College Basketball Bracket Predictions: Duke or San Diego State?

Bracketology Bracket Predictions

Can Tre Jones and Duke get the final No. 1 seed? (Photo by Mary Holt/Icon Sportswire)

In this post, we’ll provide weekly college basketball bracket predictions in the run up to the 2020 NCAA men’s basketball tournament. These predictions are based on our data-driven bracketology models, though we do make some judgment calls and manual adjustments in cases where we believe they are warranted.

How Our College Basketball Bracket Predictions Are Different

Most bracketologists simply analyze team results to date when they make college basketball bracket predictions. That’s a pretty crude and simple approach. Projecting the NCAA field as if teams will not play 18 or more conference games, or as if team strength has no impact on likely future results, will not result in the most accurate bracket projections — especially when there is a lot of season left to play. In short, it’s a lot harder to predict this year’s bracket in February than it is on the day before Selection Sunday.

Unlike most bracketologists, we use our college basketball season simulations (which we update daily) to forecast where teams will end up at the conclusion of the season. Then we project the NCAA Selection Committee’s decisions using a purpose-built model we designed, based on historical selection and seeding data. Even though our predictions will never be perfect, that approach is still far more sophisticated than the prevailing standard.

The process of modeling all this stuff in a 100% automated way is quite complex, though, and there are still some corner cases where we think our models need further refinement. So the final step in our bracket prediction process for this article is having human eyes review the data, and adjust some teams up or down if we think it makes sense.

2020 College Basketball Bracket Prediction: As of Results through February 12th

 

[click image to enlarge]

Here are some observations and notes about some of the bracketing decisions, based on results through Wednesday, February 12th.

There are really about four tiers of teams when looking through the bracketing decisions, for the at-large pool:

  • The first tier consists of five teams for the four No. 1 seed spots. As noted in the Bracket Reveal analysis, we currently project Duke as a No. 1 seed, though the Committee has them and San Diego State swapped. It’s really close right now, and if San Diego State finishes undefeated, they will get that spot. If the Aztecs win at Boise State this week, the odds swing in their favor to get the final No. 1 slot.
  • The next tier has the other teams mostly on the No. 2 and 3 seed lines. These teams would have to make major runs and likely not lose another game to get in the top seed mix, but would also have to slump to fall out of this group.
  • Next comes a large cluster of teams appearing at seeds 4 to 7. There were a lot of tough calls here, and you could still see plenty of movement in this group. How the Pac-12 (Oregon, Colorado, and Arizona), Big East (lots of teams) and Big Ten (lots of teams) shake out will determine the ultimate pecking order. One or two key results could move teams a seed line or two right now.
  • After that is the big mix of teams from the No. 8 seed line to the end of the at-larges, and beyond to the other side of the Bubble. These teams are very much in the mix but also at risk of missing the tournament, and week-to-week results could alter the outlook. In our look at parity in college basketball, we noted that the bubble looked strong in terms of team quality this year. In terms of resumé and how the committee usually decides such things, though, most of these teams need to add a big win or two over a top team down the stretch to feel safe.

Biggest Weekly Movers in our College Basketball Bracket Predictions

Here the biggest positive and negative changes over the last seven days in tournament odds

TeamCurrent OddsChange TeamCurrent OddsChange
Michigan94%+29%Mississippi State41%-32%
Oklahoma97%+25%Arkansas51%-24%
NC State35%+17%VCU32%-23%
Cincinnati84%+14%Wichita State45%-21%
Tennessee54%+14%Providence40%-21%

And here are the biggest positive and negative changes in average projected seeding over the same time frame. In this case, remember that negative is good (as the average seed is going down) and positive is bad. These are limited to teams with at least a 10% chance of reaching the tournament.

TeamAvg. SeedChange TeamAvg. SeedChange
Creighton4.6-1.6Arizona4.6+1.2
Penn State5.3-1.3West Virginia4.3+1.2
Oklahoma9.7-1.1Yale11.3+1.2
Michigan9.8-1.1Stanford8.9+1.1
Texas Tech8.0-1.1

Creighton got a big road win at Seton Hall that boosted them up the seed list. Penn State continued their winning streak (now seven games) by beating Minnesota and winning at Purdue and have realistic visions of a protected seed. Meanwhile, it was a bad week for the SEC Bubble. Arkansas lost three straight to move onto the bubble, and Mississippi State suffered a bad 25-point loss against Ole Miss.

