I'd assume that it's a timezone issue. The author probably didn't check to see if the labels on the x-axis matched his own timezone and just pointed out where h

I'd assume that it's a timezone issue. The author probably didn't check to see if the labels on the x-axis matched his own timezone and just pointed out where he knew the drop happened, instead of where it actually happened (on his graph).