<
>

Connelly: Using SP+ spreads to help determine CFB win totals

Notre Dame is a big favorite in nine of its matchups. Quinn Harris/Getty Images

You know it's been a weird year when you see how much even Vegas' practices have been affected. College football win totals were released in April (a bit earlier than normal) and then were rendered entirely moot by July. And when the season officially began to look like it was going to pick up again, finding win totals to wager on became awfully hard.

You can kind of understand why. While the meager college football schedule to date -- one FCS game in Week 0, nine FBS/FCS games in Week 1 -- has gone off without a hitch, the big waves are coming. TCU-SMU has already been postponed by COVID-19 to an undefined future date, while Tulsa-Oklahoma State has been pushed back a week. Bigger schools, with bigger coronavirus case totals, are approaching, and we're watching the water, waiting to see if this is a ripple or a tidal wave.

Aside from being a good betting exercise for a number like SP+ or FPI, win totals also help us set pretty good expectations for given teams, especially in a season's early going. They can help to keep us from overreacting.

For our own purposes, then, let's take a look at the college football landscape using the win totals construct. And instead of using projected SP+ win totals, as I am wont to do and as I shared in last week's 2020 SP+ projections post, let's flesh out an idea that I mentioned later in that post: marginal wins. We're going to derive win totals with these criteria: If you're favored by more than 7.5 points per SP+ (approximately one possession), you're given a win. If you're an underdog by more than 7.5, it's a loss. Anything in between is a tossup, for which you get 0.5 wins.

(Two notes: First, as mentioned elsewhere, due to small or nonexistent crowds at games, I'm starting the season using only a one-point home-field advantage for projections instead of the normal 2.5. This will get adjusted as we begin to understand the effects of these crowd changes. Second, last week's results are, for obvious reasons, in the sure win and sure loss piles.)

Here's how the fall FBS field lays out with this method:


10.5 wins: Appalachian State, Clemson

Your most likely unbeaten Power 5 and Group of 5 teams combined for just one regular-season loss last year. Both play 11 games, and both are given just one relative tossup game per SP+: Appalachian State hosts a strong Louisiana team on Oct. 7, while Clemson plays at a top-10 Notre Dame team on Nov. 7. With wins in those games, it will take a pretty significant upset -- like, it must be said, App State's loss to Georgia Southern last year -- to dent unbeaten records.


10 wins: Notre Dame

The Irish picked a really good year to "join" a conference. Brian Kelly's team starts the year ninth in SP+ and is a projected favorite of at least 9.8 points in nine of 11 games. The exceptions: Clemson's visit and a Nov. 27 trip to North Carolina.


9.5 wins: Alabama, Oklahoma, WKU

Both Alabama and Oklahoma are, like Clemson, healthy favorites in all but one game. For Alabama, it's Georgia's Oct. 17 visit (they're 5.2-point favorites), and for Oklahoma it's the Red River Rivalry game against Texas on Oct. 10 (5.6 points).


9 wins: Memphis, Louisiana, North Carolina

Memphis labored a bit more with Arkansas State than SP+ expected but looks to be in pretty good shape moving forward, facing tossups against only UCF and Cincinnati and a near-tossup with SMU. Louisiana is an 8-point projected underdog at Iowa State this weekend but faces mostly likely wins after that. The most interesting team here: Mack Brown's Tar Heels. If they get past mid-October tossups against Virginia Tech and Florida State, they could be awfully high in the rankings when Notre Dame comes to town.


8.5 wins: Cincinnati, Florida, Georgia

Florida and Georgia are both projected top-five teams per SP+, but 10-game SEC schedules will leave you with quite a bit of work to do. Georgia heads to Alabama, hosts Auburn and faces Florida in Jacksonville; Florida also hosts LSU and travels to Texas A&M. This could be a fascinating CFP race, as there really might not be more than two or three Power 5 teams finishing with zero or one losses. We could theoretically see a three-loss No. 4 seed.


8 wins: Texas, UAB, UCF

After being extremely bearish on Texas last season, SP+ agrees with the AP poll in ranking the Horns 14th to start the season. But they're not completely distanced from a batch of second-tier Big 12 contenders -- Iowa State, Oklahoma State, Baylor -- and that gives them quite a few tossups through which to navigate.


7.5 wins: Louisiana Tech, LSU, Miami, SMU, Troy, Virginia Tech

Figuring out how to set expectations for LSU this year is nearly impossible. The defending champs suffered enough departures to render them 13th in the updated SP+ projections, and since the AP poll came out before a few opt-outs, it's easy to call them overrated heading into the year. But while depth and experience are concerns, they still have more upside than anyone in the SEC besides Bama. Both 9-1 and 6-4 are on the table, which I guess makes 7.5 wins a good starting point.


