Bracketology – Bracket Projection

Bracket Projection

From games played through 2/9/18.

This projection is based on team ratings from my ranking website, TheRankerPit.com (whose ratings have proven better than ESPN’s BPI at picking against the spread). I used those ratings as a starting point and made adjustments based off of projections of what the selection committee will do.  I don’t yet attempt to avoid same-conference matchups in early rounds or give any consideration to regions.

Check out the Bracket Matrix to compare the bracketology projections of many bracketologists.

Bracket Projection

1 Virginia 1 Villanova
16 Rider 16 Penn
8 Nevada 8 Seton Hall
9 Arizona St. 9 New Mexico St.
5 Clemson 5 Texas A&M
12 Arkansas 12 Syracuse/UCLA
4 Gonzaga 4 Ohio St.
13 Murray St. 13 Loyola (IL)
6 Butler 6 St. Mary’s (CA)
11 St. Bonaventure 11 Marquette
3 North Carolina 3 Auburn
14 South Dakota 14 Northern Ky.
7 Houston 7 Louisville
10 Maryland 10 La.-Lafayette
2 Duke 2 Cincinnati
15 Montana 15 Bucknell
1 Michigan St. 1 Purdue
16 Hampton/Prairie View 16 Winthrop/UC Santa Barbara
8 Michigan 8 Rhode Island
9 SMU 9 Virginia Tech
5 TCU 5 Florida St.
12 East Tenn. St. 12 Texas/Southern California
4 Wichita St. 4 Tennessee
13 Vermont 13 Buffalo
6 Oklahoma 6 Creighton
11 Western Ky. 11 Florida
3 Kansas 3 West Virginia
14 Central Ark. 14 Towson
7 Kentucky 7 Arizona
10 Alabama 10 Miami (FL)
2 Xavier 2 Texas Tech
15 FGCU 15 Wagner

Overall Seeds

1 Virginia
2 Villanova
3 Purdue
4 Michigan St.
5 Duke
6 Cincinnati
7 Texas Tech
8 Xavier
9 North Carolina
10 Auburn
11 West Virginia
12 Kansas
13 Gonzaga
14 Ohio St.
15 Tennessee
16 Wichita St.
17 Clemson
18 Texas A&M
19 Florida St.
20 TCU
21 Butler
22 St. Mary’s (CA)
23 Creighton
24 Oklahoma
25 Houston
26 Louisville
27 Arizona
28 Kentucky
29 Nevada
30 Seton Hall
31 Rhode Island
32 Michigan
33 Arizona St.
34 New Mexico St.
35 Virginia Tech
36 SMU
37 Maryland
38 La.-Lafayette
39 Miami (FL)
40 Alabama
41 St. Bonaventure
42 Marquette
43 Florida
44 Western Ky.
45 Arkansas
46 Syracuse
47 UCLA
48 Texas
49 Southern California
50 East Tenn. St.
51 Murray St.
52 Loyola (IL)
53 Buffalo
54 Vermont
55 South Dakota
56 Northern Ky.
57 Towson
58 Central Ark.
59 Montana
60 Bucknell
61 Wagner
62 FGCU
63 Rider
64 Penn
65 UC Santa Barbara
66 Winthrop
67 Prairie View
68 Hampton

Projected Conference Champions – Automatic Bids

Conference Predicted Champion
America East Vermont
American Athletic Cincinnati
Atlantic 10 Rhode Island
Atlantic Coast Duke
Atlantic Sun FGCU
Big 12 West Virginia
Big East Villanova
Big Sky Montana
Big South Winthrop
Big Ten Michigan St.
Big West UC Santa Barbara
Colonial Athletic Association Towson
Conference USA Western Ky.
Horizon League Northern Ky.
Ivy League Penn
Metro Atlantic Athletic Rider
Mid-American Buffalo
Mid-Eastern Athletic Hampton
Missouri Valley Loyola (IL)
Mountain West Nevada
Northeast Wagner
Ohio Valley Murray St.
Pac-12 Arizona
Patriot League Bucknell
Southeastern Auburn
Southern East Tenn. St.
Southland Central Ark.
Southwestern Athletic Prairie View
Sun Belt La.-Lafayette
Summit League South Dakota
West Coast Gonzaga
Western Athletic New Mexico St.

Last Four In – the last four teams projected to earn at-large bids

  • Syracuse
  • UCLA
  • Texas
  • USC

First Four Out – the best four teams projected to be just short of getting at-large bids

  • Penn State
  • Oregon
  • Boise State
  • Mississippi St.

Next Four Out – the next four best teams

  • Missouri
  • MTSU
  • UNLV
  • NC State

52 thoughts on “Bracketology – Bracket Projection

    1. gotstremph Post author

      Great question! Virginia’s resume is a little bit better than NC State’s right now. The Duke win (by 2) is nice. NC State beat em by 22. I’ve been expecting Virginia losses, but they keep skating by on 1-3 pt wins. They’re my first team out right now. My bubble group of teams is large and could change a lot based on what happens in the next two weeks.

