*****ColleyvilleAg06 Bubble Watch*****

856 Views | 11 Replies | Last: 6 min ago by rhutton125
ColleyvilleAg06
How long do you want to ignore this user?
AG
Thoughts on Tuesdays win I still cant believe we were down 9 at half and by the 8 minute timeout I felt completely comfortable and no stress about cruising to a win. The win didn't do much from a metrics standpoint but good to avoid a loss, watch a few more teams around us take an L and keep chugging towards some huge quad 1A opportunities down the stretch to clinch a 2 seed or potentially even challenge for a 1.

If you are looking for a quick primer on A&M's seeding, my prognostication is roughly a finish in the last 7 games would be equal to the following seeds (subject to the rest of the national landscape so +/- 1 seed line

7-0 or 6-1 = 1 seed
5-2 = 2 seed
4-3 = 3 seed
3-4 = 4 seed
2-5 = 5 seed
1-6 = 7 seed
0-7 = 9 seed

Given we are comfortably in the field, I will keep this BW a bit more brief than prior years and focus the analysis on those near A&M's seed and around the last 4 in/out and the rest just list a ranking, but happy to talk more about each individual team in the more detailed posts.

This is just to get the conversation started. I will add some more detail on those near the cut line tomorrow.

1 Seeds

1) Auburn (Lock) clear #1 with an insane 13 quad 1 wins (no one else has more than 8)
2) Alabama (Lock) 6-1 in quad 1A games and a pretty clear #2
3) Duke (Lock) outstanding predictive numbers
4) Florida (Lock) Lacked the non conf SOS and high end elite wins that the rest of the SEC had, until the last week winning at Auburn and at Mississippi State to for now claim the 4th one seed.

2 Seeds
5) Tennessee (Lock) Even after losing on Tuesday night Tennessee has both a better resume than A&M and clearly better predictive #s. But the gap has closed considerably

6) Texas A&M (Lock) Outstanding resume numbers (better than even Duke) but lagging predictive numbers (for a 2 seed). Ultimately it feels like A&M will get rewarded for going out of conference and racking up several high end wins on a neutral court (Purdue, Texas Tech, Creighton).

7) Houston (Lock) Predictive metrics darling (#2, #1, #3), it will be interesting to see where the committee has them during the sneak peak reveal on Saturday morning. There are 38 teams that have at least 2 quad 1A wins this season. Houston is not one of those teams. Their lone Q1A is at Kansas which is great but beyond that…they have built their resume on pounding bad teams

8) Purdue (Lock) Very similar overall profile as A&M, A&M has a bit better on the resume numbers, Purdue has better predictives. I am glad A&M owns the head to head on a neutral (Indy) court.

3 Seeds
9) Iowa State (Lock) 7-4 vs quad 1, 9th in SOR, solid overall resume but still behind the top 8, the last team that feels currently in the same league as A&M if this was selection Sunday
10) Arizona (Lock) great predictives, subpar resume #s, didn't do much in non conference but has been elite in the Big 12
11) Michigan (Lock) - rocketing up the seed list particularly after knocking off Purdue, predictive numbers wont give them consideration for anything more than a 3 seed right now
12) Kentucky (Lock) 7 quad 1A wins ties Auburn for the most in the country. That number alone hides a lot of warts with this resume which otherwise would be a few spots lower.

4 Seeds
13) Wisconsin (Lock)
14) Texas Tech (Lock)
15) Kansas (Lock)
16) St. John's The loss to Villanova on Wednesday knocks them down a full seed line and they only have 1 quad 1 game left on the schedule


5 Seeds
17) Marquette
18) Ole Miss
19) Illinois
20) Michigan State

6 Seeds
21) UCLA
22) Missouri
23) Memphis
24) Clemson

7 Seeds
25) Louisville
26) UConn
27) Maryland
28) Creighton

8 Seeds
29) St. Mary's
30) Mississippi State
31) Baylor
32) Oregon

9 Seeds

33) Gonzaga
34) Utah State
35) New Mexico
36) Nebraska

10 Seeds
37) Ohio State
38) West Virginia
39) San Diego State
40) Drake (Missouri Valley Placeholder if they lose and drop out it does NOT open up a slot for someone else as MVC will still have a bid)


11 Seeds
41) Vanderbilt
42) Oklahoma

Last 4 in
43) SMU It is very rare a "power" conference team will have NCAAT worth numbers (top 45 across the board in resume and predictive, yet has not yet scored a single quad 1 win. But then again is the ACC really still a power conference?

44) Arkansas 3 quad 1A wins against Michigan, Kentucky and Texas all away from home, for now have the pigs in. Just 4-7 in the SEC, and if the magic number is 7 wins it is hard to find 3 more wins on this schedule without stealing one somewhere.

