PDA

View Full Version : AP & Coaches Poll: Weekly Update Thread



thespywhozaggedme
12-24-2018, 08:29 AM
http://www.espn.com/mens-college-basketball/story/_/id/25608294/duke-blue-devils-reclaim-no-1-ap-poll-kansas-jayhawks-drop-5th

If we win out I can't see how we are not a one seed.

GoZags
12-24-2018, 09:24 AM
And up to #3 in KenPom, behind Duke and Virginia.

JPtheBeasta
12-25-2018, 08:38 AM
Will there be 4 teams each with a better body of work than GU? It’s very possible with Duke, Michigan, Virginia, Tennessee and Kansas in the mix. North Carolina and Kentucky could put things together. Nevada could have an excellent record by come March. I agree that the Zags have put themselves into a decent position but you have to at least consider that other team’s could end up with better resumes.

Bogozags
12-25-2018, 06:28 PM
I don’t know if that is possible UNLESS we dominate conference play once Tillie gets back into the grove again...

I do believe we could metriculate back to #1 as teams above us lose games...

MontanaCoyote
12-26-2018, 07:55 AM
And up to #3 in KenPom, behind Duke and Virginia.

Can some math major school me on this. What should I make of these Kenpom stats?:

Virginia #2 Kenpom with AdjEM of 27.93. BUT.....#270 NCSOS Adj EM -3.02. How does Kenpom statistically reconcile such a high #2 Adj EM ranking with a pretty miserable #270 strength of schedule NCSOS AdjEM -3.02 ranking.

To this philosophy major seems that Virginia has the record they do because they’ve primarily played what many posters
refer to as cupcakes. But Kenpom has them as the second best team in the nation, at least “on paper.”

What am I missing here? What gives? PS Almost 60 years removed from college classrooms so don’t be too harsh on me.

amaronizag
12-26-2018, 08:25 AM
Pomeroy did away with his Pythagorean winning percentage and replaced it with his AdjEM. AdjEM is the difference between offensive and defensive efficiency. AdjEM is the number of points a team would be expected to outscore the average D-I team over 100 possessions. If GU’s offensive efficiency is 10% higher than the national average and Denver’s defensive efficiency is 10% higher than the national average, then GU’s offense when playing Denver would be expected to be 20% higher than the national average. If GU has an offense than is 120% of the national average and Virginia has a defense 80% of the national average, then GU’s offense should be exactly average when it plays Virginia.

bartruff1
12-26-2018, 08:30 AM
As Barbie so often says...." Math is hard....lets go shopping "......

BobZag
12-26-2018, 09:26 AM
As Barbie so often says...." Math is hard....lets go shopping "......

+1

gonzagafan62
12-26-2018, 09:46 AM
Can some math major school me on this. What should I make of these Kenpom stats?:

Virginia #2 Kenpom with AdjEM of 27.93. BUT.....#270 NCSOS Adj EM -3.02. How does Kenpom statistically reconcile such a high #2 Adj EM ranking with a pretty miserable #270 strength of schedule NCSOS AdjEM -3.02 ranking.

To this philosophy major seems that Virginia has the record they do because they’ve primarily played what many posters
refer to as cupcakes. But Kenpom has them as the second best team in the nation, at least “on paper.”

What am I missing here? What gives? PS Almost 60 years removed from college classrooms so don’t be too harsh on me.

https://kenpom.com/blog/ratings-explanation/

Read the whole thing. You cannot skew numbers according to his numbers and methodology..cupcakes and that sort of thing actually hurt more than help.

Very insightful, and comes from Ken himself

bartruff1
12-26-2018, 10:46 AM
No matter how many times you multiply or divide by a variety of factors you still have to make assumptions about the value of those factors...that skews the numbers....you can run a sensitivity analysis to test the validity of those assumptions and make adjustments...I see he does that...but it is still a guess... he says it is not a rating system...it is a predictive model ….likely more precise than accurate....in Physics it is the.... law of uncertainty...

Right now, some 75% or more of equities are traded according to predictive algorithms ….you will go broke betting on them if you have a investment horizon longer than a week..today was a perfect example...

The proof is in the puddng...is ken pom any better at picking winners and losers and point spreads than the odds makers at Vegas ???

I don't know....I am asking if anyone has compared the two...

