Sometime before 6 p.m. ET on Sunday, the NCAA tournament selection committee will email CBS executives a completed bracket to unveil on the selection show. Below is a look at some of the most difficult decisions the committee must finalize between now and then:

1. Is Duke or Virginia the No. 1 overall seed?

If it appeared Virginia had emerged as the obvious choice to be the NCAA tournament’s overall No. 1 seed only a couple days ago, that’s no longer the case.

It’s now a two-horse race after the Cavaliers fell to Florida State in the ACC semifinals and Zion Williamson returned to carry Duke to the ACC tournament title.

On the surface, it’s a close call. Duke owns the superior collection of high-level victories, but Virginia has two fewer losses and better computer metrics. Duke swept a pair of games from Virginia during the season, but the Cavaliers finished two games clear of the Blue Devils in the ACC standings.

VIRGINIA (29-3, 16-2, NET 1, KenPom 1)

Q1 record: 12-3; Q2 record: 4-0

Biggest wins: at North Carolina, Florida State, Virginia Tech (2), Louisville (2), Wisconsin

Losses: Duke (2), Florida State

DUKE (29-5, 14-4, NET 3, KenPom 3)

Q1 record: 11-4; Q2 record: 6-1

Biggest wins: Virginia (2), North Carolina, Kentucky, Texas Tech, Florida State (2)

Losses: Gonzaga, North Carolina (2), at Virginia Tech, Syracuse

What might tip the scales in Duke’s favor is if the committee considers that Williamson missed six games while recovering from the knee injury he suffered Feb. 20. The Blue Devils went 26-2 with the consensus national player of the year on the floor and 3-3 without him.

Of course, injuries are a factor for every team and the committee won’t just wholly disregard the games Williamson missed, but it would be silly to think Duke’s performance with and without him won’t be discussed. Committee chair Bernard Muir already said earlier this week that he would be paying especially close attention to how healthy Williamson looked in his return and how Duke played with him back in the lineup.

The No. 1 overall seed matters to Duke and Virginia because it determines which team goes to the East Regional in Washington D.C. and which goes to the South Regional in Louisville. Not only is Washington D.C. more geographically favorable for both teams, it doesn’t come with the risk of having to play Kentucky in a potential regional final in Louisville.

2. Who should the other two No. 1 seeds be?

Auburn’s 84-64 rout of Tennessee in Sunday’s SEC title game raises an obvious question: Did the Vols already do enough to lock up a No. 1 seed with Saturday’s semifinal victory over Kentucky, or did they need to beat the Tigers too?

Tennessee is one of four teams who went into Sunday thinking they had a chance to join Duke and Virginia on the top seed line. Also still in contention with the Vols are Gonzaga, North Carolina and Michigan State. Kentucky locked itself into a No. 2 seed on Saturday afternoon when it suffered its second loss in three meetings with Tennessee in the SEC semifinals.

Compare the resumes of the four teams vying for the last two No. 1 seeds, and it’s easy to see why it will be a difficult choice.

Gonzaga has the gaudiest record and the strongest computer metrics, but the Zags’ collection of quality wins pales in comparison to their peers. Aside from being the only team to beat full-strength Duke this season, Gonzaga’s most notable victories came against a murderer’s row of Washington, Saint Mary’s and Creighton.

GONZAGA (30-3, 16-0, NET 2, KenPom 2)

Q1 record: 4-3; Q2 record: 6-0

Biggest wins: Duke, Washington, at Creighton, Saint Mary's (2)

Losses: Tennessee, at North Carolina, Saint Mary’s

Michigan State would have the most quadrant 1 wins of any team in the nation if it defeats Michigan for a third time in Sunday’s Big Ten title game, but the Spartans also have more head-scratching losses than any other No. 1 seed contender. In addition to being swept by 17-15 Indiana, they also dropped a game at Illinois.