45) BYU pretty good predictive numbers for a bubble team, 3 quad 1 wins look nice but not all quad 1 are created equal UCF, West Virginia and Baylor don't exactly move the needle.

46) Georgia similar boat as Arkansas. The wins over Kentucky and St. John's are nice but the losses are piling up and at 4-8 in the SEC you have to squint a bit to see where they can pick up at least 2 more wins needed to have a case.


Projecting 22 one bid leagues regardless of conference tournaments, for spots 47-68 (will be adjusted for stolen bids and some at large bids above this line will go away)

One bid leagues
47) Atlantic 10 -note, several bubble teams that are out now could work their way in to make this a 2 bid league
48) Atlantic Sun
49) America East
50) Big Sky
51) Big South
52) Big West Note, 2 bubble teams that are OUT right now but could possibly win out and get at large consideration
53) Conference USA
54) Coastal
55) Horizon
56) Ivy
57) MAAC
58) MAC
59) MEAC
60) NEC
61) OVC
62) Patriot League
63) SWAC
64) Southern
65) Southland
66) Summit
67) Sun Belt
68) WAC


--------------------------------------------
69) Texas The horns have much better predictive number than Georgia, Arkansas or even Vanderbilt but at a certain point it has to translate to wins and so far there haven't been enough of them for Texas. Really only 3 good wins to hang their hat on vs. A&M, Mizzou and at Oklahoma and not much even in the decent win category. The path here is to win the 3 @South Carolina, Georgia and Oklahoma, if they can do that they will be in the hunt. Given their lack of results in non conference, the magic number for Texas to get a bid may be 8 SEC wins however which means they may need 1 more in addition to those 3.

70) Wake Forest took a horrible loss on Wednesday night to Florida State on their home court. Those are the kind of losses that get brought up on Selection Sunday as the reason you are headed to the NIT. The resume numbers are not bad…the predictives are horrendous.


71) VCU
72) Indiana
73) North Carolina
74) Kansas State
75) UC San Diego
76) UC Irvine if either UC San Diego or UC Irvine win out until the tournament they would potentially have a chance to sneak in
77) San Francisco
78) George Mason
79) Cincinnati
80) USC
81) Boise State
82) UCF
83) Pitt
84) Xavier
85) Arizona state
86) Dayton
87) Villanova
88) Santa Clara
89) Iowa
90) TCU
91) Rutgers
92) Oregon State
93) Colorado State
94) Georgetown




These teams are not really in the same category as those above them, there is a considerable gap in profile strength but they are still technically alive since they have so many elite wins left on their schedule if they were to win out, but effectively there is no realistic path. Not eliminating them quite yet, since they can technically get to an at large

95) Northwestern
96) LSU
97) Utah
98) Washington
99) Penn state
100) Minnesota
101) Oklahoma State
102) South Carolina
ColleyvilleAg06
How long do you want to ignore this user?
AG
This is your primer for some of the frequently used abbreviations and what is important to the committee. The first thing to know is there are 6 key metrics for the committee, 3 are based on your resume and 3 are "predictives"
Resume Numbers these are based entirely on who you play, where and who you beat, with slightly different formulas and weightings given.
The 3 Resume numbers are:

KPI Key Performance Index
SOR Strength of Record
WAB Wins above Bubble (how many more wins you have vs. a team right at the bubble cut line would be expected to have against your schedule)

Predictive Numbers as the name suggests they can be used to predict future performance, but more importantly they are data driven metrics based on much more than W/L in each game and they analyze things like offensive efficiency, defensive efficiency, FG%, scoring margin, and even game control. So if we run that 20-0 run play in the 1st half and then cruise from there it is a lot more impressive than losing all game and then going on a big run late to win it. So sure you can (and institutions like Vegas do) use them to predict future performance, but more importantly they are a deeper evaluation of how good teams are based on data.

The 3 predictive numbers are:
BPI Basketball Power Index
KenPom Rating System
Torvik Rating System

A&M all year has not dominated teams, has gone on 2nd half runs, and won close games, that tends to play very well with the resume numbers and much less so on predictive. Think of it this way a close win at home against Arkansas will result in a positive move up the resume numbers (the win is all that matters) and probably a slide down on the predictives. A close loss to Auburn or at Florida would be the oppositive, the predictives would go up, the resume ranks would go down based on taking on a loss.
These 6 numbers are not the ONLY thing the selection committee looks at. They also look at record against the various quadrants, non conference strength of schedule, head to head outcomes, road & neutral wins, etc.
ColleyvilleAg06
How long do you want to ignore this user?
AG
At this point in the season, most teams numbers are pretty well baked, we have had 24 data points and unless the 7 remaining are drastically different, you shouldn't see a whole lot of movement up or down. So if we see, for example a 4-3 finish to the season, I would expect these numbers to look almost identical on Selection Sunday. If we finish 5-2 or 3-4 that will look a bit different…but not much. At 6-1 or better or 2-5 or worse and you may see a much more different numerical strength of the resume.
Here is a look at the trend line for A&Ms metrics for the ones I have handy

