The top three teams in our pre-season Top 128 struggled to unconvincing victories over much lower-rated opponents on Saturday, giving other teams hope that there may be slots open in the inaugural College Football Playoff come December.
- Top-rated Florida State couldn't put away 37th-rated Oklahoma State, stumbling to a 37-31 victory in Arlington, Texas. FSU was favored by 18 1/2 while we had the Seminoles a 27-point favorite, and that's with giving Oklahoma State home field advantage. Florida State led the whole game and had a 17-0 lead early but the Cowboys kept coming back to keep things close, intercepting Jameis Winston twice and getting within 3 points in the 4th quarter. Winston had an incredible touchdown run and a long TD pass but was pressured effictively throughout.
- #2 Alabama had a tough time with West Virginia, who was 4-8 last year and ranked 57th in our pre-season rankings. The Tide held on to win 33-23 after entering a 26 1/2 point favorite (27 by our projections). Alabama led only 20-17 at the half but they were ahead from the 2nd quarter on. Blake Sims was at quarterback for Alabama the whole game and went 24 of 33 with no TDs and one interception. The game was played in Atlanta.
- Our #3 team, UCLA, had a terrible day offensively and needed its defense to top Virginia 28-20. They scored 21 points on two interception returns and a fumble return but led just 21-17 before Brett Hundley put the offense on the board with a 6-yard run. They won by 8 points after being 21 point favorites by the oddsmakers and 25 point faves in our book. Virginia was just 2-10 last year and were #63 in our pre-season top 128.
It's not possible to make sweeping generalizations from one game, but it's also wrong to completely ignore a game's results. Obviously based on these results the top teams were a bit overrated, if not overranked. In other words, they might still be the three best teams in the country but they aren't as far away from the rest of the pack as we had them.
But in terms of offense and defense, how much of their performance was due to their shortcomings, and how much might have been because their opponents were better than we thought?
For the Florida State game, the outcome probably means that Oklahoma State is better than we expected. They finished #9 in our Strength power rating last year and we downgraded them to #37, with both the offense and defense falling quite a bit.
The Seminole offense with Jameis Winston is probably still good, and they did score 30 points (7 points were from an interception return). But we downgraded Oklahoma State's defense from #18 to #64, and now it looks like that decline won't be nearly as sharp. The Cowboys offense fell from #9 to #22, and that might be overdone, too, though perhaps FSU's defense fell further than we estimated.
Is Alabama in trouble or West Virginia on the rebound? Two things were surprising here: that Alabama couldn't score at will on the Mountaineer defense, and that West Virginia had such success passing the ball. The former must be due to defensive improvement by West Virginia since we know Alabama's running game is great regardless of who is at quarterback. The latter has to be party on Alabama's defense, even though we did rank West Virginia's passing attack #9 in the nation for 2014.
UCLA's offense is probably the biggest stunner of the three games. The Bruins had the #11 adjusted scoring offense of 2013 and we projected it to be #4 in 2014. But they scored just one touchdown against the Cavaliers, who were the #78 defense last year. We did project a big upgrade to #42, and maybe this game shows that it's even bigger. But more likely, UCLA has some major kinks to work out of their offense.
In the end the assumption we make when modifying our ratings is that half of a game's effect is due to one team being better than expected, and the other being worse than expected. So look for the ratings of the top three teams to fall by about 3 points each, and the other teams' ratings to rise.
This won't topple the top three from their perches (unless Baylor goes haywire tonight vs. SMU). They were pretty far above the rest of the pack, and FSU very far above the other two, so it's going to take more than a close game or two to cause that kind of revision. But it will propel the losing teams upward. We see:
- #37 Oklahoma State will rise to approximately #28
- #57 West Virginia may jump to #46
- #63 Virginia should rise to roughly #55
That's not a bad deal, gaining 8 to 11 spots for losing a game. They might deserve even more upward movement if they were the cause for the top teams' underperformance. Or they might eventually drift back downward if the top teams continue to struggle. Lou Holtz and Mark May argued over how much you can judge a team by its first weekend performance, and it's a worthwhile argument. You don't know if a team's performance is representative, and you don't know how good their opponent is. So you make the conclusions you can, but keep them muted; it's only one game out of 12 after all, and the other team might prove to be a lot better or worse than you're assuming.