MICHIGAN STATE (27-6, 16-4, NET 8, KenPom 4)

Q1 record: 12-4; Q2 record: 6-2

Biggest wins: Michigan (2), Purdue, Wisconsin (2), Maryland, Iowa (2)

Losses: at Purdue, Kansas, at Louisville, Indiana (2), at Illinois

Tennessee and North Carolina have similar resumes, right down to their head-to-head wins over Gonzaga and some other overlapping marquee wins and losses. Choosing between them is truly splitting hairs.

NORTH CAROLINA (27-6, 16-2, NET 7, KenPom 5)

Q1 record: 10-6; Q2 record: 6-0

Biggest wins: Gonzaga, Duke (2), Florida State, Louisville (2), Virginia Tech, at Wofford

Losses: Virginia, Duke, Kentucky, at Michigan, Louisville, Texas

TENNESSEE (29-5, 15-3, NET 6, KenPom 7)

Q1 record: 9-5; Q2 record: 6-0

Biggest wins: Gonzaga, Kentucky (2), Louisville, Mississippi State (2),

Losses: at Kentucky, at LSU, Auburn (2), Kansas

You can make a credible argument in favor of any of these four teams, but I think Tennessee and North Carolina have the best case based on their combination of quality wins, no bad losses and head-to-head wins over Gonzaga. The Zags would settle for a No. 2 seed in the West with Michigan State, Michigan and Kentucky joining the Zags on that seed line.

3. How should LSU be seeded without its coach?

An injury to a star player has occasionally impacted a team’s NCAA tournament seeding in past seasons. Now the Selection Committee must decide how to seed a team that’s missing its head coach.

On the eve of its regular season finale, LSU suspended coach Will Wade indefinitely in the wake of reports that FBI wiretaps intercepted discussing a “strong-ass offer” he had made in hopes of landing promising recruit Javonte Smart. Interim coach Tony Benford then guided the Tigers to an SEC title-clinching victory over woeful Vanderbilt on March 9 and a 79-76 upset loss against Florida in the conference tournament quarterfinals.

LSU might have been able to make a push for a No. 2 seed with a deep SEC tournament run, but exiting without a single victory leaves the Tigers with a resume that doesn’t compare with the likes of Kentucky, Michigan or Michigan State. As a result, a No. 3 seed is the ceiling for LSU, though that’s no certainty either as you’ll see from the comparison below between the Tigers and four other contenders for that seed line.

LSU (26-6, 16-2, NET 14, KenPom 18)

Q1 record: 9-3; Q2 record: 7-3

Biggest wins: at Kentucky, Tennessee, at Mississippi State, Auburn

Losses: at Houston, Florida State, Florida, Arkansas, Oklahoma State

HOUSTON (31-2, 16-2, NET 4, KenPom 13)

Q1 record: 6-2; Q2 record: 10-0

Biggest wins: LSU, Cincinnati (2), at UCF, Utah State

Losses: at Temple, UCF

KANSAS (25-9, 12-6, NET 19, KenPom 16)

Q1 record: 11-8; Q2 record: 7-1

Biggest wins: Tennessee, Michigan State, Texas Tech, Wofford, Villanova, Marquette, Kansas State, Iowa State

Losses: at Kentucky, at Texas Tech, at Kansas State, Iowa State (2), at Oklahoma, at Texas, at Arizona State, at West Virginia

FLORIDA STATE (27-7, 13-5, NET 15, KenPom 14)

Q1 record: 8-5; Q2 record: 6-1

Biggest wins: Virginia, LSU, Purdue, Virginia Tech (2), Louisville

Losses: at Virginia, Duke (2), at North Carolina, Villanova, at Pittsburgh, at Boston College

TEXAS TECH (26-6, 14-4, NET 10, KenPom 9)

Q1 record: 8-5; Q2 record: 8-0

Biggest wins: Kansas, Kansas State, at Iowa State, Oklahoma (2)

Losses: Duke, at Kansas, at Kansas State, Iowa State, at Baylor, West Virginia

Kansas has the best collection of marquee wins of that group. Texas Tech ranks the highest in the computer metrics. Houston and Florida State both own head-to-head wins against LSU in addition to strong resumes of their own.

