Quad One Wins Minus All Losses : Power Rating

[visitingcock]

These are the only 23 teams that DO NOT have more losses than Quad One wins. I looked at 2022 stats. It shocked me to see how few teams had Q1 Wins that exceeded or equaled all losses.

Anyway, here's 2023 through Sunday.

1. tOSU 7 -0 +7
-----------------------
2. SCAR 6 -0 +6
-----------------------
3. UConn 6 -2 +4
-----------------------
4. Stanford 4 -1 +3
4. Indiana 4 -1 +3
4. Utah 4 -1 +3
------------------------
7. Columbia 5 -3 +2
8. UCLA 4 -2 +2
9. Duke 3 -1 +2
10. Michigan* 5 -3 +2
11. LSU 2 -0 +2
-----------------------
12. NDame 3 -2 +1
13. Arizona 3 -2 +1
13. Gonzaga 3 -2 +1
-------------------------
15. NC State 3 -3 0
15. Iowa St. 3 -3 0
15. Baylor 3 -3 0
15. FSU 3 -3 0
15. Va Tech 3 -3 0
15. Villanova 3 -3 0
21. Kansas 2 -2 0
22. St. Johns 1 -1 0
22. MTSU 1 -1 0

 

[Bone Dog]

LSU should be +2, right?

I must admit to being puzzled by the category “quad 1.” Can you help me understand it?

 

[visitingcock]

LSU should be +2, right?

I must admit to being puzzled by the category “quad 1.” Can you help me understand it?

Oops. You are right. I'm prejudiced. Correction made.

Quad one is basically the top quarter of wins against teams ranked 1-75 . It gets jacked around some whether you are playing them at home, on the road or at a neutral site. Consider them "power wins." I used NET quads for my ranking. If you used RPI quads it would be slightly different.

Quadrant 1 (Q1)
Home (1-30)
Neutral (1-50)
Away (1-75)

Quadrant 2 (Q2)
Home (31-75)
Neutral (51-100)
Away (76-135)

Quadrant 3 (Q3)
Home (76-160)
Neutral (101-200)
Away (136-240)

Quadrant 4 (Q4)
Home (161-361)
Neutral (201-361)
Away (241-361)

See: The NET Nitty Gritty Report for NCAA Women's College Basketball | WarrenNolan.com

 

[dabears17]

LSU should be +2, right?

I must admit to being puzzled by the category “quad 1.” Can you help me understand it?

Quad 1 wins are wins that are:
home games against Net top 30
neutral site games against Net top 50
away games against Net top 75

 

[Bone Dog]

Quad 1 wins are wins that are:
home games against Net top 30
neutral site games against Net top 50
away games against Net top 75

Thanks

 

[PoppaBear]

I'm no expert on this system but shouldn't Maryland be on this list?

 

[visitingcock]

I'm no expert on this system but shouldn't Maryland be on this list?

You are correct. They have 4 Quad One wins and three losses, so they are net +1

 

[Gamecockfan77]

Is this how the committee looks at these? It's an awkward way to express the statistic

 

[visitingcock]

Is this how the committee looks at these? It's an awkward way to express the statistic

quad one record is a factor, but that's only part of a teams sheet analysis.

Here are the team sheets: https://www.warrennolan.com/basketballw/2023/net-teamsheets-plus

Here is the "nitty gritty report" on all teams last year (which is a more abbreviated form Quad record breakdown) FOR LAST SEASON. The NET Nitty Gritty Report for NCAA Women's College Basketball | WarrenNolan.com

 

[Bone Dog]

Is this how the committee looks at these? It's an awkward way to express the statistic

I agree. It might be more revealing (at least in a rough and ready way) in the early part of the season to look at the ratio

Q1 + Q2 : Q3+ Q4

without even factoring in wins and losses. This would also be a way of measuring strength of conference by schedule.

Thus, as of today (thru 1/13), you get:
UConn 12:3

UNC 11:5
Creighton 11:5
UCLA 12:6
Duke 10:5
ND 10:5

NC St 11:6
Villanova 11:7
Indiana 10:6
Iowa St 9:5
Stanford 10:6
Indiana 10:6

FSU 11:8
Michigan 10:7
VPI 10:7
Iowa 8:7
Utah 8:7
Oklahoma 8:7
Colorado 9:8
Tennessee 10:9

tOSU 8:9
Maryland 8:9
Alabama 8:9
SCar 7:10

Arizona 3:11

and of course,
LSU 3:14

They can be grouped according to broader ratios. For example, UConn's group is 4:1, the next group (with UNC, Duke, ND, etc) is 2:1 or better, the next is 1.5:1 or better, then 1:1 or better, then less than 1:1, less than 1:3, and finally less than 1:4.

If you factor in W/L records, the list begins to reflect more what we come to expect, with SCar and Stanford closer to the top of the list.

 

[visitingcock]

I agree. It might be more revealing (at least in a rough and ready way) in the early part of the season to look at the ratio

Q1 + Q2 : Q3+ Q4

without even factoring in wins and losses. This would also be a way of measuring strength of conference by schedule.

