11-30-2015, 04:34 PM
I was reading over the reddit thread on this hire, and I saw a few optimistic takes (mainly this one) and got curious about the relative performances before and after Sark was fired. He was fired after the loss to Washington, so I assigned every game until that point to Sark, and every game after that to Helton. I then looked at USC's offensive YPP in that game, and compared that to the opponents average ypp on the season on defense.
The result? Sark's team was better. They averaged 1.58 ypp more on offense than the opponent typically gives up, Helton averages 0.07. Yes, Helton's team is barely better than an average opponent against their defenses. Interestingly his only loss, against ND, was his best game, averaging more than 2 ypp above ND's typical game. I did the same for defense, and Sark's team held opponents 0.83 ypp below their regular season average, and Helton's held them 0.02 yards below. So on both offense and defense Helton's team appears to be performing worse than Sark's.
I thought another way of looking at this would be to use FPI, since that's the only advanced metric I'm aware of that's readily available week-by-week. In week 6 (after the Washington game) USC was ranked #5 with an FPI of 22.8, currently they're #8 with an FPI 20.0. It could be that the week 6 FPI was held up by the high preseason expectations (it's not clear what kind of preseason voodoo ESPN does and how long the metric holds onto it), but you'd still expect better than just treading water if Helton was such an improvement.
So I don't see it. I think this USC team is worse than the one we played and beat in their own house. The only difference I see is turnover margin, USC was +4 with Sark (all of that coming against Arkansas St), and is +9 under Helton. Their big wins under Helton were all multi-turnover games, Utah with 4 turnovers and UCLA with 3. Turnovers matter, obviously, but it looks to me like Helton took Sark's team, made them a bit worse, and then got lucky with turnovers where Sark didn't. Not a good sign if you're a USC fan.
The result? Sark's team was better. They averaged 1.58 ypp more on offense than the opponent typically gives up, Helton averages 0.07. Yes, Helton's team is barely better than an average opponent against their defenses. Interestingly his only loss, against ND, was his best game, averaging more than 2 ypp above ND's typical game. I did the same for defense, and Sark's team held opponents 0.83 ypp below their regular season average, and Helton's held them 0.02 yards below. So on both offense and defense Helton's team appears to be performing worse than Sark's.
I thought another way of looking at this would be to use FPI, since that's the only advanced metric I'm aware of that's readily available week-by-week. In week 6 (after the Washington game) USC was ranked #5 with an FPI of 22.8, currently they're #8 with an FPI 20.0. It could be that the week 6 FPI was held up by the high preseason expectations (it's not clear what kind of preseason voodoo ESPN does and how long the metric holds onto it), but you'd still expect better than just treading water if Helton was such an improvement.
So I don't see it. I think this USC team is worse than the one we played and beat in their own house. The only difference I see is turnover margin, USC was +4 with Sark (all of that coming against Arkansas St), and is +9 under Helton. Their big wins under Helton were all multi-turnover games, Utah with 4 turnovers and UCLA with 3. Turnovers matter, obviously, but it looks to me like Helton took Sark's team, made them a bit worse, and then got lucky with turnovers where Sark didn't. Not a good sign if you're a USC fan.