Skip to main content

Bracketology 2021

Selection Sunday is finally here.

From a very young age, I have been obsessed with March Madness. The games themselves are full of drama, excitement, Cinderallas, buzzer beaters, celebrations, and heartbreaks. But, it is also the structure of the tournament that has always intrigued me.

The bracket itself is a thing of beauty. It has an air of a mathematical work of art in the way that it reduces from 64 (or so) to 32 to 16 to eight and then finally to four, two, and then one. It is parade of powers of two, marching to a glorious conclusion.

For this reason, I has also long been obsessed with the concept of bracketology. First, it was in the hand-made paper brackets that I made as a child, and patiently filled out as each game game to its conclusion. Then it was the study of how the bracket is assembled, from the s-curve, to avoiding conferences rivals, to geographic optimization. Finally, it was the use of analytics to see if I could use math to dominate my office pool.

Last year, this was all cruelly taken away from us just nine days before the most holy day on the entire sports calendar. We have had to wait 24 long and difficult months. But, that magical day is here once again. It feels like a celebration. Just making it here almost feels like victory.

For the last several years, I have gone through my own exercise in bracketology. Yesterday and last night, as the results came in, I updated my seed like and rearranged my bracket. This morning, I wanted to show you what I came up with. In addition, I have a few notes on the biggest challenges in the construction of this year's bracket as I see it. This will give us a few things to watch for as the real brackets are rolled out.

So, here are my four projected NCAA Tournament region brackets, arranged from No. 1 to No. 4 based on the rankings of the No. 1 seeds. In the Final Four, region No. 1 (the "West") would face regions No. 4 (the "East"), while region No. 2 (the "Midwest") would face region No. 3 (the "South").





Key Questions:

1) Who fell off from the bubble yesterday?

With upset wins by Georgetown and Oregon State in the Big East and Pac-12 Tournament finals, those two teams both earned bids, which will cost two other teams bids. In my estimate, it is Colorado State and Wichita State (who was also upset by Cincinnati) who will lose out in this transaction. ESPN basically agrees, while CBS favors Colorado State over Utah State and Wichita State over Drake.

Georgetown and Oregon State will now likely enter the tournament as No. 12 seeds, which means that the bottom two No. 11 seeds will now be filled by the winners of the two First Four games.

There is also one more bid stealer lurking on Sunday: the Cincinnati Bearcats. If they knock off Houston in the American Athletic Conference this afternoon, one more team drops out of the field. I agree with both CBS and ESPN that this would be the other Mountain West team, either Colorado State (if they are not already out) or Utah State.

2) Where will MSU be seeded?

I believe that the Michigan State Spartans are right on the cusp between a high No. 11 seed and a low No. 10 seed. I have them as the top No. 11 seed right now, and safely out of the First Four. ESPN and CBS agree, but the Spartans are dangerously close to the First Four... and the NIT. 

While I still think that there is a s 99 percent chance that MSU is in the Tournament, with the upsets yesterday and the Spartans' NET of 70, I cannot rule out the idea that the Spartans somehow get left out. It is very unlikely... but it is still possible.

That said, I have MSU in the "Midwest" region with Illinois as the No. 1 seed. In my bracket, the Spartans would face No. 6 BYU in the first round with No. 3 Arkansas lurking. ESPN has MSU also as an No. 11 seed lined up to face No. 6 USC with No. 3 Texas in the same pod. CBS has the Spartans as a No. 10 seeds matched up with No. 7 Oregon with a possible match-up with No. 2 Alabama in the second round.

3) Who will be in the First Four?

In my bracket, the last four teams in are Drake, Louisville, Syracuse, and Utah State.  If Cincinnati beats Houston this afternoon, I would slide Utah State off of the bubble and out of the tournament and UCLA would play in the First Four. ESPN basically is in agreement with me. CBS, though, has the First Four as Wichita State, Syracuse, UCLA, and Colorado State in the First Four, with the Rams as the team that would get cut.

4) How will the committee rank the No. 1 seeds?

