
College Football Week 3 Predictions: Picking Top 25 Games Against the Spread
Week 3 could not have come fast enough.
I have never picked games well in Week 2, so I was happy to walk away with a clean, symmetric, 9-9-1 record against the spread and get on with the rest of the schedule.
Why is Week 2 so difficult? Because it's national overreaction week. Sometimes I fall victim to this and overrate a team's first game (see: Utah and Utah State), while other times I overcompensate and actively ignore Week 1 (see: Georgia Tech and USC).
Either way, things get easier once teams play more games and data becomes more reliable. Week 3 is still pretty early, but most teams have played 100 percent more games than they had before last week, which means we have 100 percent more game data.
The Week 3 slate features four games between ranked teams, and all eight of those teams are ranked in the top 20. It also features numerous conference openers and the debut of midweek ACC madness.
That should get me back into rhythm.
As always, feel free to chime in below with your opinion or with questions about the picks. I'll explain my rationale beneath each game, but of course we can always dive deeper. Just remember to keep it civil because, like, what's the point in yelling on a comment board?
Remember: We are all on the same team here. The line is our only enemy.
Note: Top 25 rankings refer to the Associated Press Poll. All Week 3 spreads via Odds Shark. All historical spread info via TeamRankings.com. All betting percentages (which side has seen more action) via Sports Insights.
No. 11 Clemson at Louisville
1 of 11
The Line: Clemson (-6.5)
Louisville hosts Clemson after dropping its first two games of the year. Its back is planted firmly against the wall, it knows it needs a win to save the season and it's playing on a chaos-friendly weeknight—all of which made it tempting to back the underdog.
I expect the Cardinals will come out with a jolt, perhaps even jumping to a lead the way they did against Florida State on a Thursday Night last season. Even after last week's loss to Houston, this crowd will show up, get behind its team and make an impact.
But just like last year's Seminoles, Clemson has too many weapons to suppress for 60 minutes. Florida State mounted a comeback against what appears to be a much better Louisville team, and Clemson, if it falls behind, has the talent to follow suit.
Louisville quarterback Lamar Jackson is explosive, but he's also a mistake-prone true freshman and playing without his best receiver, James Quick. The gap between him and Clemson quarterback Deshaun Watson, who again ranks near the top of the national QB rating leaderboard, is enough to make me side with the road dog.
The Pick: Clemson (-6.5)
No. 9 Florida State at Boston College
2 of 11
The Line: Florida State (-8.5)
Boston College entered the year with question marks, and those question marks remain after back-to-back wins over FCS teams.
Florida State entered with question marks as well, and although there were shaky moments in wins over Texas State and South Florida, it at least diagnosed itself against FBS competition.
The Golden Eagles ranked No. 127 in the country on Phil Steele's returning experience chart, and no matter your opinion of Steele's formula, that's a troubling number. SB Nation's Bill Connelly proposed a better way of measuring experience, but even then Boston College ranked in the nation's bottom six.
The jump in competition from Maine and Howard to Florida State should be too much for a young team with little continuity. I hate picking against Boston College head coach Steve Addazio, who has quietly done one of the best jobs in the country the past few seasons, but his roster is overwhelmed by FSU's talent, and the Noles should come out hot against a team they needed all 60 minutes to beat last season.
The Pick: Florida State (-8.5)
Connecticut at No. 22 Missouri
3 of 11
The Line: Missouri (-21.5)
Both teams enter 2-0, but neither has played a single impressive game.
Missouri beat Southeast Missouri by 31 and Arkansas State by seven, while Connecticut beat Villanova and Army by five apiece.
The Tigers do not crack my personal top 25, but at the same time…this is Connecticut. The Huskies are even worse than Arkansas State, ranking 28 spots lower on Football Outsiders' F/+ ratings, and Missouri gets them at home on the heels of a poor performance, when logic suggests it should come out extra-focused.
Brian Fremeau's FEI projections have Missouri winning this game by 34—a plus-12.5 margin over the spread as a home favorite. The two games this season that most closely resembles, Purdue at Marshall (-7.5) and Texas at Notre Dame (-10) in Week 1, both came home as covers for the favorite. Fremeau projected Marshall to win by 20 and Notre Dame to win by 22, and they won by 10 and 35, respectively.
The sample size is small, but I'll roll with that.
