The bracket below is our best current prediction of the final NIT bracket. It is predictive of where things will end up. It is not a reflection of where things currently stand.
We aren’t currently running our full model. That will come soon, but in the meantime, here’s how this works.
If you notice any irregularities or have questions of any sort, please say something in the comments or on twitter: @joestunardi.
- If you want NCAA Tournament Bracketology, we have that too.
- Regions are ordered as follows: first overall seed; fourth overall seed; second overall seed; third overall seed.
- A single asterisk designates an exempt bid, awarded to the best available teams from certain conferences. A double asterisk designates an automatic bid, awarded to certain conference champions who lose in their conference tournament. We currently estimate five to be the likeliest number of automatic bids, and we’ve given these five to the conference favorites we estimate to be likeliest to receive them.
Last Updated: Preseason
Mississippi Region
| 1. Mississippi* |
| Virginia |
| 4. Northern Iowa* |
| Nebraska |
| 3. Louisville* |
| Ohio** |
| 2. Missouri* |
| New Mexico |
LSU Region
| 1. LSU* |
| College of Charleston** |
| 4. Louisiana Tech* |
| SMU |
| 3. UAB* |
| Georgia |
| 2. Northwestern* |
| Notre Dame |
Miami (FL) Region
| 1. Miami (FL)* |
| Wofford** |
| 4. San Francisco* |
| Nevada |
| 3. Seton Hall* |
| Syracuse |
| 2. Boise State* |
| Penn State |
VCU Region
| 1. VCU* |
| Appalachian State** |
| 4. Seattle* |
| Minnesota |
| 3. Arizona State* |
| South Carolina |
| 2. NC State* |
| High Point** |
Note: With the College Basketball Crown expected to take two teams each from the Big 12, Big East, and Big Ten, we have removed TCU, USC, Rutgers, Providence, UCF, and Butler from our projected field. In our projection, they would each be higher in line for an exempt bid than the recipient from their respective league.
What are your thoughts on Davidson’s NIT chances?
They were our twelfth team out as of this morning. That isn’t out of our model’s margin of error, but it’s not a great spot to occupy. Probably need a lot of opt-outs.
Kentucky and Penn State aren’t even close to .500, on what planet are they getting an NIT bid?? I know the rule has technically changed, but there’s no way they’re picking a team multiple games under .500…even if you project both teams to win out, UK still 2 games under .500 heading into SEC tournament and PSU 1 game heading into B1G tournament.
Meanwhile St. John’s will finish 1 game above .500 even if they lose out (I think they win at least 1 of PC and Hall, both at home), is slightly worse in NET than UK (admittedly significantly worse than PSU who is a NET darling for some reason), better SOR than either UK or PSU, has the same amount of Ws and a better Q1&2 win %age than PSU and UK (5-8, 5-11, 5-13, respectively), and has same # of Q3 losses as PSU. The ONLY thing really dragging SJU behind PSU and UK is it’s single Q4 loss.
Somehow that’s the difference between a 3 seed for PSU, a 5 seed for UK, and not even being in the NIT for St. John’s? I’m not buying it.
This is a fair critique. Without any real word on how sub-.500 teams will be treated, with little historic basis to build an estimation off of, and with fewer nonconference games this year, we chose to not explicitly include overall win percentage as a variable in our model’s selection formula. This might have been a mistake, but we chose to err on the side of simplicity.
The biggest thing dragging St. John’s down in our model’s eyes is its KPI. KPI doesn’t get a ton of attention, but it’s on at least the NCAAT committee’s team sheet, it’s shown a strong correlation with NIT seedings in the past, and it has SJU 94th right now (with PSU 79th and UK 76th, with none of the three expected to significantly change over the rest of the season, where the model does see SJU splitting these next two games and UK and PSU each losing one of their next three).
UK and PSU’s schedules are a lot stronger than that of St. John’s, so while SOR has St. John’s the best of the three, KPI and NET both prefer UK and PSU, and we have more data on how committees’ choices align with KPI/NET/SOR than we do for how they treat teams playing hard schedules and finishing under .500 with a shortened non-conference piece of the season. The model might be wrong on UK and PSU, but we really don’t have enough precedent to know, and it does have St. John’s right on the bubble if it’s a 32-team field.