January 3rd (prior to conference play)
KenPom 16
Torvik 19

January 5th (after Texas blowout win)
KenPom: 11
Torvik: 17

January 21st after a 3-2 start to SEC play
BPI: 19
KenPom: 18
Torvik: 20

January 24th after Ole Miss win
BPI 18
KenPom 15
Torvik 19

January 30th after beating Oklahoma the 2nd time
KPI 9
SOR 12
WAB 7
BPI 17
KenPom 16
Torvik 19

February 2nd after winning at South Carolina
KPI 5
SOR 6
WAB 4
BPI 17
KenPom 14
Torvik 18

February 7th before Mizzou win
KPI 6
SOR 7
WAB 5
BPI 16
KenPom 14
Torvik 17

February 10th after Mizzou win
KPI 6
SOR 7
WAB 5
BPI 16R
KenPom 14
Torvik 17

Current: 2/13
KPI 6
SOR 6
WAB 5
BPI 14
KenPom 13
Torvik 16
panhandlefarmer
How long do you want to ignore this user?
AG
Awesome! Been waiting for this!
Topher17
How long do you want to ignore this user?
AG
Thanks for putting this together. You always do a great job!

I posted the same in the other games thread, but the cut line for the SEC teams involved will be fascinating to watch and I'm sure glad we aren't involved!
Texas and Vandy might come to regret their non-con scheduling at 7-11, while I'm not sure OU and Georgia can win enough games. Arkansas may have the best chance just based on remaining schedule and I don't love their chances.
Jonah2012
How long do you want to ignore this user?
Thanks for all your work on this!

It's interesting to think what would the SEC prefer. All of the top five teams trade wins with each other and take care of the bubble teams and have five 1 and 2 seeds and less teams in. Or those bottom four all sneak some wins in and get 14 teams in, regardless it's the strongest conference ever it seems.
Marsh
How long do you want to ignore this user?
AG
I'm curious how the seeding will actually look and if the SEC will take a collective reduction in seed # to prevent so many possible #1 vs #2 seed matchups being from the same conference.

Hasn't the committee historically done everything in their power to prevent a #1 and #2 seed in the same bracket being from the same conference? That's impossible with potentially 5 SEC teams in the top 8. The pessimist in me says one of the five arbitrarily gets dropped to a 3 for no reason other than optics...
WhataMaroon88
How long do you want to ignore this user?
AG
Marsh said:

I'm curious how the seeding will actually look and if the SEC will take a collective reduction in seed # to prevent so many possible #1 vs #2 seed matchups being from the same conference.

Hasn't the committee historically done everything in their power to prevent a #1 and #2 seed in the same bracket being from the same conference? That's impossible with potentially 5 SEC teams in the top 8. The pessimist in me says one of the five arbitrarily gets dropped to a 3 for no reason other than optics...
We are in rare territory as far as the number of high seeds from one conference. However, 2/3 is along the same bracket line. A 2/3 seeds plays a 1 seed in the Elite Eight regardless, so no, it won't matter in this case. We would likely get lined up with one of the SEC teams, but if there was only one SEC 2 seed, perhaps we'd be put in Duke's bracket.
Sterling82
How long do you want to ignore this user?
Thanks for the rundown, particularly on the various metrics.
ColleyvilleAg06
How long do you want to ignore this user?
AG
Jonah2012 said:

Thanks for all your work on this!

It's interesting to think what would the SEC prefer. All of the top five teams trade wins with each other and take care of the bubble teams and have five 1 and 2 seeds and less teams in. Or those bottom four all sneak some wins in and get 14 teams in, regardless it's the strongest conference ever it seems.


More teams = more money. I am sure they would love it if Kentucky lost in Austin this weekend.
ColleyvilleAg06
How long do you want to ignore this user?
AG
The NCAA bracketing rules state that a conferences top 4 teams cannot be in the same region if they are in the top 4 seed lines. There is no rule preventing a 1 and a 2 both being in the south region if that 2 seed is the 5th best team from the SEC.
rhutton125
How long do you want to ignore this user?
Refresh
Page 1 of 1
 
×
subscribe Verify your student status
See Subscription Benefits
Trial only available to users who have never subscribed or participated in a previous trial.