I wasn't trying to make a special significance by looking back exactly four years -- I picked 2016 as a starting point because I felt that it gave enough contex

I wasn't trying to make a special significance by looking back exactly four years -- I picked 2016 as a starting point because I felt that it gave enough context to show that the big drop in December 2018 was a little blip in what is actually a huge uptrend.