Bubble Watch: Bracket Predictions for the Final At-Large Teams

To see where your team stands, check our Bracketology page for up-to-date predictions, as these numbers could shift based on each incoming result.

Here’s a rundown of some of the teams closest to the bubble this week, in particular teams just on either side of the cutline in the most recent Bracket Matrix.

Cincinnati/Wichita State

Cincinnati is still on the outside of the bubble in the Bracket Matrix, while we have projected them in the field for several weeks in a row. Meanwhile, Wichita State is holding on as a 10-seed in the Bracket Matrix. This is your reminder that Cincinnati won at Wichita State last week, and sits tied for second in the American. Wichita State fell to 5-5 in conference, and has lost three straight. That includes important losses to Cincinnati and Houston.

We feel pretty good about this projection. Wichita State is currently just on the outside of the automated bracketology. They are just barely in the field in our projections here. But either way, this has not been a great run for them and they are very much in bubble trouble. The Bearcats are in much better shape, and still get the Shockers at home in their remaining head-to-head matchup.

VCU

VCU is currently in the field on the Bracket Matrix, and would be slotted for a play-in game in Dayton. But we have them out. They have already lost twice to Rhode Island, and lost the first game to Dayton. They also went 1-3 in the non-conference against LSU, Tennessee, Wichita State, and Purdue. VCU will almost certainly need to beat Dayton next week to keep at-large hopes alive.

Indiana

The Hoosiers are still somewhat comfortably in the Bracket Matrix field. They are in almost 75% of entries, and just above the cutline to go to Dayton for the play-in games. They are also in the last four at-larges in the automated bracketology predictions, holding on at a 46% chance. I’ve moved them out of the current projections for a few reasons, though. First, I think as a practical matter, the Big Ten is going to have a hard time getting more than 10 teams. Indiana has now lost four in a row. If they hit their projected record, they will be 8-12 in the Big Ten, and likely fall outside the Top 60 in the NET. I think those factors will give the committee enough reason not to strain to work another Big Ten team in at the bubble.


2020 College Basketball Bracket Prediction: As of Results through February 5th

 

Bracket for 2020 NCAA Tournament Predictions February 6

[click image to enlarge]

Here are some observations and notes about some of the bracketing decisions, based on results through Wednesday, February 5th:

  • The top seed line continues to have five strong candidates, and San Diego State’s case becomes stronger the longer they remain undefeated. It’s hard to imagine them not getting the No. 1 seed if they can finish the year without a loss, though right now they are slotted as the top No. 2 seed.
  • There was plenty of movement in the 3-6 seed positions in the bracket this week, and that may be an indicator of what it will be like the rest of the season. Michigan State dropped with two losses, Auburn moved up with two big wins, and the top of the Big East looks like it could swing back and forth for the rest of the season.
  • A handful of teams made big moves near the bubble or to jump in the field. Xavier (road wins over Seton Hall and DePaul) and Providence (wins over Butler and Creighton) are two teams from the Big East that improved their fortunes.
  • We slot Northern Iowa and East Tennessee State into the 12-seed line for now, but do not be surprised to see these teams finishing higher. They are the two most likely at-large candidates from mid-majors. Both have a decent case if they do not win the conference tourney auto bids.

Biggest Weekly Movers in our College Basketball Bracket Predictions

Here the biggest positive and negative changes over the last seven days in tournament odds

TeamCurrent OddsChange TeamCurrent OddsChange
Providence61%+43%USC40%-25%
Xavier75%+34%North Carolina14%-25%
Purdue73%+20%Rutgers60%-17%
Arizona State48%+19%Washington10%-15%
Penn State93%+17%Alabama74%-15%

And here are the biggest positive and negative changes in average projected seeding over the same time frame. In this case, remember that negative is good (as the average seed is going down) and positive is bad. These are limited to teams with at least a 10% chance of reaching the tournament.

