re: My Melt is Real, but I Have Data

Not all of the data include Hatfield and Holtz, those are just notable examples raised in discussions of important Arkansas coaches in the later portion of the 20th century. Otherwise all data are in the SEC (and the West), and Petrino, for example, coached against Saban. The whole point of doing this over time is to account for short-term fluctuations. If that is not good enough for you, then we can re-estimate the model controlling for Saban the interpretation of results do not change (see below) at accepted cutoffs for significance. I am no fan of CBB, but all this is saying is that it will take a few more years of .5 or less to really make him historically awful for us in the SEC. The graphs on the first page were without statistical control and only designed to show trends for comparison, one of which was with other recent coaches in the West. I argue that is more important than the historical aspects of Holtz and Hatfield anyway.