Re: Phil Steele Averages Offensive & Defensive Production Compared to Opponent's Numb
Yardage is a nice talking point, but there's not a direct correlation between it and the effectiveness of a unit on a team. I'll take a team that gives up 600 yards a game as long as all those yards are between the 20s and I'll take an offense that only gains 200 yards a game if they're always scoring more than the opposition.
Tennessee is ranked above us in offensive performance, but against common opposition (Mississippi State, Missouri, and Georgia), we outgained UT's offense 486 yards to 476 yards. We did it on 10 less offensive plays per game, on average, so our yards per play is more than a yard greater (7.14 to 6.09) than theirs.
So how did UT get ranked higher than us? It was probably the 633 yards against Akron (giving up 450 yards per game, on average against their daunting MAC schedule) and the 713 yards against Troy (Sun Belt school, giving up 434 yards per game) that skewed their average enough to outrank us. Of course, you can't beat on the viles for running up the score, because they were only up against Akron by two TDs midway through the 4th quarter and needed an insurance TD pass from Tyler Bray with a little over 6 minutes left before going into protect the lead mode. Against Troy, they were TIED with the Trojans until the winning score with about a minute and a half left in the game. Troy subsequently drove to the Tennessee 40 and just couldn't connect on two pass attempts to take the game to OT.
So, UT is ranked above us because they had to fight tooth and nail to beat a couple of schools that are barely above FCS level and should have been playing fourth stringers by the third quarter. These statistics are about as useless as Roy Adams' liver. Is there anyone that truly believes if UT and Alabama played the same defense with all other things equal, that UT's offense would gain 50+ yards more than Alabama's offense? That's what these numbers imply and using the numbers against common opponents from last season, that's been shown to be empirically incorrect.