The Pick: Missouri (-21.5)
No. 23 Northwestern at Duke
4 of 11
The Line: Duke (-3)
Northwestern catching three points at Duke is the "Line I Have to Pick For This Article But Would Never Put Actual Money On" of the week.
Do yourself a favor and stay away.
Since I have to pick it, my base urge is to always go contrarian. The public sees that Font-eight number next to Northwestern's name, then doesn't see a Font-eight number next to Duke's name, then thinks it's pulling a fast one on the sportsbook by taking a ranked team plus points, as if there must be some sort of clerical error.
But there's a reason oddsmakers have Duke as a three-point favorite: These teams are roughly even, and three points is a baseline home-field advantage. Fremeau's FEI projections have Duke winning 26-20 and give it a 67 percent chance to win outright.
Sports Insights lists a moneyline option, which allows bettors to forgo the spread and pick which team will win outright, adding juicier odds for the underdog and worse odds for the favorite—and 97 percent of bets are on Northwestern. Ninety-seven percent! Have 97 percent of bettors won anything? Ever? In the history of handicapping?
Maybe this line doesn't stink so much after all…
The Pick: Duke (-3)
No. 14 Georgia Tech at No. 8 Notre Dame
5 of 11
The Line: Georgia Tech (-2.5)
Notre Dame has lost quarterback Malik Zaire, running backs Tarean Folston and Greg Bryant, tight end Durham Smythe and defensive linemen Jarron Jones and Ishaq Williams to season-ending injuries.
Cue up the "Nobody Believed In Us!" theory.
The Irish have recruited for depth and have the bodies, in the short term, to survive most of those losses. Attrition has shifted the line toward Georgia Tech and made Notre Dame a short home underdog, but its biggest stars remain intact.
Inside linebacker Jaylon Smith, the No. 1 overall draft prospect on Matt Miller's B/R big board, and defensive tackle Sheldon Day, whom Miller has called the best run-stopping tackle in the draft class, are made as if by design to stop Georgia Tech's triple option. By virtue of playing Navy every season, they also have experience defending it.
The offense still boasts a pair of All-America candidates in left tackle Ronnie Stanley and wide receiver Will Fuller. DeShone Kizer threw a handsome clutch touchdown to Fuller to help the Irish escape at Virginia, and although Georgia Tech will have the benefit of game-planning around him, the redshirt freshman quarterback has the physical tools, dominant offensive line and deep cast of receivers to build on that early success.
"Certainly DeShone Kizer doesn't have the experience that Malik (Zaire) has, but we can run our offense through DeShone," head coach Brian Kelly told reporters after Saturday's win. "He has a lot of weapons around him, and we saw that tonight."
I like that Notre Dame is the underdog, and I like that three in four spread bets have been for Georgia Tech. My contrarian alarm is sounding, and it's telling me to take the higher-ranked team, which has already played two power-conference opponents, over the lower-ranked team, which has only played Alcorn State and Tulane.
The Pick: Notre Dame (+2.5)
No. 18 Auburn at No. 13 LSU
6 of 11
The Line: LSU (-7.5)
It takes a stubborn jerk to keep pimping Auburn—his preseason pick to win the SEC and make the College Football Playoff—after watching the Tigers need overtime against Jacksonville State.
But I guess I've been called worse.
Prior to the season, I would have happily laid three points with Auburn in Death Valley, and I'm not sure one bad game from AU and one good (decent?) game from LSU should change my mind on that.
Now Auburn is getting 7.5 points against a team that lost defensive coordinator John Chavis, doesn't trust its quarterback to make plays and nearly pooched a game it dominated against Mississippi State.
Maybe I'm seeing Auburn through the lens of confirmation bias, and LSU is really the better team. Even so, I don't see Les Miles winning this game by more than a touchdown. Miles has a unique way of turning blowouts into close games late, and Auburn did win 41-7 when these teams met last season.
That still has to mean something, right?
The Pick: Auburn (+7.5)
South Carolina at No. 7 Georgia
7 of 11
The Line: Georgia (-17)
Georgia beat Vanderbilt with bend-but-don't break defense, allowing 14 points on nine scoring opportunities (when the offense has a first down inside the opponent's 40-yard line) and forcing three turnovers to Vanderbilt's zero.
Even without starting quarterback Connor Mitch, who is out for the season with a separated shoulder, South Carolina will do a better job finishing drives. Former walk-on Perry Orth provided a spark against Kentucky, completing 13 of 20 passes for 179 yards (9.0 yards per attempt), one touchdown and one interception.