      Like

      Reply
      1. TJ

        This projection for Virginia is out of whack. Granted when you wrote it, NC State hadn’t been pounded by Duke.

        I understand that you were projecting Virginia to lose – but they keep winning. At some point, probably now, I think you need to adjust your methodology. Virginia is currently 22-7, 14-5 in the ACC (controlling their destiny for second place), with three wins over teams in the top 10 at the time they played, and ranked in both major polls.

        I’d be willing to bet no team with a resume like that has ever been left out of tournament.

        Like

      2. gotstremph Post author

        Thank you for your comment. I totally agree. I will have an update posted later today.

        Like

    1. gotstremph Post author

      Thank you for the comment. I appreciate your passion for SoCon and completeness. I’m in the minority by projecting UNC Greensboro to win the conference tournament. I’ve got them as a 13 seed. I used an alternate spelling to save space, but they’re in there.

      Like

      Reply
  1. John Smith

    This bracket is laughably bad sorry to say. Some of the teams you leave out are mind baffling (Arizona State). I looked at Bracket Matrix standings and this bracket has performed near the bottom of all bracketologists the last couple of years, so take this one with a big grain of salt folks.

    Like

    Reply
    1. gotstremph Post author

      Thank you for your comment and feedback. It looks like Arizona State is in a good spot, doesn’t it (currently 1st in the Pac-12)? According to my ratings and power scores, they’re not actually that good. I project they will lose 3 of their last 4 reg season games and lose to Colorado or Oregon in the quarterfinals of the Pac-12 Tournament. Selection Sunday is 19 days away. There is still enough time for AZ St to prove who they really are and lots of other bubble teams to finish strong.

      Like

      Reply
      1. John Smith

        This is a really poor answer. Have you even watched their games this season? If the program you’re using has contributed to you finishing bottom 5 in bracket matrix back to back years, it may be time for a refresh.

        Like

      2. gotstremph Post author

        Thank you for your concern. You’re right, I’m a below average bracketologist. My specialty and focus is analyzing the tournament, not predicting who will be selected. But I do aspire to improve my bracketology skills and hope that your criticism is not meant to discourage me from continuing my journey. Tread softly, John, because you tread on my dreams.

        Like

  2. HTPinUtah

    Objectively PIttsburgh is trending up and has a nice last 6 games. Q index is 78/0.67 and should improve. Q1 2/6; Q2 2/1; Q3 6/1; Q4 5/1. 1-2 bad losses (Nichols and Wake) but they may be in Q3 soon. 3/3 down the stretch and 1-2 wins in the ACCT (likely good wins)….4th-5th place in the ACC may get them in…Again – objective numbers are not bad.

    Like

    Reply
    1. HTPinUtah

      Some very nice OOC wins too.. FSU, KState, Northwestern, Rutgers. Beat UNC twice. Played Duke and Louisville close (should have won Louisville).

      Like

      Reply
  3. Mike

    How is Creighton both an 11 seed and first 4 out? Also interesting that you have Seton Hall nowhere in sight (and I believe you’re the only one on bracketmatrix that has them out right now)…I’m assuming those 2 things are related?

    Like

    Reply
    1. gotstremph Post author

      Great catch! Thank you, Mike. I’ve corrected my First Four and Next Four Out lists. A lot of positioning will get sorted out this week. I’m looking forward to seeing how the Big East Tournament plays out. Seton Hall is playing really well lately. Do you think they’ll avenge their 2OT loss to Georgetown on Thursday?

      Like

      Reply
      1. Bryan

        Why have St. Mary’s in the tournament? They don’t have a single win against another at-large team. They’re 3-9 against quadrant 1 and 2 teams while Seton Hall has 7 wins against at-large teams in your bracket and is 12-10 against Q1/Q2 teams.

        Like

      2. gotstremph Post author

        Thanks for the question, Bryan. A good case can be made for Seton Hall. It seems most bracketologists consider them a lock, which I assume would remain the case even if they lose to Georgetown Thursday night. Looking at Seton Hall’s rank in several of the key advanced metric rating systems gives me pause – especially their NET rank. NET: 62, BPI: 58, KenPom: 59, Sagarin: 51. St. Mary’s is ranked significantly higher in each one (NET: 37, BPI: 39, KenPom: 33, Sagarin: 42). I know the Selection Committee will not only be looking at Q1/Q2 losses, so I’m trying to take everything into consideration. Anyways, these are both bubble teams to me and I’ll be shuffling and sorting those teams until Sunday. I wouldn’t be surprised if St. Mary’s wins tonight and get’s an auto-bid.