LSU has a strong argument to jump Kansas, Texas Tech or Florida State among those four teams, but the margin is paper-thin. That could be the excuse the committee needs to drop the Tigers to the No. 4 line based on Wade’s absence or the desire to try to avoid the negative publicity that would accompany a scandal-tainted program making a deep run.

4. How many mid-major bubble teams belong in the field?

For the past few years, the committee has placed the greatest emphasis on the number of top 50 wins or quadrant 1 wins teams amass. As a result, it has gotten harder for mid-major at-large contenders to make the NCAA tournament because they typically have fewer opportunities to play those games.

Consider Middle Tennessee, left out of last year’s field despite a gaudy KenPom ranking, a 24-7 record and a recent history of NCAA tournament success. Or Monmouth, somehow snubbed in 2016 despite a conference title and non-league victories over Notre Dame, USC, UCLA and Georgetown.

This year’s committee has a chance to make a statement by swinging the pendulum back in the other direction. There are two deserving bubble teams in particular who could capitalize if the committee favors successful mid-majors who didn’t have many cracks at marquee wins over middling power-conference teams who had lots of opportunities but squandered most of them.

BELMONT (26-5, 16-2, NET 47, KenPom 53)

Q1 record: 2-2; Q2 record: 2-1

Biggest wins: at Murray State, Lipscomb (2), at UCLA, Western Kentucky

Q3, Q4 losses: 2 (at Green Bay, Jacksonville State)

UNC GREENSBORO (28-6, 15-3, NET 59, KenPom 80)

Q1 record: 2-5; Q2 record: 2-0

Biggest wins: Furman (2), East Tennessee State (2)

Q3, Q4 losses: 0

Belmont’s winning record in games in the top two quadrants bolsters its case, but the committee will have to overlook the fact that 65 percent of its wins this season have come in quadrant 4 games. UNC Greensboro can boast that it’s only lost once all season to a team outside the NET’s top 15, but the committee will have to forgive its glaring lack of notable victories.

It would be fun to see accomplished mid-majors like this get a chance over the likes of TCU, NC State or Alabama, but recent history suggests Belmont and UNC Greensboro should both be worried. My guess is the committee gives Belmont the chance to prove itself in the First Four but relegates UNC Greensboro to the NIT.

5. What about 17-15 Indiana or ... gulp … 16-16 Texas?

Team A boasts the most Quadrant 1 wins of any bubble team including a season sweep of Big Ten champion Michigan State and three other victories over opponents projected to be seeded seventh or better in the NCAA tournament.

Team B has a top 40 NET ranking and seven victories over projected NCAA tournament teams including ACC co-champion North Carolina, Big Ten co-champion Purdue and Big 12 powers Kansas, Kansas State and Iowa State.

Those two bubble teams sound like strong candidates for at-large bids, right? Well, what if I also told you they have a combined record of just 33-31?

The above helps explain why Indiana (17-15) and Texas (16-16) likely pose more problems for the committee than any other bubble teams this year. They both have the quality wins of a team safely in the NCAA tournament and the overall records of a team that would be lucky to make the NIT.

There is precedent for teams similar to Indiana being selected in recent years. Alabama snagged an at-large bid last March with 15 losses, as did Vanderbilt the previous season. Villanova and Georgia also both received at-large bids at 16-14 in 1991 and 2001, respectively.

A bid for Texas, on the other hand, would be uncharted territory. Never before has a .500 team received an at-large bid to the NCAA tournament, nor has a team with 16 losses before Selection Sunday.

It would take guts for the committee to include Texas given the inevitable backlash, but the committee could have incentive to take a hard look.

A ridiculous 19 of the Longhorns’ 32 games this season came against opponents in the NET top 50. Including them would send the message that the committee rewards a tough non-conference schedule, something that is good for the health of the sport.