After the game, Travis Haney of ESPN.com called Mitch's injury "the best thing that could have happened for that position," saying Orth and true freshman Lorenzo Nunez "have higher ceilings" anyway.
Based on one small sample, I agree.
Georgia won't cover this spread unless it gets more production from its offense—specifically quarterback Greyson Lambert. Vanderbilt proved last weekend that you can dare Lambert to beat you with his arm. Georgia won, sure, but Lambert posted a sub-100 quarterback rating. Without Nick Chubb's heroics, it might have lost.
Chubb will get the Bulldogs to 3-0, but it won't be as easy as it sounds. South Carolina is down right now but always plays Georgia well. Every time I count out Steve Spurrier, he comes up with an unexpected win. Fool me 100 times, shame on me.
The Pick: South Carolina (+17)
Stanford at No. 6 USC
8 of 11
The Line: USC (-10)
USC has been a team-shaped wrecking ball, beating Arkansas State and Idaho by a combined score of 114-15.
The Idaho win means next to nothing, but the 55-6 win over Arkansas State, which stayed within seven points of Missouri the following weekend, shines a favorable light on the Trojans.
Stanford rebounded from its lifeless Week 1 loss at Northwestern by beating Central Florida 31-7, but UCF is not the same team it has been the past few seasons. It's fitting, then, that it played the Cardinal in what appears to be their worst year post-Jim Harbaugh.
If you look back at my Week 1 spread picks, you'll see I took Northwestern over Stanford because I was turned off by Stanford's defensive line depth. I pointed out how nose tackle Harrison Phillips, an undersized sophomore, essentially had no backups.
Phillips tore his ACL against the Wildcats and is out for the season, and his loss depletes an already depleted resource. This is not your older brother's Stanford defense, and there's no way I'm counting on the offense to keep this game close after watching it lay an egg at Northwestern.
The Pick: USC (-10)
No. 15 Ole Miss at No. 2 Alabama
9 of 11
The Line: Alabama (-6)
Prior to last year's upset, Alabama had its way with Ole Miss. It won the previous 10 meetings by an average of 19.1 points, most recently posting a 25-0 shutout when the Rebels came to Tuscaloosa two years ago.
Ole Miss was undefeated before that 2013 meeting as well, but Alabama made it look lost offensively. It did the same for three quarters last season, but then Bo Wallace played a career-best 15 minutes. He finished the upset with 251 passing yards, three touchdowns and zero interceptions, overshadowing how much Ole Miss struggled to run the ball (76 yards on 32 carries).
So basically, Ole Miss failed to hang with Alabama for 10.75 straight games until its quarterback willed it to victory. The only way I see it winning in Tuscaloosa, where it essentially has no chance of running the ball, is if Chad Kelly repeats Wallace's heroics.
Kelly leads the FBS is quarterback rating, but Alabama is a big step up from Fresno State, Tennessee-Martin and the JUCO defenses Kelly played last season. You rarely find Nick Saban laying less than seven points at home, but I think this line has more to do with the public getting bored of Alabama and always being high on Ole Miss than it does any flaws in the Crimson Tide.
The Pick: Alabama (-6)
No. 19 BYU at No. 10 UCLA
10 of 11
The Line: UCLA (-16)
BYU's first two games have made #Pac12AfterDark look like #BigTenAtNoon. The Hail Mary win at Nebraska was probably the saner of its finishes, which also included last week's comeback over Boise State.
The Cougars could have easily lost either or both of those games, but freshman quarterback Tanner Mangum has conjured miracles in crunch time. Between him and UCLA freshman quarterback sensation Josh Rosen, this game should be rebranded "The Hype Bowl." It might as well feature the Clegane brothers.
As for what to expect, here's a stat: Only one of BYU's past 10 and two of its past 16 losses have come by 16 points or more. Virginia lost by 18 at UCLA two weeks ago, and although the Cavs are highly underrated, the Cougars are at least a little better.
I don't think BYU is one of the 25 best teams in the country—at least not without star quarterback Taysom Hill and running back Jamaal Williams—so I'm not exactly stoked about this pick. There's a chance its luck will regress to the mean and UCLA will win by three or four touchdowns. But for one week, I'll ride with superstition and take Mangum in a game with a 10:30 p.m. ET kickoff.
How could this one not go down to the wire?