Thanks for the feedback. Again, it’s a fair criticism—we just really haven’t found a good way to incorporate the optical piece of sub-.500 teams when by every other indication, their résumés line up with NIT résumés historically.
I saw that re: KPI. And I can’t speak to any historical trends, but I would imagine when every other metric on the team sheet falls between 63 SOR and 76 NET, and one falls at 94, a significant outlier, a committee (be it NCAA or NIT isn’t going to put as much stock in the outlier). I also think it’s just a weird year for metrics with limited or in some cases even no non-conference play. I mean look no further than Colgate’s rankings. I also think that’s particularly true with KPI because I’m not sure anyone believes a 15-9 Mercer or 19-7 Chattanooga teams outta the SoCon or 18-5 Wright State outta the Horizon or 12-5 Georgia State outta the Sun Belt or 16-3 UCSB outta the Big West is better than St. John’s this year, but that’s what KPI would have us believe…that the Johnnies are on par with 12-6 North Texas and 10-7 Santa Clara, both out of significantly weaker conferences.
That makes sense, but I’m not sure the impact of outlier ratings shows up in the way we think it does—the committee might write them off, but they still see things the ratings systems “see” (SOS, good wins, bad losses, etc.) and incorporate those. So the committees might not hold St. John’s back *because* of KPI, but our data from the last few years indicates the St. John’s KPI is a bad sign for them, whether because it influences the committee or because the committee sees the same things that hold back that KPI number.
Colgate’s NET ranking is wild, and is a big indictment of NET (very forseeable issue, too—nuts that the NCAA didn’t change the NET formula in anticipation of this exact scenario), but I’m not sure its KPI ranking would be much lower had it played nonconference games. Would depend what the games were, but 11-1 will do a lot for you in KPI and SOR, so they could schedule Binghamton five times and be in the same position in those. For what it’s worth, KenPom—which tracks very closely with Vegas lines, and Vegas lines make their oddsmakers a lot of money—would have Wright State favored over SJU on a neutral floor right now, with SJU/North Texas and SJU/UCSB close to tossups. Your point stands with Mercer/Chattanooga/Georgia State, though, and quite a few others.
We’ll see. Lot of games left to happen, too, and if the field does drop to 16 teams, it’ll change things further.
There’s a slight bit of difference between Kentucky and Western Kentucky.
Yes! At least someone noticed!
Totally Agree with You. Having at least a 500 record and Quad 1 Wins are both very important
Please note. That’s Western Kentucky not Kentucky
It’s been confirmed, 16 teams in NYC for the 2021 NIT, no automatic bids. That makes it very interesting. It should be very competitive without the small conference garbage teams that can’t win their conference tournament. I hated when the NIT did that, it watered the tournament down so much.
Dallas, not NYC.
Do you actually think there will be an NIT with all this clamor?
No. We obviously have no idea, but it seems really unlikely that the NIT will take place this year. That said, if the NIT doesn’t take place, we’re working on simulating a virtual NIT on this site. More to come, but check back if you, like our founder, love the NIT enough to want to follow a digital equivalent.
Would love simulations…..
They’re happening! Information on our home page. Virtual NIT beginning Tuesday, ending by Saturday. Selection Show today, bracket challenge will open later today as well.
i love the NIT as well, could you please make a bracket sometime today on your site. i would give Little Rock, North Texas, Prairie View, Eastern Washington and Akron the benefit of the doubt and not include them in your NIT bracket. Thanks very much for considering!!
Duquesne come on man. They should be in NIT as a lock big wins against St Louis, VCU and The Bonnies on the road no home gym all year. I have seen them as high as an NIT 5 seed on another site. Give me the RPI any day over the New NET rankings.
You’ll note that of those big wins, the ones against Saint Louis are the only ones that came over an NIT team or better. The committee might like the 21 (probably 22) wins. It won’t like the 279th-ranked nonconference strength of schedule. Their chances aren’t great, regardless of what you or I think. But they’re at least in the mix. Could make it with a win this week, should make it if they get two.
About South Dakota State
No love to Memphis, the actual #1 seed. 8… Ha ha ha.. FU
It sounds like Oregon St. might have a decent chance if they can beat Utah tomorrow. Last season they had their best conference record in almost 30 years and didn’t get a sniff by the NIT at 18-13, 10-8. The conference was way down. This season with the conference trending much stronger, they are in the conversation at 17-13, 7-11. I’m assuming if they beat Utah and upset Oregon (already beat them once), they would be solid for the NIT at 19-14?
