I went back and looked at the numbers in detail once the pattern became apparent. In competitive games last year, the LSU defense typically saw a significant down tick in performance around the 45-50 snaps defended mark. The 70 snaps I cited earlier simply gave enough time after that inflection point to have a large impact on the final results. But it was actually part of a larger pattern on further examination.
Miami: In their first 50 offensive plays, the Canes had 212 yards (4.2 yards per play) and 3 points. In their last 19, they got 130 yards (6.9 YPP) and 2 TDs.
Florida: First 52 snaps went for 270 yds (5.2 YPP) and 14 points. Last 18 snaps – 105 yds (5.8 YPP) and 14 points. You can see the run defense falter late. 2nd half rush sequence: 3, 1, 2, 6, 2, 0, 10, 6, 9, (-1), 17, 12, 2(TD). After that UF is trying to kill the clock with a lead: 4, 5, 5, 1, 8, 3, (-1). LSU definitely struggled stopping the run when they needed to late and the 2 first downs on Florida's final drive were killers.
Georgia: First 47 plays = 156 yds (3.2 YPP) and 3 points . Their last 18 snaps yielded 166 yds (9.2 YPP) and 2 TDs. Good win for LSU but probably the biggest example of hitting the wall at 50 snaps.
Bama: The first 47 plays gained 323 yards (6.9 YPP) and 16 points. Their last 31 plays generated 253 yards (8.2 YPP) and 2 TDs.
MSU: First 50 snaps good for 190 yds (3.8 YPP) and 3 points. Last 11 yielded 70 yds (6.4 YPP) and no points.
Arkansas: Shouldn’t have been competitive but it was. Arky’s first 50 plays gained 145 yds (2.9 YPP) and 3 points. The last 8 plays resulted in 71 yards (8.9 YPP) and 2 TDs (doesn’t this number sound familiar?)
aggy: First 52 = 266 yds (5.1 YPP) and 24 points. Last 55 plays = 270 yds (4.9 YPP), 6 TDs and 2 FG. FGs are almost automatic in OT but LSU still had 7 blown chances to make a game winning stop on TD drives late. That’s a huge contrast to forcing 3 and out 5 times earlier.
UCF: First 48 = 192 yds (4.0 YPP). Last 13 plays = 58 yds (4.5 YPP), 1 TD. The variance was small in this game. UCF should have been dead in the water with a non-passing QB once they got down late. The fatigue shows up more clearly in the 2nd half run sequence: 7, 1, 3, 2, (-5), 5, 2, 8, 12, 11, 22, 2(TD). Bold denotes play count 50+
There is a pretty clear trend showing LSU’s defensive performance dropping off late in games – specifically after approximately 50 snaps. I suspect that this relates mostly to fatigue of their very heavy DL since the run defense was impacted significantly. It happened way too many times to be a coincidence. This is why I think it's kinda crazy for LSU to invite inflated play counts by going HUNH.
Using no huddle and quick throws is totally rational. There is no need to use tempo as the standard on offense. Using it selectively or threatening to do so prevents substitution. Using it constantly and targeting 75-80 snaps puts the defense in the 50+ danger zone for 35-40% of the game instead of 15-20% like most games last year. Like I said earlier I'm not sure they've really thought through all the implications of changing the offense.