Bill Connelly is once again doing the Lord’s work, breaking the blanket of boredom laid over the offseason with little CFB nuggets every weekday morning. If you’re obsessive like me, you can look over his entire catalog of work to date here. They’re all interesting, insightful and stat-based. But only one is worthy of a separate write-up, and that’s your Colorado Buffaloes.
Before we dive in, it is important to remember something. The words, the description of the team’s chances? That’s Bill Connelly, explaining his math. The actual prediction part is handled entirely by his S+P formula, which does a lot of things really well, but it has a few known blind spots.
Well, none of the information presented at the jump should be surprising. The Buffs were an anomaly in 2016, and 2017 regressed to a below average team that was still better than the pre-2016 teams. The lack of leadership combined with a lot of turnover made it hard to sustain success. Honestly, the most surprising stat was the comparison of MacIntyre’s first 5 year to McCartney’s:
If you’re the optimistic type, you could say that this trajectory remains on par with McCartney. But it takes some positive spin.
Win percentage, first 3 years: MacIntyre 0.250, McCartney 0.227
Win percentage, next 2 years: MacIntyre 0.577, McCartney 0.542
I did not expect that symmetry. Obviously the situations are different, but it’s something to keep in mind going forward.
The stats show that the 2017 offense was unremarkable, at best. The single best measure for the Buffs was their stuff rate, which can be attributed almost entirely to Phillip Lindsay’s indomitable will. The sack rate was horrendous, which can be attributed to a line that never gelled and a young QB who ran into his share of sacks.
Other than those stats, last year’s offense is surprising in its consistent mediocrity. CU was slightly below average across the board, which Connelly remarks, and is now losing a lot of production.
This is the point where national publications usually differ from the beat reporters in their predictions for the Buffs. Connelly actually does a great job mentioning that Travon McMillian is coming in to replace Lindsay, though he downplays his ability, and he also mentions Juwann Winfree and Laviska Shenault as “small-sample all-stars”. While we here in Buffland know that Winfree and Shenault, as well as Ento and KD Nixon, can actually exceed their predecessors productio. Many national writers will just see the experience and yardage leaving and despair for the Buffs without considering how talented the players who are replacing them.
Basically, Bill Connelly sees a mediocre offense losing a lot of production and its coordinator, and while he likes the potential, he’s not as sold as some of us.
A final note to point out: There is one small paragraph about the offensive line, which is about right. Lynott and Haigler are the elder statesmen of the line now, and they both have a ton of potential, but there are not a lot of sure things in the OL.
Colorado’s defensive radar is not fun to look at. We all knew that last year’s defense cost the team wins, but in case you didn’t, let Bill Connelly tell you. Here is a series of damning sentences about 2017:
Still, the identity shift confused me. Leavitt’s is an aggressive, efficiency-first defense, and it looked good on the Buffs. Even Eliot’s less poor units have been mostly bend-don’t-break oriented. Sure enough, everything about CU’s defense changed last year.
Success rate: 24th in 2016, 105th in 2017 (minus-81 spots)
IsoPPP (big-play prevention): 70th in 2016, 59th in 2017 (plus-11)
Points allowed per scoring opportunity: 16th in 2016, 64th in 2017 (minus-48)
Colorado grew far less efficient, didn’t improve enough at big-play prevention to pick up the slack, and — while taking on bend-don’t-break tendencies — grew far worse at red zone defense. Not the greatest combo.
Like Bill C says, the loss of all that talent and a fantastic coordinator like Leavitt would have led to regression regardless, but it didn’t have to lead to this. The rushing defense in 2017 was... not good. You can reasonably expect that to improve with the addition of Kwahn Drake as D-line coach and the infusion of young talent into that unit. Connelly mentions Nate Landman as a rising star, and I am similarly high on him. He can be a all-conference player. The biggest issue I take with this part of the review is his take on the secondary:
In the back, well, we’ll see. Worthington and Fisher are keepers at safety, but the cornerback position is uncertain after the loss of second-round draft pick Isaiah Oliver. From a foursome of sophomore Trey Udoffia (six passes defensed in 2017), junior Dante Wigley (six), redshirt freshman Chris Miller, and JUCO transfer Delrick Abrams Jr., a reliable duo must emerge.
I have the utmost faith that a Coach MacIntyre defensive backfield will always be solid, especially with equally competent coaches ShaDon Brown and Ashley Ambrose back there. Those cornerbacks are the main candidates to take a starting position, but Udoffia and Wigley are Pac-12 level players, and Abrams and Miller are at worst fantastic depth. The backfield is much more certain than a “let’s see” to me.
Let’s take a moment and thank Ross Els for what he has done to the special teams unit. That unit jumped from 103rd to 12th in S+P in one year. 91 SPOTS IN ONE YEAR.
This is as good of a place as any to go over his overall prediction for CU. The numbers see 4.3 wins, with toss-ups against CSU and ASU. If the Buffs win four games next year, there’s a good chance they’re looking for a new coach. I don’t see it happening.
There are a few things working against CU numbers-wise. One, their 5-year recruiting ranking is last in the conference, but that doesn’t adequately reflect the talent on campus, given that about half of the scholarship players are from these last two classes, which are much higher rated. Along these recruiting lines, a common refrain with the S+P formula is that junior college transfers throw it for a loop. Second, JuCo recruiting rankings are often depressed compared to high school recruiting rankings. The number one junior college player in the country is rarely, if ever, a five star. With that statistical bias, JuCo players count the same as freshmen, which means that a player like Davion Taylor is given the same weight as a player like Dimitri Stanley, even though Taylor will undoubtedly play a bigger role this year. Number three, in my opinion, the formula is vastly overrating teams like ASU and UCLA. Their talent is fantastic, but with a new coach (especially one like Herm Edwards), I can’t see them being much better than the Buffs.