Showing posts with label Bracket analysis. Show all posts
Showing posts with label Bracket analysis. Show all posts

Friday, March 9, 2018

Bubble Profiles- Notre Dame


So the ACC Tournament has been on ESPN, so you probably have not heard the talking heads shut up about Notre Dame making the tournament.  This is certainly an argument one can make.  Notre Dame, despite a seven game losing streak, was still a good team even without Colson, Harvey, and Farrell.  They just didn’t have the ability to finish games anymore.  The Irish have remained in the Top 40 in efficiency in KenPom, Top 40 Rated by Sagarin, and they started 3-0 in the ACC. I had them as my preseason favorite to win the ACC.  But injuries happen, but honestly, it is more than just injuries. Some of it is just luck, which the Irish have had little to none of this season, which is something that will get covered here.

The Irish Profile:
15 games without Colson (6-9)
5 games without Farrell (1-4)
1 game without Pflueger (0-1)
15 games (and counting, so really shouldn’t roll into their case, or then we have to put Maryland in the field) without Harvey (7-8)

The Irish have not played well since the injury bug hit.  But what about before?
Win @DePaul (Q3, not useful but a road win to bank)
Win vs LSU (Neutral, Q2, LSU was much better than expected)
Win vs Wichita St (Neutral, unbelievable steal of a Q1 here, as ND almost never led, and had 3% chance of winning with 21 seconds left)
Bottom Line: they won the Maui, which Michigan, Marquette, and Wichita State cannot claim (though with Cal and VCU down, this was a shit field… the fact that Chaminade did not finish last is also evidence of that).
Then:
Loss at Michigan St (Q1, OK)
Loss vs Ball St (WHAAAA, OK I get it, buzzer beater, but still Q3)
Loss vs Indiana (also, not good, Q3)
They sneaked past Dartmouth too.

Uh, that looks like an NIT team.  So now the luck factor.  Here is how Pomeroy defines his “Luck Rating”:
Luck – A measure of the deviation between a team’s actual winning percentage and what one would expect from its game-by-game efficiencies. It’s a Dean Oliver invention. Essentially, a team involved in a lot of close games should not win (or lose) all of them. Those that do will be viewed as lucky (or unlucky).

I like to look at it as the team with the ball last is most likely to win, just like football.  If your offense averages more than a point per possession, you have to be feeling pretty good in a tie game.  The Irish struck the pot of gold against the Shockers, but had been snake eyes since.  Buzzer beater to Ball State, OT to Indiana (98.5% chance of winning), 1 pt loss to UNC, OT loss to Louisville.  I eyeballed the chart a month ago and saw them hanging around -.120 in the luck category, indicating that they had probably coughed up 3 wins along the way (standard deviation is actually around .051).  They finished the season at .055, which is a standard deviation out side the norm, but really only cause to believe they LOST one game by a bad break.

The 41 team Bubble (that reaches down and grabs “not gonna happens” like Davidson and Maryland, as well as some auto qualifiers) 3rd in basic efficiency, and 4th in efficiency strength.  This reflects a very good team.  But reaching back to the Maryland example, they are 15th and 12th here, and not even in the discussion.  Their Quadrant Win value is 36th, which means their results profile like a mid-major.  Their case is similar to Penn State’s actually, with the injury wrinkle.  

I have them a few spots out right now, with no way to make up ground, and Nevada potentially stealing a spot, as well as potentially an A-10 team. Duke probably would have been enough, but as it stands, Ball State at home with the whole team intact killed them.

Saturday, March 3, 2018

3 MARCH Bracket












Pods

1
Virginia



Kansas
1


San Diego

 16
Texas Southern



Hampton
16


Boise

16
Radford







Nashville
Pittsburgh







Wichita

Detroit

8
Butler



Middle Tennessee
8


Wichita

9
Southern California



Arizona State
9


Dallas










Pittsburgh

4
Kentucky



Clemson
4


Charlotte

13
Vermont



Buffalo
13



Boise







Boise



5
Texas Christian



Nevada
5




12
Louisiana



Kansas State
12









Missouri
12






ATLANTA

OMAHA




Last 5 Safe

3
Tennessee



Auburn
3


Florida State

14
East Tennessee St



Northern Kentucky
14


Oklahoma
Nashville







Dallas

Baylor

6
Ohio State



Houston
6


NC State

11
Oklahoma



Florida State
11


Louisville












7
Miami



Creighton
7


Last 5 In

10
Saint Bonaventure



Louisville
10


Missouri
Nashville







Detroit

Kansas State

2
Cincinnati



Purdue
2


Alabama

15
Charleston



Stephen F. Austin
15


Texas










Providence




SAN ANTONIO
















First 5 Out

1
Xavier



Villanova
1


Mississippi State

16
Niagara



Saint Francis PA
16


Notre Dame






Florida Gulf Coast
16


UCLA
Detroit







Pittsburgh

Oklahoma State

8
Seton Hall



Florida
8


Marquette

9
Saint Mary's



Virginia Tech
9













Next 5 Up

4
Gonzaga



West Virginia
4


Penn State

13
South Dakota State



Murray State
13


Syracuse
San Diego







San Diego

Nebraska

5
Texas Tech



Michigan
5


Georgia

12
Providence



New Mexico State
12


Utah

12
Alabama











LOS ANGELES

BOSTON




The Rest

3
Wichita State



Michigan State
3


Western Kentucky

14
Montana



Bucknell
14


LSU
Dallas







Wichita

Washington

6
Arizona



Texas A&M
6


Saint John's

11
Texas



NC State
11


San Diego State










Boise State

7
Arkansas



Rhode Island
7




10
Baylor



Loyola
10



Charlotte







Charlotte



2
Duke



North Carolina
2




15
Santa Barbara



Penn
15




So, Nebraska's stay in the bracket lasted all of 1 day. A close win that would have maintained their KenPom and Sagarin numbers could have kept them on the cut line, but like Western Kentucky, their trend rating plummets with no way to make it up.  They need a lot of help from Providence, Alabama, Texas, and Boston College.  Penn State beating Purdue today would not help, as the Big 10 isn't getting 6 bids.

Of course the MAAC and Big South abandons all convention and the top 2 teams are dead coming out of there.  Iona  could squeeze out a 15, but those other potential winners aren't getting off the 16 line. My preseason picks had Manhattan and Liberty, but I'm sliding in Niagara and Radford for today.

Rhode Island stands pat at 7... The Bonnie could totally end up with a better seed if they beat them again in the A-10 tourney.  And Davidson could make it a 3-bid conference.  They are playing better than anyone right now.

The quadrant formula seems to work, but like everything else, it is flawed because of arbitrary cut lines.  I think I may tweak it to reduce the value of Quad 2/3/4 a little.  Right now it is .9/.3/.15/.075 for wins and .2/.4/.6/.8 for losses.  Rewards for getting wins, particularly big wins, but does punish for losses, and can't boost a resume by getting your teeth kicked in by Q1... need at least 25% to push the needle.