TeamAvg. SeedChange TeamAvg. SeedChange
Auburn3.3-1.5Michigan State5.9+1.7
Penn State6.6-1.3Villanova4.7+1.2
Providence9.1-1.2Alabama9.8+1.2
Creighton6.2-1.1Georgia State12.9+1.0
Xavier9.8-1.0UNC-Greensboro10.7+0.9

Our computer projections are even lower on Stanford than the above bracket, but either way, it was a bad week for Cardinal, losing at Cal and then at home to Oregon State. Memphis is also continuing to go the wrong direction. Florida missed two opportunities at home against tournament-type teams, losing to Baylor and Mississippi State. Meanwhile, Virginia got a much-needed win over Florida State, and Villanova won twice on the road in the Big East in the last week.

Bubble Watch: Bracket Predictions for the Final At-Large Teams

To see where your team stands, check our Bracketology page for up-to-date predictions, as these numbers could shift based on each incoming result.

Here’s a rundown of some of the teams closest to the bubble this week, in particular teams just on either side of the cutline in the most recent Bracket Matrix.

But first, a look at some of our recent bubble discussions from the last two weeks:

Last week, we noted that the Bracket Matrix had both NC State and Virginia Tech in the field. Virginia was out, which was the complete opposite of our position. In one week, consensus moved toward us quickly. NC State and Virginia Tech have now virtually dropped out of the field across all brackets, while Virginia is now in the field in 80% of brackets. Last week, we also noted that Memphis was in and Cincinnati out in most brackets, while we had the opposite. So far, that one remains unchanged. Cincinnati’s home win over Houston did not give them a boost for whatever reason in many brackets. Cincinnati is ranked higher predictively and is projected to finish with the better record in the American Conference (and we give them the second-best chance to win the conference regular season and tournament).

Two weeks ago, we mentioned DePaul as a team that was in most brackets, but not ours. They’ve continued to lose and have now dropped out of the Bracket Matrix. We also mentioned Alabama as the opposite case. So far, that has not gone as well. Alabama has lost three in a row. We are still projecting them right at the bubble cut line, but they are generally out of the Bracket Matrix still.

Now, to this week:

USC/Stanford

USC and Stanford are both solidly in the tournament per the Bracket Matrix results. Both are on the No. 9 seed line, with at least 95% of brackets having them in. Both are out according to our computer simulation projections right now. I narrowly put Stanford in the last four after locking in the home win over Oregon, while USC did fall out. USC lost a key home game to Colorado. That was a blown opportunity for a Quad 1 win, and now they have just three left: at Arizona, at Colorado, and Arizona at home. We project the Trojans to lose more than half their remaining games.

Stanford, meanwhile, is a NET favorite (which is why I’ve adjusted them in the field) but the Oregon game was their first Quad 1 victory of the year. They are in better shape with home games against Arizona and Colorado looming, but they are not in the safest position either.

Xavier

Xavier is just on the outside of the Bracket Matrix (first team out, with 45 votes of 99). We moved them in after a big week that included a road win at Seton Hall. They had slumped with five losses in six games, but turned things around. They will get plenty of more Quad 1 opportunities in a conference with lots of good teams.


2020 College Basketball Bracket Prediction: As of Results through January 30th

Bracket Predictions 2020 NCAA Tournament

[click image to enlarge]

Here are some observations and notes about some of the bracketing decisions, based on results through Thursday, January 30th:

  • Our top seed line remains the same, and right now it looks like there are five primary contenders for a No. 1 seed. We’ve got Kansas, Duke, Baylor, and Gonzaga in our projection, and San Diego State would certainly be a strong candidate if they complete an undefeated regular season, and in the mix at 31-1 with a Mountain West title. Other teams could get in that conversation with a great closing month.
  • Some of the closest calls were deciding the last No. 2 seed (Michigan State as projected Big Ten champ over Louisville and Oregon), which team got bumped down to the No. 4 line (Florida State, versus Seton Hall or Villanova), and the group of teams on the 4/5 line and the top of the No. 6 seed line.
  • Just as with last week, the biggest seeding difference between this bracket and our automated version is the placement of Big Ten teams.
  • We will likely delve into this in more detail in a full post, but the bubble can shrink due to some current projected auto-bids falling to the at-large pool. So teams currently projected in the Dayton play-in games can realistically be thought of as at risk.

