In case you’re interested in my NIT bracket, or how I fill it out (as I’m being told you are by our content team), here you go.
I used our model to inform me on making these picks, which is part of why the picks are grouped together as they are. While this isn’t the most likely individual bracket scenario, that scenario would involve picking so many favorites that it wouldn’t be any fun. So instead I picked a realistic bracket that has some flavor to it. Here’s how I did it:
First Round:
Indiana over St. Francis (PA)
Alabama over Norfolk State
TCU over Sam Houston State
Clemson over Wright State
In our model, the most common occurrence among the 1/8 and 2/7 games is for exactly one seven or eight-seed to win. I’m not sure if this happens in the majority of simulations, but it’s the expected value.
These four games are, in the order I’ve listed them, the least likely to see such an upset. And while I do have some mild qualms about how much each of the four teams will care, I don’t see any as a clear-cut example of the “this team doesn’t want to be here” NIT phenomenon.
These next four were all about equally likely to see an upset, with Loyola/Creighton at 22% and the other three at 20%.
UNC-Greensboro over Campbell
I’m anticipating a Chris Clemons scoring night to remember, especially given that UNC-Greensboro is in the bottom half of Division I when it comes to three-point defense, but UNCG’s calling card all year has been their consistency. Their worst loss was a road loss to Furman. Their undefeated record in games they’re expected to win is statistically significant.
Creighton over Loyola
Our model can wobble up to two or three percent in any direction with an individual game, just because our sample size of simulations is only 1,000. So I’m not viewing this as necessarily the most mathematically likely upset. Additionally, I’m guessing Loyola will be a more common upset pick in our pool because of our Midwestern base and their performance last year, so there’s some value in taking Creighton.
Lastly, Creighton can shoot the ball, and while Loyola’s combination of a slow tempo and limiting offensive rebounds is a traditional recipe for a Cinderella team (because it limits possessions, decreasing the sample size contained in an individual game and increasing variability), Loyola’s defense just doesn’t look good enough on paper to handle Creighton’s offense.
North Carolina State over Hofstra
I’m concerned about NC State in this one because of their propensity to foul and Hofstra’s propensity to get to the line and make free throws. I don’t like Hofstra as much as South Dakota State, though, and I think NC State and Texas are equally valuable later in the tournament.
South Dakota State over Texas
So here’s the pick, of the eight. Texas should struggle to contain Mike Daum, especially behind the three-point line. I also have questions about how the Longhorns are meshing with Kerwin Roach back in the lineup. Texas and North Carolina State both have a lot to be disappointed about, but I’ll take Texas as the one to lose.
Between the 3/6 games and 4/5 games, our model expects the lower-seeded team to win in three matchups (again, this isn’t more likely than the combination of all the other possibilities, but it’s the median outcome). On average, about 1.3 of those come from 6-seeds, and 1.7 come from 5-seeds.
Furman over Wichita State
Memphis over San Diego
Wichita State and San Diego are the two least likely 6-seeds to win an NIT game.
Wichita State should get a lot of second chances against Furman, but while the Shockers have played well lately, their best win came over Temple, who might not be as good as Furman, straight up. In the end, I don’t trust Wichita State to play efficient enough offensive basketball to beat Furman.
San Diego will limit Memphis’ second chances, but Penny Hardaway’s team generates a lot of possessions, offensive rebounds or not, and San Diego lacks good matchups on their offensive side of things.
Georgetown over Harvard
Toledo over Xavier
Our model prefers Toledo to Harvard as a first-round winner, and the matchups say it’s right.
Georgetown’s going to move the ball up and down the court and spread it around, pulling Harvard out of its comfort zone. Harvard should be able to give the Hoyas a game, but there isn’t anything the Crimson are so excellent at that they’ll clearly be able to exploit a mismatch.
Toledo, on the other hand, is just a good all-around team on paper. Yes, they haven’t beaten anybody, but they played Buffalo tough back around Valentine’s Day, and Xavier, while playing much better these days than they were in January, hasn’t done anything spectacular aside from beating Villanova at home and playing them to overtime in New York. Toledo’s an experienced team with a strong defense, and Xavier’s willingness to slow things down should keep it close enough to give Toledo a great opportunity.