        Like

      3. Mike

        I think they’re the better team, but Georgetown will be the more desperate team so it could go either way. I’m in agreement with the majority that Seton Hall is pretty much a lock at this point. I agree that their numbers that you listed below are slightly concerning, but they also have top wins that the majority of the bubble simply can’t compete with. If you look at their scores, they just play a lot of close games, which is related to why their efficiency numbers don’t look very impressive. And I know it’s no longer an official criteria, but their old RPI rank is 46 which would generally have them in fairly safe territory.

        Like

      1. gotstremph Post author

        Thanks for the comment, Oliver. I know. I agree it’s shocking and hard to believe that Gonzaga would be seeded so low. Thank you for reminding me how biased people (including many bracketologists and even Selection Committee members) can be.
        My metrics and models help eliminate bias, and this early in the season, I rely on them heavily. Gonzaga is not as good as they have been in recent seasons. Some of the other top WCC teams have improved, making an undefeated conference schedule less likely. I’m excited to see how Gonzaga does in the next two months, especially given what my models are saying at the moment.

        Like

  4. Henry J Muto

    Kentucky as a 3 seed was laughable. The committee almost gave them a 1 seed today. Probably should stick with ranking the teams as they stand now and not try and predict 6 weeks down the road you would come out looking better though your 47th ranked out of 60 newbies on the site so not sure that will help you out either. I scored more points than anyone 2 out of the last 4 years I don’t have a blog or get paid like Jerry Palm (who I have beat 13 out of the 15 years ) but I post my seed list on multiple sites before the selection Show each year.

    PS Nebraska lost their 2nd best player probably drop them from the bracket now.

    Like

    Reply
    1. gotstremph Post author

      Thank you for the comment. While my specialty and expertise lie in analyzing the teams once they’re in the tournament, I’ve still got a bit of room to grow as a bracketologist.

      Like

      Reply
  5. Henry J Muto

    Kentucky as a 3 seed was laughable. They have been a 2 seed for a while now and the committee almost had them as a 1 seed toady (5th overall).

    Like

    Reply
    1. gotstremph Post author

      That’s a great question, Matthew. Thank you for asking. After a 5-game losing streak, we should definitely be questioning Nebraska’s tournament-worthiness. My quick answer: I think Nebraska is a better team than their record shows, and I project them to finish 9-11 (or better) in the Big Ten, one of the strongest conferences this year. Other metrics are in line with my assessment of Nebraska (NET: 33, BPI: 22, POM: 27, SAG: 21).

      Not so quick answer: My bracket is a prediction of what the bracket will look like on Selection Sunday, not if the tournament started today. I mostly use my own team ratings to project future performance. My numbers indicate Nebraska is good enough to warrant a 6-9 seed, so even according to my own numbers, a 6-seed is probably too optimistic. So I’ll agree with you on that point. My methodology also keeps me from overreacting to a loss or even a recent string of losses. Most bracketologists overreact to “what happened yesterday.” Only 5 days ago, Nebraska was in 91 out of 99 brackets. Now they’re in 62 of 102. All that said, all models are wrong and sometimes very wrong. We have a month of basketball left and Nebraska can get back on track tonight against Maryland (they’re currently a 2.5 point favorite). Let’s see how it goes…

      Like

      Reply
    1. gotstremph Post author

      Great question, David. I use my statistical models to project how a team will perform between now and Selection Sunday. As we now are only a week and a half out, there is less uncertainty and I can focus more on what the committee will do. Your timing is perfect – I am shifting several projections tonight, and FSU will be a team with more movement than the rest (along with Miami, Butler, and Arkansas).

      Like

      Reply
      1. BleedinGold

        Your statistical models are flawed.

        During the Early Release by the committee on Feb 9th, Purdue was ranked as the Top 3 seed. Since then their only loss was on the road @Maryland. Which won’t hurt them considering everyone around them has since lost as many or more games than them….

        Does your model have them losing the last two (@Minn @NW), and then dropping the first game of the BTT? That’s the only way they even sniff a 5 seed.

        Like

      2. gotstremph Post author

        I agree. My model is flawed. I’d go so far as to totally agree with the great statistician George Box, who famously noted that all models are wrong (https://en.wikipedia.org/wiki/All_models_are_wrong). I’m not sure why the BracketMatrix was showing Purdue as a 5-seed. I’ve had them as a 4 and close to a 3 for over a week. I definitely don’t see Purdue falling below the 4 line, and their performance over the next two weeks will show whether they should be bumped up to a 3-seed. Thanks for the comment.

        Like

    1. gotstremph Post author

      Thanks for the question. I agree Washington has some good wins. ASU has now lost more conference games than they’ve won, so that one’s losing some luster. There are a lot of teams this year with multiple good wins. Looking past the good wins, they’ve had close games against much weaker opponents and too many of their losses were not close. Another thing hurting the Huskies is how the Pac-12 isn’t as strong this year as it often is. If they finish the season strong, they’ll definitely make the tournament. But their performance against some of the middle-range teams isn’t giving me a high level of confidence that they’ll finish strong. I’m excited to see how it plays out!