The Pick: BYU (+16)
Other Top 25 Games
11 of 11
Tulsa at No. 16 Oklahoma (-33)
Beating Tennessee in overtime, on the road, after trailing by 17 points was a season-defining win for Oklahoma. The Sooners would be excused for starting slow against Tulsa, looking ahead to next week's bye and the start of Big 12 play. But the Golden Hurricane look frisky under new head coach Philip Montgomery, who knows Bob Stoops well from his time as offensive coordinator at Baylor. They've scored 40-plus points in each of their first two games, so even if Oklahoma comes out firing, I'd feel good about a backdoor cover.
The Pick: Tulsa (+33)
Nevada at No. 17 Texas A&M (-34.5)
I rode Texas A&M as one of my favorite Week 1 picks against Arizona State, and it came through again by covering a big spread against Ball State last weekend. This time, however, I think the Aggies might get caught in a letdown spot. There's almost no chance Nevada beats them, but covering five touchdowns against a viable Mountain West team is harder than it sounds. It's even harder after two emotional victories (last week marked the debut of Kyle Field's renovations) and one week before the start of SEC play.
The Pick: Nevada (+34.5)
Georgia State at No. 12 Oregon (-45)
No spread south of 50 is too high that I won't pick Oregon. The Ducks are typically money after a loss, and third-year FBS program Georgia State does not have the backbone to keep up with them. After losing to Arizona last season, Oregon raced to a 42-10 lead at No. 10 UCLA before easing off the gas in the fourth quarter. Georgia State has looked improved this season, but sometimes you amble into the wrong place at the wrong time.
The Pick: Oregon (-45)
Northern Illinois at No. 1 Ohio State (-34.5)
I backed UNLV (+28.5) against UCLA last week because I liked how the Rebels stayed close at Northern Illinois in Week 1. After watching the Bruins make quick work of Tony Sanchez's team, I think that result said more about the Huskies than UNLV. Ohio State played as poorly as it might play all season in the first three quarters against Hawaii, then still used a late run to push against a 38-point spread. Five touchdowns sounds small enough.
The Pick: Ohio State (-34.5)
Troy at No. 24 Wisconsin (-34.5)
Wisconsin quarterback Joel Stave has been a pleasant early surprise. He wilted after a strong first half against Alabama, but even one strong half against the Crimson Tide showed progress. If running back Corey Clement returns from a groin injury, he could look to pad his stats after doing nothing versus Alabama and missing last week's 58-0 blowout of Miami (Ohio). But even if Clement stays sidelined, the Badgers can beat this number against a Troy team that ranks No. 123—two spots below Miami (Ohio)—on Football Outsiders' F/+ ratings.
The Pick: Wisconsin (-34.5)
UTSA at No. 25 Oklahoma State (-24)
Oklahoma State is ranked, for some reason, after struggling on the road against Central Michigan and beating Central Arkansas, 32-8. Central Arkansas lost to Samford by 29 points in Week 1, so it's fair to say the Cowboys have struggled twice. UTSA has played a pair of quality opponents, Arizona and Kansas State, and will not be unfamiliar with the level of competition in Stillwater. The Roadrunners aren't great or anything, but they can hang within 24 points.
The Pick: UTSA (+24)
SMU at No. 3 TCU (-37)
How good is SMU really? It hung with Baylor for 35 minutes in Week 1 and beat North Texas by 18 points in Week 2, but it will need to be even better to cover this line against TCU in Week 3. The Horned Frogs know Baylor struggled against the Mustangs, and after watching Art Briles' team jump them in last year's CFP rankings, they will not waste an opportunity to bludgeon a common opponent. It's stupid and political, but hey—it's college football.
The Pick: TCU (-37)
Week 2 Record: 9-9-1 (50.0%)
Overall Record: 20-16-1 (55.4%)
Note: There is no official early line for No. 21 Utah at Fresno State. Most books won't post that number until they learn the fate of Utes quarterback Travis Wilson (shoulder), since quarterback injuries almost always skew the line a couple points. The same concept stopped most books from posting Air Force at No. 4 Michigan State since Falcons quarterback Nate Romine reportedly tore his ACL, according to his private Facebook post, though nothing has been made official. Therefore, I have abstained from picking those games.
Brian Leigh covers college football for Bleacher Report. You can follow him on Twitter @BLeigh35
.jpg)