Nebraska over Butler
Our model adores Nebraska, partially because it doesn’t fully appreciate the impact of injuries on the Cornhuskers. But even though our model is slanted towards Lincoln, I still trust it when it says this is less likely to be an upset than any of the 3/6 games (it’s worth noting that thanks to home court advantage, our model doesn’t view any poorer-seeded teams as first-round favorites), especially since no matchups obviously favor Butler.
Lipscomb over Davidson
Our model has Lipscomb winning this 45% of the time, meaning it’s nearly a tossup. Were Lipscomb at home, that number would be in the 60%’s, meaning that Lipscomb probably is the better team, and will just need to handle playing Davidson on the road.
Davidson likes to play slowly and shoot a lot of three’s. Lipscomb plays up-tempo basketball, but its defense is relatively effective at neutralizing shots from behind the arc. This game very well might come down to which team can control the pace. I’ll take a Lipscomb team that’s played Louisville tightly and smoked Vermont (not to mention smoked Liberty once) over a Davidson whose best result is a seven-point home win over VCU.
Colorado over Dayton
Providence over Arkansas
Were Daniel Gafford playing for Arkansas, I could go either way on both of these, and our model views them as similarly likely to see the 5-seed win.
On paper, Dayton needs to score inside, which Colorado will probably allow them to do. Dayton doesn’t run a very deep rotation, though, and while they manage this by slowing things down, it could cause them some issues late in the second half a mile above sea level.
Arkansas can’t allow Providence to beat them at the free throw line, but they very well might. What gives Arkansas an upper hand is their ability to force turnovers, and Providence’s relative inability to protect the ball. Still, with Gafford out, it’s silly for me to not take Providence.
Second Round:
Our model expects the most likely number of one and two-seeds to win their second round game to be either four or five. It says the most likely number of three-seeds to win is one. It says the most likely number of four or five-seeds to win is one or two, and the remaining place in the quarterfinal will more likely than not be taken by a six, seven, or eight-seed.
Toledo over South Dakota State
I’m bound to have one six or seven-seed in my quarterfinals, by virtue of picking these teams to play each other. I’m taking Toledo as a value play, as I’m guessing South Dakota State will be a bit of a crowd favorite thanks to Mike Daum.
If it comes to pass, this would be a fun matchup. Mike Daum is so difficult to stop, but Toledo is comfortable defending teams that play at South Dakota State’s pace, and the Rockets are a good defensive team on the whole.
At the other end of the court, South Dakota State could be in trouble. Toledo just has more weapons than the Jackrabbits are used to dealing with. While South Dakota State did give Nevada (who similarly has a lot of weapons) a run for their money in December, the Jacks’ three-point defense was abnormally successful that day, and Nevada doesn’t shoot as well as Toledo.
Lipscomb over UNC-Greensboro
UNC-Greensboro is our model’s most likely 1-seed or 2-seed to not make the quarterfinals (46% chance of making it that far). On paper, they just aren’t that strong of a team, to the degree where Lipscomb might actually be the favorite if this matchup comes to pass.
If these teams do play, Lipscomb needs to protect the ball. Aside from that aspect of the game, Lipscomb holds the advantage, but Isaiah Miller could cause a whole lot of trouble for Kenny Cooper. I’ll take my chances that Lipscomb’s efficient scoring on possessions where they don’t turn the ball over will be enough.
North Carolina State over Georgetown
Indiana over Providence
Clemson over Furman
All three of these teams are about 64% likely to win their first two games, according to our model. I like NC State to put up a whole lot of points against Georgetown. I like Indiana to keep Providence off the free-throw line and take care of business elsewhere. I think Clemson might have trouble with Furman if the Paladins make their three’s, but I don’t think Furman can live on three’s alone, and Clemson’s defense looks too strong elsewhere.
Alabama over Colorado
Colorado’s best road win this year came against USC, and I don’t see them improving that down in Alabama, though it may be an ugly game offensively.
