Didn’t look at the details but how did they arrive at the conclusion that Stockholm’s broadband dropped to 1/3rd the speed from 2018 to 2019, then shot back up

Didn’t look at the details but how did they arrive at the conclusion that Stockholm’s broadband dropped to 1/3rd the speed from 2018 to 2019, then shot back up in 2020? Did they test one connection on one day and that place had some technical difficulty? Seems unlikely that the broadband in a big city with dozens of service providers could vary by that much year on year.