I think there is another post about this "study" somewhere on the forum. The study was really flawed in terms of number of cars, location of cars, miles driven, etc. with no attempt to normalize the data. I think the one 2015 LEAF that was included in the study was from a hot climate and probably was stored at full charge for extended time. A lump of coal might be more appropriate for this study than even a box of salt! I will post comparison of 2011 (replacement battery), 2015, and 2019 Plus after I have a full year of data on the 2019.
Edited to add: I could not find the reference to 2012 when I went back to look at the study again so I deleted references to it from this post.