Thus, as of today (thru 1/13), you get:
UConn 12:3

UNC 11:5
Creighton 11:5
UCLA 12:6
Duke 10:5
ND 10:5

NC St 11:6
Villanova 11:7
Indiana 10:6
Iowa St 9:5
Stanford 10:6
Indiana 10:6

FSU 11:8
Michigan 10:7
VPI 10:7
Iowa 8:7
Utah 8:7
Oklahoma 8:7
Colorado 9:8
Tennessee 10:9

tOSU 8:9
Maryland 8:9
Alabama 8:9
SCar 7:10

Arizona 3:11

and of course,
LSU 3:14

They can be grouped according to broader ratios. For example, UConn's group is 4:1, the next group (with UNC, Duke, ND, etc) is 2:1 or better, the next is 1.5:1 or better, then 1:1 or better, then less than 1:1, less than 1:3, and finally less than 1:4.

If you factor in W/L records, the list begins to reflect more what we come to expect, with SCar and Stanford closer to the top of the list.

Do Quad 3 and 4 really matter other than for showing WEAKNESS of schedule? ( Your SOS formula really points out Arizona and LSU’s shortcomings from the pre conference season

For my methodology of ranking, you just ignore all Quad 2,3,4 wins but count all loses. I’ve thought about expanding it to Quad 1 and 2 wins because so few teams have winning records of just Q-1 minus losses. But I’m lazy.

 

[GamecockFam]

Do Quad 3 and 4 really matter other than for showing WEAKNESS of schedule? ( Your SOS formula really points out Arizona and LSU’s shortcomings from the pre conference season

For my methodology of ranking, you just ignore all Quad 2,3,4 wins but count all loses. I’ve thought about expanding it to Quad 1 and 2 wins because so few teams have winning records of just Q-1 minus losses. But I’m lazy.

I think it all depends on the intended goal. I'm no statistician but I would think that the lack of data points makes the data less reliable.

A random thought, but I wonder if there is a correlation between teams in the NCAA tournament with statistical outlier Quad 3/4 losses and upsets in the tourney. Like, my running hypothesis is that teams that have good Quad 1/2 win ratios with a couple of head-scracher Quad 3/4 losses (i.e., Iowa last season) have a historically higher statistical likelihood of being upset in the tourney by a lower seed

 

[UcMiami]

I have always thought that quality of opponent should be considered on a sliding scale ... For a team aspiring to be a top ten team bad losses start around 15 and meaningless wins start around 25 while for a team rank around 25 the meaningless win column starts around 50. It is all relative to the terrain around where a team is ranked. LSU with its best win I believe being Arkansas at #38 has a 0-0 record against 'quality' teams for a program ranked in the top 10 and by years end they will have played two teams in the teens and 1 team in the top 10 - a really miserable schedule compared to pretty much every other ranked team.

 

[DefenseBB]

Being an analytical geek, Kudos to @visitingcock for putting this together!

 

[Ladeeeeda]

I think it all depends on the intended goal. I'm no statistician but I would think that the lack of data points makes the data less reliable.

A random thought, but I wonder if there is a correlation between teams in the NCAA tournament with statistical outlier Quad 3/4 losses and upsets in the tourney. Like, my running hypothesis is that teams that have good Quad 1/2 win ratios with a couple of head-scracher Quad 3/4 losses (i.e., Iowa last season) have a historically higher statistical likelihood of being upset in the tourney by a lower seed

Good question...I would be curious to know this as well!

 

[visitingcock]

Just for fun, I'm going to print an analysis of last year Q-1 wins minus losses compared to the tournament round of 32 qualifiers.

The only real fly in the ointment was . . . you guessed it - Creighton. But they blew all the sober predictions out -hums and computers.

 

[tigers4ever]

I think it all depends on the intended goal. I'm no statistician but I would think that the lack of data points makes the data less reliable.

A random thought, but I wonder if there is a correlation between teams in the NCAA tournament with statistical outlier Quad 3/4 losses and upsets in the tourney. Like, my running hypothesis is that teams that have good Quad 1/2 win ratios with a couple of head-scracher Quad 3/4 losses (i.e., Iowa last season) have a historically higher statistical likelihood of being upset in the tourney by a lower seed

I think that is probably accurate and is why LSU gets the high net ranking under this formula. They beat the teams they were supposed to beat and beat them by a lot. I know LSU's SOS bothers many but Mulkey may have unintentionally stumbled onto a possible loophole in the system.

 

[Ladeeeeda]

I think that is probably accurate and is why LSU gets the high net ranking under this formula. They beat the teams they were supposed to beat and beat them by a lot. I know LSU's SOS bothers many but Mulkey may have unintentionally stumbled onto a possible loophole in the system.

I would tend to think she does it intentionally.....but who knows

 

[tigers4ever]

I would tend to think she does it intentionally.....but who knows

 

[tigers4ever]

It prob was intentional but it's wild to think that's LSU's SOS with no losses gets a net ranking of 2 or 3.

 

[Gamecockfan77]

LSU's schedule is in the top half now so it's not that surprising.