Nebraska over TCU
I decided to go with two 4/5-seeds in the quarterfinals, mainly because Nebraska is more likely, according to our model, to make the quarterfinals than any team with a poorer seed than two (our model has Nebraska making the quarterfinals in 32% of simulations).
These are two of the best teams in this year’s NIT, and if Nebraska does win, it won’t come easy. TCU won’t let them score many from deep, so Nebraska will need to penetrate and get as many second chances as they can. Defensively, they’ll need to limit TCU’s good looks, which can be tough against a team that moves the ball as well as Jamie Dixon’s.
Memphis over Creighton
Our model likes Memphis just as much as it likes Nebraska, and it isn’t too high on Creighton. It has Memphis reaching the quarterfinals in 31% of simulations, with Creighton only at 50%.
Creighton will have to run with Memphis to beat them. Unlike many of Creighton’s victims, Memphis plays adequate perimeter defense. I’ll take the Tigers.
Quarterfinals:
Our model doesn’t like any six, seven, or eight-seeds to make the Final Four, as the combined Final Four probabilities of those teams sum to only 23%. For the three, four, and five-seeds, that number is 106%, meaning that the median number of those teams to make it is likely one.
North Carolina State over Lipscomb
NC State is our most likely team to make the NIT Final Four, with a 44% probability of getting that far. Some of this can be owed to UNC-Greensboro’s weakness, but the Wolfpack are just a solid team—Sagarin and BPI both have them as the best team in the field, and KenPom doesn’t have them far behind Clemson and Texas for that designation.
I like Lipscomb a lot, but NC State is capable of running with the Bisons, challenging them on Lipscomb’s defensive glass, and forcing turnovers.
Memphis over Nebraska
Our model has Memphis and Nebraska both as decently likely Final Four teams (14% and 16%, respectively). I like Memphis slightly more because I just don’t know how long Nebraska can keep winning with their depleted rotation. Memphis is a good pick to take them down because the Tigers play at such a fast pace, and because Nebraska’s weak free-throw shooting negates the issue of Memphis fouling too much.
Alabama over Toledo
Our model says Alabama’s the sixth-most likely team to make the Final Four, giving them a 33% likelihood of winning their first three games. It has Texas as a slight favorite to win this region, but I already eliminated Texas, for better or for worse.
Toledo has only a 5% chance of making the Final Four, according to our model. And while this would be a great matchup (Alabama is one spot behind Toledo on KenPom), the game would be in Tuscaloosa, and Alabama has to go through less to get to this point.
Clemson over Indiana
This looks like a great matchup on paper, especially in Bloomington. Luckily for Clemson, Indiana hardly utilizes the three at all, and isn’t great at forcing turnovers. With those weaknesses of the Tigers somewhat negated, and with questions about Romeo Langford’s back, Clemson seems like a valuable pick.
Final Four:
North Carolina State over Clemson
NC State has won both of their matchups with Clemson so far by a combined total of three points. I could go either way on this, but I’m more confident in NC State getting to New York than I am in Clemson doing the same, and I’m guessing Clemson may be a popular pick among a certain demographic in our pool given that they’re the highest-ranked NIT team on KenPom. Also, NC State’s our model’s overall NIT favorite, and while the field is favored over the Wolfpack by a ton, our model is on to something as far as the ease of their path goes.
Memphis over Alabama
If this comes to pass, it could be a great undercard to the NC State/Clemson matchup. The teams are well-matched, and while our model likes Alabama more than Memphis to win the whole tournament, if both get to this point, the game would be a tossup. As with NC State, I’m picking Memphis somewhat in an attempt to go against the grain. They strike me as a good value play.
Championship:
North Carolina State over Memphis
In the end, NC State just strikes me as too good for Memphis. They’re our overall favorite for a reason: Not only are they arguably the best team in the tournament, but if they don’t face Clemson, they’ve got a very favorable path, especially once they get past Hofstra. We’ll see what happens, but I’m picking Kevin Keatts’ team to win it all.