      Like

      Reply
    1. gotstremph Post author

      Good question, Brian. Thank you for your comment! 103 out of 108 brackets in the matrix have them in. If today was Selection Sunday, I’d have them in too, but we still have a month of basketball left. Providence has a tough final stretch, where they will most likely go 3-4, to finish the season 19-12 (10-8). They have some good wins, but so do lots of other teams. I consider scoring margin, and Providence had some close wins against weaker opponents (Belmont-1 pt, Rider-4 pts, Brown-5 pts, Stony Brook-2 pts). They’re sitting at a 10-seed in the matrix right now, and losing 4 of their last 7 and an early exit from the conference tournament would very well knock them out.

      Like

      Reply
      1. Brian

        Wow so you are banking on them losing 5 out of 8, even though they are 7-4 with the third toughest sos so far in the BE.
        Pretty ridiculous.. only 1 team from the big east has not gotten in with 10 wins in the last 5+ years… andy you think pc would be one even though they would finish the year with 5+ quad 1 wins, an rpi of around 40 and sos under 20.

        Okay so this isn’t worth my time.

        Like

      2. gotstremph Post author

        Probably more like 5 out of 9 (they’ll win in the first round of the Big East tournament). They might still make it. You’re right about the 10-win trend. Last year was the perfect example. 4 teams were 10-8 in conference and all made it in. 11-seed Providence lost their play-in game. 6-Creighton, 9-Seton Hall, and 10-Marquette all lost in the first round. Maybe the committee will think twice this time?
        Seriously, though, it’s really too soon to say Providence is a lock. Please check back in 4 weeks.

        Like

  6. TJ

    It is time to adjust your team rankings when you are consistently sliding Duke up to the top line. They lost to Virginia — you keep them on the top line. They lose to St. Johns — you keep them on the top line.

    Any ranking system that still projects them to be seeded ahead of Virginia is just wrong.

    Like

    Reply
    1. gotstremph Post author

      Thanks for the comment, TJ. I agree that I need to adjust. In fact, I was in the middle of updating when you commented. I disagree, however, that projecting Duke ahead of Virginia is absolutely wrong. There’s still a whole month of the season left before Selection Sunday. I’ll agree with “probably wrong” but not “just wrong.”

      Like

      Reply
  7. GPJ73

    Curious why you would have Old Dominion – with its zero wins over the top 100 RPI or KenPom – over teams like Alabama and Missouri … or, frankly, any of the other teams in your first 8 out. Their resume is *terrible*.

    Like

    Reply
    1. gotstremph Post author

      That’s a really good question. Thank you for bringing that up. This far away from Selection Sunday, my projections are still largely (though not entirely) based on my model’s team ratings, and Old Dominion scores well in a lot of the key team stats that go into that calculation. As we head down the stretch, I will shift more to a prediction of how the committee will make their selections. Old Dominion’s team rating is close behind MTSU’s rating for tops in C-USA, so I wouldn’t be surprised if ODU wound up winning the C-USA Tournament and got the automatic bid.

      Like

      Reply
    1. gotstremph Post author

      Great question, Ed. If Selection Sunday was yesterday, Virginia would’ve been given a 1-seed, without a doubt. This projection is my best guess on how the bracket will look on March 11th. Duke and Virginia are really close, which last week proved. My model helps look at the whole season and not put too much weight on any single game. Given the season performance of all teams to this point, Duke scores a little bit better than Virginia. I’m really excited to watch the ACC conference race down the stretch.
      Thanks for the question. Please check back later to see how my projections look as we get closer to Selection Sunday.

      Like

      Reply
  8. You can't Forecast

    I think you need a new computer. Seton hall is a lock and you have them as the. Next four out. LMAO…..every other site has them easily in. There are other laughable seedings in your bracket as well. Too many to mention. LOL

    Like

    Reply
    1. gotstremph Post author

      You can’t, thank for your comment! You’re totally right about my projection for Seton Hall. I based my projections (maybe too much) on my computer ratings for all of the teams. I think my model may need some tweaking to account more for strength of schedule. Nonetheless, playing Villanova close and winning by 9 @ Hawaii isn’t impressing me much. I think I might be the one laughing when Seton Hall makes their early exit from the tournament. Joking aside, I will be adjusting my bracket projection soon to how I think the committee will make their decisions. Stay tuned! And thanks, again, for the comment!

      Like

      Reply
      1. You can't Forecast

        Joking aside, both providence and Michigan state are getting in as well. also Houston is not making the tournament. I believe you have them as a 9 seed. Just a few more observations

        Like

Leave a Reply to chicago dude Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s