I'm hoping, foolishly, that people will actually read the explanatory text of this statistical study instead of merely looking at the charts and using them as evidence to "prove" their pre-determined conclusions. The text explains the parameters and limitations of the study - caveats that are crucial to understanding the meaning of the charts.
The first caveat to note is that that the study only covers injuries during the 2012-2015 period. Many of these stadiums may have switched turf types since then, perhaps even more than once - the Pats switched out their turf TWICE in 2017.
The second caveat to note is that the study apparently doesn't distinguish between contact and non-contact injuries. TJ Ward throwing himself at Gronk's knees is equal to a guy tearing an ACL while making a sharp cut with no one within a couple yards of him.
The third caveat to note is that these are IN-GAME INJURIES that have occurred during the regular season. The study does NOT include injuries that have occurred on practice fields during the season OR during OTAs and training camps, activities that are often held on fields outside the stadium or even off-site in another state - fields that may have entirely different surfaces.
Lastly, for the sake of perspective, there's the small data set and the confidence intervals to consider, plus the fact that the difference between the "best" and the "worst" turf type is less than 4 injuries per 1,000 "athlete-exposures".
So, FWIW ...
Turf Type and NFL Injuries: Part I | Football Outsiders
Turf Type and NFL Injuries: Part II | Football Outsiders