Biggest Weekly Movers in our College Basketball Bracket Predictions

Here the biggest positive and negative changes over the last seven days in tournament odds

TeamCurrent OddsChange Team Current OddsChange
Virginia60%+29%NC State 24%-36%
North Carolina40%+25%Tennessee40%-21%
Illinois86%+22%Stanford11%-20%
USC68%+22%Memphis42%-20%
Arizona State34%+16%Richmond18%-19%

And here are the biggest positive and negative changes in average projected seeding over the same time frame. In this case, remember that negative is good (as the average seed is going down) and positive is bad. These are limited to teams with at least a 10% chance of reaching the tournament.

TeamAvg. SeedChange TeamAvg. SeedChange
Villanova3.6-1.4Florida8.4+1.8
Kent State11.4-1.3E. Tenn State8.7+1.6
Butler4.6-1.1Richmond9.2+1.5
LSU4.7-1.1Stanford8.5+1.1
Maryland4.8-0.9BYU8.6+1.0

Our computer projections are even lower on Stanford than the above bracket, but either way, it was a bad week for Cardinal, losing at Cal and then at home to Oregon State. Memphis is also continuing to go the wrong direction. Florida missed two opportunities at home against tournament-type teams, losing to Baylor and Mississippi State. Meanwhile, Virginia got a much-needed win over Florida State, and Villanova won twice on the road in the Big East in the last week.

Bubble Watch: Bracket Predictions for the Final At-Large Teams

The bubble is still fairly large at this point. Our predictions give only 28 teams a greater than 85% chance of making the field.

That still leaves 20 teams that have between a 45% and 85% chance of making the field (and at least a 10% of doing so specifically as an at-large).  Eighteen of those 20 are currently projected into the field as either at-larges or auto bids. An additional 13 teams have less than a 45% chance of making it, but are given at least a 10% chance of earning an at-large. To see where your team stands, check our Bracketology page for up-to-date predictions, as these numbers could shift based on each incoming result.

Here’s a rundown of some of the teams closest to the bubble this week, in particular teams just on either side of the cutline in the most recent Bracket Matrix.

NC State/Virginia/Virginia Tech

These ACC teams are likely to be in the middle of the bubble talk for the next month and a half. Interestingly, even after the results of this week, the Bracket Matrix has NC State and Virginia Tech just in (last four), and Virginia in the first four out. Virginia beat Florida State this week, and our projections now give them a 60% chance of making it. In comparison, NC State lost twice, at Georgia Tech and at home to North Carolina. Virginia Tech also took two potentially costly losses in the last seven days, at Boston College and at Miami. Add in that Virginia Tech only gets Louisville and Duke on the road, and their outlook took a big hit.

So we are opposite the consensus here, with Virginia the most likely team now to make the field of these three.

Cincinnati/Memphis

Memphis is still in the projected field for 70% of the Bracket Matrix, while Cincinnati is rarely in (8% of the time). But this is a case of predictive measures and where we think the teams are more likely to be on Selection Sunday. These two teams are going in opposite directions.

Right now, Cincinnati does not have a “signature” win which is why Brackets that look only at current resume are largely ignoring them. Memphis, meanwhile, have the slightly better resumé, but are trending the other way without star freshman James Wiseman, who was initially declared ineligible and then opted not to return. After a 12-1 start, they have lost 4 of the last 7, and have dropped out of the top 50 in predictive ratings.

Cincinnati still has two games left against both Wichita State and Houston, and a home game against Memphis, to add to their results. We are currently projecting them to finish 2nd in the American, and for Memphis to finish up in 5th. That’s why Cincinnati is in our field and Memphis is not.