Definitely would be in if they get through Oregon. Very good chance if they come back here against Utah. And even if they lose, they aren’t completely out of it—just need automatic bids to break right, and some favorable decisions by the committee.
Update: freshman Jarod Lucas nailed a 3 with 1.8 seconds left to give the Beavs a 71-69 win over Utah. Before the game some bracketologists had OSU as a NIT 5 seed.
How does Syracuse not make the field?
Syracuse is in one of the toughest conference how are they left out?
There are only so many spots. Syracuse’s résumé is very close in strength to that of St. John’s, but it just isn’t projecting to end quite as strong, as the committee evaluates such things. It’s close enough that our model does still have them slightly more likely to make the NIT than miss it, but they’d do well to take care of business tomorrow, and maybe win one more.
Syracuse is not in the ” toughest conference”.
Explain to me how Texas A&M is not in place of AR, AL and Tennessee.
A&M beat Tennessee going away at their place.
A&M beat Auburn at their place.
A&M beat Bama at their place.
A&M is one of 2 SEC’s teams with a winning road record, UK is the other
A&M has a winning conference record, UT, AL and AR do not.
Hey Bob, I wrote more on this yesterday a little further down the comments section, so you can read more there if you’d like. The short story is, while those pieces of information do favor A&M, the overall body of work favors those other three teams, in large part because the committees (at least the NCAAT committee, but likely the NIT committee too) are instructed to weight all results equally, regardless of when in the year they happened. Which means A&M’s disastrous non-conference portion of the schedule holds them back.
Explain to me how Texas A&M is not in place of AR and Tennessee.
A&M beat Tennessee going away at their place.
A&M beat Auburn at their place.
A&M is one of 2 SEC’s teams with a winning road record, UK is the other
A&M has a winning conference record, UT and AR do not.
What would Mizzou have to do to make NIT? Beat A&M, Auburn, and Miss St/LSU? I.e. go to SEC championship?
It’s hard to say with a lot of confidence, but some back of the envelope math makes it look like getting through Auburn might be enough to get them to the bubble, especially if automatic bids break their way. So much going on around them in that portion of the seed list, though. A lot can change.
How about Oregon State? They have wins over Oregon and Stanford.
They came into tonight 28% likely, by our model’s measurement, which is close enough to say they’ve got a solid shot right now, if not a great one. Certainly on the bubble. Beating Utah might be enough. Keep a close eye on automatic bids.
How Tennessee, Alabama and especially Arkansas, and no Texas A&M? They had one really bad loss and a few questionable losses early, but finished above these other teams in conference. Even beat Auburn on the road.
The Fairfield, Harvard, and Temple losses really hurt Texas A&M, especially because the Harvard and Temple losses both came by double digits. They also have a poor non-conference strength of schedule (184th right now), compared to Arkansas (12th), Alabama (53rd), and Tennessee (16th), which makes the 6-6 non-conference record look even worse. Arkansas’s worst loss is to A&M. Tennessee’s worst loss is to A&M. Alabama’s worst loss is to Penn, who’s over 100 places better than Fairfield in every ranking system the NCAA Tournament Selection Committee uses. It’s amazing they’ve recovered as much as they have, and they deserve a ton of credit, but their NIT hopes look slim.
Your arguments are compelling. Very interesting statistics that I (or much of general fandom) don’t get exposure to. I recognize that body of work for the season is important, it just seems that more emphasis should be given to where you end up. A team becomes a team during the season, not simply where they start. Regardless, in this instance, I can understand why A&M is given less consideration.
Thanks for the thoughtful reaction! I try not to personally opine on what the committees “should” do—I don’t think my opinion on it’s all that important. Just trying to accurately predict what they will do, and understand/communicate why.
Would the NIT take into consideration UNC’s injuries and the fact that they are playing much better now and have most of their best players back (Anthony, Brooks, Robinson)? 3 straight wins (NCSU, at Syracuse, and Wake Forest) and then a loss at Duke. They also have 7 games decided by 6 points or fewer. They have lost at least 3 on buzzer beaters, so they haven’t been stomped in several of their losses.
I know 13-18 is a stretch and they would need to win at least 2 ACC tourney games.