With Michigan and Baylor losing early in their conference tournaments, the overall rankings of the No. 1 seeds is in question. As the only undefeated team, Gonzaga has to be the top ranked No. 1 seed, but the final placement of the other three teams is in question. This does matter as it relates to which teams would meet in the Final Four round were they to advance.

I would personally place Illinois as the No. 2 overall seed if they beat Ohio State today. If not, I would place Baylor in that spot. As for Michigan, their loss to Ohio State on Saturday, and more importantly, the loss of Isiah Livers drops them to the lowest of the No. 1 seeds in my estimation.

To me, this alignment also has some practical implications. It would keep Michigan and Illinois from potentially facing each other until the final game and it does the same thing for Baylor and Gonzaga, who were considered the two best teams for most of the year. To this point, this arrangement feels correct to me, and it looks like ESPN and CBS came to roughly the same conclusion.

5) Who is the last No. 2 seed?

Selecting the top seven teams is pretty easy this year, but the eighth overall team (the weakest No. 2 seed) is a bit tougher. ESPN has Houston in that slot, and CBS has Oklahoma State, even though they lost to Texas in the Big 12 Final. I decided to aware the Longhorns with that spot due to that victory, but I am not super confident in that pick.

6) Who is the last No. 4 seed?

Similar to the weakest No. 2 seed, selecting the lowest ranked No. 4 seed is tricky. CBS and ESPN agree on Purdue, Virginia, and Florida State, but they also have West Virginia in that spot. That looks a bit too high to me, and I have Villanova in that place instead.

7) How high will Georgia Tech rise?

Just a few days ago, the Yellow Jackets were on the bubble, but this morning they are your ACC Tournament champions. How high of a seed will they receive? Both ESPN and CBS have Georgia Tech as a No. 9 seed. I have them a bit higher at No. 8. But, as we have discussed with MSU, the prospect of facing a No. 1 seed in the second round is no reward.

8) Where will the committee place BYU, Loyola, San Diego State, Saint Bonaventure, and VCU?

In a normal year, comparing the resumes of mid-majors to high major teams is tricky and in 2021, this job is even harder. I am really curious where BYU, San Diego State, Loyola-Chicago, Saint Bonaventure, and VCU will get seeded. 

I personally have BYU as a No. 6 seed, which is where ESPN has them, but CBS has the Cougars as only a No. 8 seed. I have Loyola as a No. 7 seed, but that is higher than either ESPN (No. 8) or CBS's placement (No. 10). As for the Aztecs, I also have them as a No. 7 seed, but ESPN has them higher as a No. 6. CBS, however, has them on the nine-line. 

VCU and the Bonnies still need to decide the Atlantic 10 Title this afternoon, which makes things even harder. ESPN just placed them both as No. 10 seeds and called it a day. I agree on VCU, but I have Saint Bonaventure as a No. 9 (assuming that they win). CBS have the two team both at one seed lower than where I have them.

9) What other teams are hard to seed?

There are a few other teams that I had trouble seeding. Clemson scores really high (No. 15) in my NEW results-based metric, but the Tigers are in the 40s in Kenpom and the NET.  As a compromise, I placed them as a No. 7 seed. CBS has Clemson as a No. 6 seed, while in ESPN, they are a No. 10 seed.

Texas Tech has a strong NET (No. 17) and Kenpom ranking (No. 24), but they score poorly in the NEW Index (No. 45) and as a result, I dropped them to a No. 7 seed. CBS agrees with this placement, but ESPN has the Red Raiders as a No. 5 seed.

Oklahoma also scores poorly in the NEW Index (No. 48) but unlike Texas Tech, the Sooners' NET (No. 37) and Kenpom ranking (No. 38) are also fairly weak. I still placed the Sooners as a No. 8 seed, which is in agreement with ESPN and CBS, but I wonder if this placement is too high. 

Similarly, Virginia Tech's metrics are not impressive to me. The Hokies rank No. 50 in Kenpom, No. 48 in the NET, and No. 42 in the NEW Index. CBS places them on the seven line, while ESPN has them as a No. 9 seed. To me, their metrics simply do not support that. I have Virginia Tech as a No. 10 seed.

10) How good of a job will the Committee do in avoiding conference rematches?