 


2020 College Basketball Bracket Prediction: As of Results by January 22nd

2020 Bracketology Predictions, January 22nd[click image to enlarge]

Here are some observations and notes about some of the bracketing decisions, based on results through Wednesday, January 22nd:

  • Kansas, Duke, Baylor, and Gonzaga currently top our list as the most likely No. 1 seed choices on Selection Sunday.
  • The top of the bracket has a pretty clear cut group of 14 teams that look targeted for the top 4 seed lines based on current performance and future schedule.
  • The most common substantive changes, compared to our automated bracketology predictions, is that the Big Ten teams are generally seeded better here. The Big Ten expanded to 20 conference games last season. This year, the conference is deep and loaded with potential tournament teams. The seeds should reflect that difficult schedule and the number of quality teams.
  • BYU’s “no games on Sunday” policy could lead to some big shifts in the bracket. The West and Midwest are this year’s Thursday/Saturday Regionals. WCC rival Gonzaga is the most likely West No. 1 seed. The other most likely West top seed, San Diego State, is a natural fit for the Sacramento site, which plays on Sunday. Kansas is the most likely Midwest No. 1 seed and would almost certainly get assigned to Omaha, another Friday/Sunday site.
  • This bracket differs on the identity of a few teams on each side of the bubble. Virginia Tech and  Stanford are two teams that our models are lower on than the consensus, and we move toward the consensus in those two cases here. Oklahoma and Mississippi State have good overall computer numbers, but they are moved out because of their poor projected Quadrant 1 records. Richmond and Rhode Island are swapped right on the cutline; Richmond won at Rhode Island recently and they otherwise are close.

Biggest Weekly Movers in our College Basketball Bracket Predictions

Here the biggest positive and negative changes over the last seven days in tournament odds

TeamCurrent OddsChange TeamCurrent OddsChange
Alabama86%+27%Purdue43%-27%
NC State 61%+25%Oklahoma State10%-22%
Rutgers72%+23%Xavier37%-20%
VCU61%+18%Oregon State5%-15%
Marquette96%+16%Cincinnati43%-14%

And here are the biggest positive and negative changes in average projected seeding over the same time frame. In this case, remember that negative is good (as the average seed is going down) and positive is bad.

TeamAvg SeedChange TeamAvg SeedChange
Seton Hall3.0-1.3Butler5.9+2.6
Marquette7.7-1.2Auburn4.6+1.7
Louisville2.8-1.2Akron9.6+1.2
Alabama8.1-1.1Purdue10.6+1.1
Houston5.8-0.9Wichita State5.8+1.1

As you can see by these results, big wins matter. Houston won at Wichita State, Alabama had a great week that included a win over Auburn. Purdue lost at home to Illinois and at Maryland.

Bubble Watch: Bracket Predictions for the Final At-Large Teams

The bubble is still fairly large at this point. Our predictions give only 30 teams a greater than 85% chance of making the field.

That still leaves 20 teams that have between a 45% and 85% chance of making the field (and at least a 10% of doing so specifically as an at-large).  Eighteen of those 20 are currently projected into the field as well as either at-larges or auto bids. An additional 17 teams have less than a 45% chance of making it, but are given at least a 10% chance of earning an at-large. To see where your team stands, check our Bracketology page for up-to-date predictions, as these numbers could shift based on each incoming result.

Here’s a rundown of some of the teams closest to the bubble this week, in particular teams just on either side of the cutline in the most recent Bracket Matrix.

Minnesota/Purdue

Minnesota and Purdue are typically projected just into the field right now (though the impact of Purdue’s home loss to Illinois may not be fully reflected in all bracket predictions). But these are the 11th and 12th teams from the Big Ten. The Big Ten, as strong as it may be, is probably not getting 12 teams in the tournament. They will cannibalize each other over the next six weeks. Right now, we project Purdue to finish 16-15 and Minnesota to finish 16-14. History shows it would be hard to make the field with those records.

DePaul

Here’s one that shows the divide between Bracketology methods that focus on current results, and those that incorporate future predictions. Based on current results, DePaul would be in. They have wins over Iowa, Minnesota, Texas Tech, and Butler. But there is no tournament taking place right now. DePaul is only 84th in our predictive ratings, and dropped to 1-5 in the Big East with a loss at home to Creighton on Wednesday. We project them to finish with 5.0 wins in the Big East, tied for last. If that projection is close, they aren’t getting in the field.

Alabama

Alabama is the opposite case. Less than 25% of bracket predictions on the Bracket Matrix have them in the field at 11-7, but we project them solidly in the field. The Crimson Tide started 4-5, but are playing much better with their current lineup. They have won 7 of the last 9, including wins over Auburn and Mississippi State. We project them to finish 11-7 in the SEC and 19-12 overall. If they hit that median expectation, that would be good enough for an at-large berth.