To be honest, I don’t know. It’s definitely an unusual situation. I know the NCAA Tournament committee is told they can consider injuries, so I’d assume the same is true for the NIT committee, but the only things of any value I can offer are that 1) our model treats all teams as equally healthy to one another, 2) our model would have projected 2018 Notre Dame (for whom Bonzie Colson was notably injured for much of the season) to be seven spots lower on the seed list than they ended up, and 3) our model is currently projecting UNC’s median overall seed to be 15 spots or so below the necessary cut line to make the NIT. That median overall seed is achieved when they beat Virginia Tech but lose to Syracuse. So, if they get past Syracuse, the gap starts to close. Lot of other factors happening around them that have an impact, but that’s the general picture.
What about Murray St. ? co-champs of the ovc? I mean Belmont lost last year and got a out large bid to the NCAA. I think the matrix still favors the P5 teams. Its so incredible that a 23-9 team that has a strong fan base and great tradition to be left out.
There are a few big differences between Murray State this year and Belmont last year. Belmont only lost five games. Murray State’s lost nine. Belmont’s KenPom ranking, which very accurately predicts how well a team will play given their performance so far and who they’ve played, was 54th on Selection Sunday. Murray State’s is currently 134th. Belmont’s non-conference strength of schedule finished last year 75th in the country. Murray State’s this year is currently 345th. That last piece isn’t necessarily Murray State’s fault—it’s possible they had a hard time getting good teams to agree to play them—but all those losses, and their best win coming at home over a Belmont team that’s not as good as last year’s, are just too much for them to be selected.
Murray State deserves an NIT bid. This is a young, talented team that plays extremely hard and has a great fan base.
Agreed the Racers also tied for regular season championship and only got the 2 seed because of an archaic tie breaker system.
Where do you have TCU??
Good question. They were only 11.5% likely to make it this morning, so they probably either need a very low number of automatic bids or a very impressive showing against Baylor to have much hope.
You acknowledge that Sub .500 teams cant make the NIT, yet you still have Minnesota in? And where is Syracuse? 17-14 but they have 6 road wins including UVA and Notre Dame.
I hope I don’t have a typo somewhere in these comments—sub-.500 teams can certainly make the NIT, so my apologies if I was unclear about that. We don’t know if the committee will take them, because it’s fairly uncharted territory, but our understanding is there’s no rule against it, so our model treats sub-.500 teams the same as everyone else (while still evaluating all those losses, of course). As for Syracuse: the model has them as a tossup right now. They do have six road wins, but only the UVA one is really impressive, and the seven home losses (two of which are in Quadrant 3) hold them back.
How do you see odds to make NIT?
They’re on our College Basketball Probabilities page! Sorry for not having that linked more clearly. I’ve added a link to it above the bracket, right after automatic bids are mentioned. Thanks for asking!
What % does St Johns have to make the NIT? And would they be 100% with a win Wednesday vs GTown?
What % does St Johns have to make the NIT? And would they be 100% with a win Wednesday vs GTown? Thank you for the info!
Thanks for visiting the site! They’re 49.7% NIT-likely and 3.7% NCAAT-likely as of this morning. Those numbers are on our College Basketball Probabilities page, and they usually update around 11 AM Eastern Time every morning if you want to keep tabs on them.
Sorry, didn’t fully answer the question. They wouldn’t be 100% likely with a win over Georgetown (the NIT at-large field might be too narrow for any team to hit 100% with our model’s error margins), but they’d be a lot closer.
UNC should be in NIT if they win two games in ACC.
The model had them 12% likely to make the NIT entering today, which implies it would probably take three victories to get them to the bubble. Even then, their overall win-loss record could be a problem, since the committee hasn’t taken any sub-.500 teams since the rule prohibiting taking such teams was abolished (to our understanding—it’s hard to find official statements on all this).
The WCC has 3 teams advancing to the NCAA Tourney but you don’t show 4th place Pacific or 5th place USF in your NIT bracket. With their WCC Tourney win over Pacific last night, USF should definitely make the NIT and Pacific is probably on the outside looking in on the NIT bubble.
That isn’t really how selections work. It depends more on how strong teams are themselves, and how strong their résumés are, than whether or not there are good teams in their conference who happen to be directly in front of them in the conference standings. It’s not impossible that San Francisco gets a bid, but it’s very unlikely at this point, and they’ll need a strong performance against Gonzaga tomorrow (if not a win) to have much legitimate hope.
