Seems disingenuous. Looking at the their data (p. 13), it does not look like a linear trend. There are two obvious change points; 1450 where rates dropped arou

Seems disingenuous. Looking at the their data (p. 13), it does not look like a linear trend. There are two obvious change points; 1450 where rates dropped around 5%, and then 1680 it dropped 5%. Without the extreme outliers in 1905 and 1930, the trend would be constant at 5% from 1680 onwards.