Perhaps the biggest criticism that I have had for the Selection Committee over the past few years is in the way that they have allowed conference teams to face each other in the early rounds of the Tournament. For example, in 2019, Michigan State played Minnesota in the second round and came very close to having the play Maryland, in Washington D.C. in the Sweet Sixteen.

The argument (i.e. excuse) has been that the Committee wants to prioritize reducing travel time for teams and fans, and sending MSU to Washington D.C. to face a Big Ten team is better than sending them to Spokane or Los Angeles. While there is some truth to that in most years, the tournament will be held exclusively in Indiana this year. So, there is no excuse for lazy bracketing (which is what this is in reality).

With nine total Big Ten teams almost certain to make the Tournament, it is mathematically impossible to completely prevent any conference foes from meeting prior to the Sweet 16 in at least one case. But, as I show above, it is not that difficult to avoid this type of rematch.

I placed No. 2 Iowa, No. 5 Wisconsin, and No. 9 Rutgers in the West. Wisconsin and Rutgers could, in theory, meet in the Sweet 16, but both teams would need to score upsets to get there. In addition, those teams did only play once in the regular season. Those two teams cannot meet Iowa until the regional final, and that is very unlikely.

The other three regions in my bracket each contain only two Big Ten teams and in none of the three cases can the Big Ten foes meet prior to the regional final. In the Midwest, I have No. 1 Illinois and No. 11 Michigan State. In the South, I have No. 2 Ohio State and No. 4 Purdue. In the East, I have No. 1 Michigan and No. 10 Maryland.

See, committee members? It's not that hard.

That is all for now. Enjoy Selection Sunday, everyone and Go Green.



Comments

Popular posts from this blog

Dr. Green and White Helps You Fill Out Your Bracket (2024 Edition)

For as long as I can remember, I have loved the NCAA Basketball Tournament. I love the bracket. I love the underdogs. I love One Shining Moment. I even love the CBS theme music. As a kid I filled out hand-drawn brackets and scoured the morning newspaper for results of late night games. As I got older, I started tracking scores using a increasing complex set of spreadsheets. Over time, as my analysis became more sophisticated, I began to notice certain patterns to the Madness I have found that I can use modern analytics and computational tools to gain a better understanding of the tournament itself and perhaps even extract some hints as to how the tournament might play out. Last year, I used this analysis to correctly predict that No. 4 seed UConn win the National Title in addition to other notable upsets. There is no foolproof way to dominate your office pool, but it is possible to spot upsets that are more likely than others and teams that are likely to go on a run or flame out early.

The Case for Optimism

In my experience there are two kinds of Michigan State fans. First, there are the pessimists. These are the members of the Spartan fan base who always expect the worst. Any amount of success for the Green and White is viewed to be a temporary spat of good luck. Even in the years when Dantonio was winning the Rose Bowl and Izzo was going to the Final Four, dark times were always just around the bend. Then, there are the eternal optimists. This part of the Spartan fan base always bets on the "over." These fans expect to go to, and win, and bowl games every year. They expect that the Spartans can win or least be competitive in every game on the schedule. The optimists believe that Michigan State can be the best Big Ten athletic department in the state. When it comes to the 2023 Michigan State football team, the pessimists are having a field day. A major scandal, a fired head coach, a rash of decommitments, and a four-game losing streak will do that. Less than 24 months after hoi

2023 Final Playoff and New Year's Six Predictions

The conference championships have all been played and, in all honesty, last night's results were the absolute worst-case scenario for the Selection Committee. Michigan and Washington will almost certainly be given the No. 1 and No. 2 seed and be placed in the Sugar Bowl and the Rose Bowl respectively. But there are four other teams with a reasonable claim on the last two spots and I have no idea what the committee is going to do. Florida State is undefeated, but the Seminoles played the weakest schedule of the four candidates and their star quarterbac (Jordan Travis) suffered a season ending injury in the second-to-last game of the regular season. Florida State is outside of the Top 10 in both the FPI and in my power rankings. I also the Seminoles ranked No. 5 in my strength of record metric, behind two of the other three candidates. Georgia is the defending national champions and were previously ranked No. 1 coming into the week. But after losing to Alabama in the SEC Title game,