The three other NIT Bracketology sites that I could find all show USF making the NIT this year, and one of them has USF in as a 5-seed. Plus, just going back 5 years, I found that the WCC has always had at least one NIT participant, including last year’s mediocre San Diego squad at 21-14. To me, USF should be included in the NIT this year, even if they lose to Gonzaga on Monday by 20 points, as expected – that’s how strong the WCC has been this year as a conference.
I try not to get into who I think should or shouldn’t make the NIT, since my opinion isn’t very important. As for what the committee will do, though, all I can tell you is that our model sees San Francisco very differently from whichever site has them as a 5-seed. One of us is wrong, of course, and I don’t know the track record of whichever site that is. For the ones with San Francisco as a six, seven, or eight-seed, just a heads up that some NIT bracketology sites don’t include automatic bids in their projections. Ours does, just to give a more accurate picture of where the cut line will be. Personally, I’d love to see the Dons do well—Todd Golden’s use of data is fun to see—but our model doesn’t at this point see their résumé as one the committee will choose, regardless of how many teams the WCC has sent recently.
Joe,
Now that UConn is a lock for the NIT in all models, it seems like they actually WON’T need the bid. If they win today at Tulane (7 pt favorites) and go 2-1 in the AAC tourney, they are DANCING! They’ll be 21-13 with a NET, BPI, POM, and SAG all in the mid 40s likely (currently they are 59, 54, 50, and 55 respectively) I’m thinking that’d give them a solid 11 seed and first four game. If they go 3-1 and make the AAC finals to finish 22-13, I’m thinking they possibly dodge the first four game. Might be a bit aggressive, but if they get a little help with favorites winning conference tourneys, it could be a surprise at-large tourney birth for the ‘Huskies in 2020!
One thing to keep in mind—KPI and SOR have been better predictors of seeding since the quadrant system was added to team sheets than KenPom, BPI, and Sagarin. There’s definitely upside for them if they can make a run in the AAC Tournament, but your scenarios sound fairly ambitious for a team that entered today 77th in SOR, and I don’t think it’s safe to call them a lock—they’re far from a lock in our model, though escaping this one would be a good start, especially if the projected number of automatic bids shrinks at all.
They survived Tully on the road yesterday. Good point about KPI and SOR, but I’m assuming both committees look at how teams are playing currently, aside from overall metrics and the full body of work. UConn has gone 8-2 down the stretch, and if they do go 3-1 in the AAC tourney, they’ll have finished 11-3 in the last month of the season. I’m hoping it will be enough for either a top seed in the NIT or an 11/12 in the dance. Championship week is here!
Winning yesterday was really important for their chances—our model wasn’t taking it for granted going in. I don’t know if this is an explicit part of the NIT selection rules, but the NCAA Tournament Selection Committee is explicitly told to count all games the same, regardless of when they occur. Unfortunately for UConn, that means March counts the same as November, and it might explain why you see their chances so differently from how our model sees them (there’s some evidence recent performance influences decisions, no matter what the committees say, but we haven’t dug into it enough to have built it into our model). Thanks for continuing to visit the site! If UConn does win three games this week, they very well might climb out of the NIT.
college basketball is just like wrestling already pre determined who wins and who losesor the players point shaving anyways i hate basketball since alabama is non competitive in basketball and needs to play in a weaker conference like the big south or better yet quit playing the sport all together shut down the basketball program and tear down coleman basketball is for the nerds anyways bama don’t need basketball anyways i just wished march madness would go away for good
You should go back under your bridge troll.
Joe
Any chance you would be interested in betting paychecks that UConn doesn’t crack the NIT? Even if they lost to Houston they are still a lock. Have a little common sense please!
I would love to bet the paycheck I receive for doing this (you’ll notice we don’t have ads yet). Let me know if that’s a serious offer. Great eROI.
More seriously, one thing this model lacks is “common sense.” Statistical models are, by definition, entirely objective. They don’t weigh subjectivity. More often than not, this is an advantage—they point out flaws in the conventional wisdom. Sometimes, though, it can be a disadvantage—the conventional wisdom can be correct. We might see which is the case with UConn. We also might not, because it’s possible they’ll play themselves into a position in which the model agrees with “common sense.” Thanks for visiting the site, Adam. Hold onto your paycheck. You don’t want mine.