7 wins: Army, Auburn, Baylor, BYU, Iowa State, Oklahoma State, Southern Miss, Texas A&M

Southern Miss' surprisingly feckless loss to South Alabama meant the Golden Eagles will be playing with an interim coach the rest of the year. It also dropped them from the eight-win pile to seven. They still have lots of potential wins if they rebound, though. Meanwhile, why is BYU this low on the list after how good the Cougars looked on Monday night? Easy: The Cougars have only eight games scheduled. They have potential challenges against Houston and WKU remaining, but their odds of going 7-1 or 8-0 are solid.


6.5 wins: Florida State, Georgia Southern, Louisville, Pitt

Another team with confusing expectations is Florida State. The Seminoles didn't top 6.5 wins in 12-game schedules in either of the past two seasons, but five of their 11 games are projected within one score. If the offense adapts well to new coach Mike Norvell's system, an 8-3 or 9-2 record is an absolute possibility. But it's hard to feel too confident in that.


6 wins: Arkansas State, Navy, Virginia

Bad news: Navy looked absolutely terrible against BYU on Monday night after what head coach Ken Niumatalolo said was a fall camp with pretty light contact. That could portend a full season of gloom.

Good news: The Midshipmen got a full offseason's worth of contact against that physical Cougars squad, and if they get up to speed, the schedule is pretty navigable.


5.5 wins: Tulane

Two likely wins, two likely losses and seven relative tossups for Willie Fritz's Green Wave this year.


5 wins: Marshall, North Texas, TCU, Tennessee

Tennessee is in a no-win situation this season. The Vols appeared in the AP Top 25, which caused plenty of scoffing -- oh look, we're overrating the Vols yet again -- and they're likely to finish pretty close to .500 against a brutal, SEC-only schedule. (Half their games are against teams in the SP+ top 11.) But going 5-5 or so would basically prove the Vols' Top 25 bona fides.


4.5 wins: Charlotte, FAU, Georgia State, Liberty

FAU was the class of Conference USA last year but has to replace an absolute ton of talent. That said, the Owls are this far down the list in part because they only have nine games scheduled. They're only a couple of conference wins away from a perfectly decent C-USA showing.


4 wins: FIU, Houston, Kentucky, MTSU, USF

I'm really intrigued by Kentucky this year. The defense should be solid, and the offense should still have a fun, physical identity without Lynn Bowden Jr. But the schedule is absolutely brutal, and even if the Wildcats field their best team since 1977, it'll be a scrap to get to .500.


3.5 wins: Coastal Carolina, Georgia Tech, Kansas State, Missouri, NC State, Ole Miss, Texas Tech, Tulsa, UTSA

Georgia Tech's original nonconference slate featured games against UCF and Georgia, plus a season opener against Clemson in cross-division play. The Yellow Jackets might be the only Power 5 team besides Notre Dame that saw their schedule get easier with the increase in conference games.


3 wins: Duke, Mississippi State, South Alabama, South Carolina, Temple, WVU

Mississippi State coach Mike Leach tends to get teams up to speed pretty quickly with his "takes three days to install" offense, and he inherits some defensive talent above what he's used to. I have no doubt MSU will be competitive. But the road to .500 is a pretty long one all the same.


2.5 wins: Boston College, ECU, Rice, Syracuse, ULM, Wake Forest

Dave Clawson has done a fantastic job at Wake Forest, but the Deacs just lose so much from last year's squad. Tough row to hoe this year.


2 wins: Texas State

Jake Spavital's second Texas State team looked awfully competitive in Saturday's tight loss to SMU. The Bobcats could be prime overachiever candidates.


1.5 wins: UTEP

I instinctively root for underdogs, but I couldn't help but root for UTEP against Stephen F. Austin on Saturday -- a win assured that the Miners wouldn't go winless this fall.


0.5 wins: Air Force, Kansas

Hiring Les Miles is a good way to assure some randomly volatile results, and even last year's 3-9 campaign featured a blowout of Boston College and a near-upset of Texas. But "Hey, maybe something weird happens!" is the only way you can talk yourself into KU winning more than one or two games. SP+ doesn't see them as a slam-dunk against Coastal Carolina, either.

(Ignore Air Force. The Falcons play only two games.)


0 wins: Arkansas, Vanderbilt

Arkansas is a projected underdog of at least 8.4 points in every game, Vanderbilt 16.7. Odds favor the Hogs picking off at least one upset, but oof.


To close, let's flip this around and look at projected loss totals. That should give us a pretty good feel for the potential CFP race. Power 5 teams are in bold.

0.5 losses: Alabama, Appalachian State, Clemson, Oklahoma

1 loss: BYU, Memphis, Notre Dame, UCF

1.5 losses: Air Force (in two games), Cincinnati, Florida, Georgia, WKU

2 losses: Louisiana, North Carolina, Texas

There is a scenario on the table in which a rampant BYU team works its way into contention and maybe sneaks into the CFP top four ... thanks to fellow indie Notre Dame losing a conference title game. I wouldn't bet on that even if it were an option, but I know what I'm rooting most heavily for this fall.