10% longer due to the scooter ban. It doesn't need critical examination to know it's wrong.

You should try reading the article (and other comments), the data collected was pre-covid.

> We report an effect each day beginning the day after the policy implementation, 10 August 2019 through 22 September 2019.

If that's the ban they studied than it's even more ridiculous because it only applied after 9pm.

The following March the scooters were effectively banned before returning later that Summer under a new permitting process.