DixieZag
12-26-2018, 11:56 AM
No matter how many times you multiply or divide by a variety of factors you still have to make assumptions about the value of those factors...that skews the numbers....you can run a sensitivity analysis to test the validity of those assumptions and make adjustments...I see he does that...but it is still a guess... he says it is not a rating system...it is a predictive model ….likely more precise than accurate....in Physics it is the.... law of uncertainty...

Right now, some 75% or more of equities are traded according to predictive algorithms ….you will go broke betting on them if you have a investment horizon longer than a week..

The proof is in the puddng...is ken pom any better at picking winners and losers and point spreads than the odds makers at Vegas ???

I don't know....I am asking if anyone has compared the two...

I would hope he is better at it, since the odds makers aren't even attempting to predict the actual spread, they will move the line however much they need in order to keep the betting as close to 50-50 as possible. So long as their line ensures they get close to their 10%, they couldn't care less about actual metrics or whatever.

MontanaCoyote
12-26-2018, 11:59 AM
Thanks All for responding to my Kenpom inquiry. I’ll study up (as opposed to study down) on what you’ve posted hoping to “get it.” I will try real hard!
Best,
MC

bartruff1
12-26-2018, 12:46 PM
I would hope he is better at it, since the odds makers aren't even attempting to predict the actual spread, they will move the line however much they need in order to keep the betting as close to 50-50 as possible. So long as their line ensures they get close to their 10%, they couldn't care less about actual metrics or whatever.

If he is....and you bet his predictions.....you could be very very wealthy....I wouldn't suggest you do that.....the odds makers are... in fact.... trying to predict the opening line....you notice it seldom moves more than a point or so over a week or so....same with the over and under....they are the gold standard...no pun intended...

DixieZag
12-26-2018, 12:58 PM
If he is....and you bet his predictions.....you could be very very wealthy....I wouldn't suggest you do that.....the odds makers are... in fact.... trying to predict the opening line....you notice it seldom moves more than a point or so over a week or so....same with the over and under....they are the gold standard...no pun intended...

Obviously cautionary words are appropriate, and you're right. But, I have read that they know that certain "Names" that have huge fan bases will "price in" some of the line, even at the opening, "Notre Dame" being the "Gold standard" (double pun intended).

But, I get what you're saying. In most betting scenarios you get a crowd sourced intelligence that will be as close to accurate as any. Still, I think KenPom is used plenty in setting lines. They know what works and what doesn't, better than me anyway. :)

bartruff1
12-26-2018, 01:10 PM
I think Nate the Great at 538 may just have made a comparison of the various models and Vegas with the actual results....Vegas won..

WallaWallaZag
12-27-2018, 12:56 AM
I think Nate the Great at 538 may just have made a comparison of the various models and Vegas with the actual results....Vegas won..

i believe models like kenpom aren't designed specifically to predict outcomes of individual games...they are more like aggregate systems that are more focused on the whole season in its entirety...body of work that should be used for things like seeding purposes.

RenoZag
12-31-2018, 10:47 AM
http://www.espn.com/mens-college-basketball/rankings

Week # 9

AP: 7th Coaches Poll: # 8

Wisconsin's loss to Western Kentucky dropped them 7 spots in both polls. . .

jazzdelmar
12-31-2018, 11:30 AM
USF disappears. Saint John’s blew chance to crack the poll. Terrible call and foul shooting. Ponds stunk.


http://www.espn.com/mens-college-basketball/rankings

Week # 9

AP: 7th Coaches Poll: # 8

Wisconsin's loss to Western Kentucky dropped them 7 spots in both polls. . .

Zagceo
02-11-2019, 08:50 AM
AP Top 25






RK

TEAM

REC
PTS

TREND





1

https://a.espncdn.com/combiner/i?img=/i/teamlogos/ncaa/500/2633.png&w=40&h=40 (http://www.espn.com/mens-college-basketball/team/_/id/2633/tennessee-volunteers)
Tennessee (http://www.espn.com/mens-college-basketball/team/_/id/2633/tennessee-volunteers)(40)
22-1
1573

-





2

https://a.espncdn.com/combiner/i?img=/i/teamlogos/ncaa/500/150.png&w=40&h=40 (http://www.espn.com/mens-college-basketball/team/_/id/150/duke-blue-devils)
Duke (http://www.espn.com/mens-college-basketball/team/_/id/150/duke-blue-devils)(24)
21-2
1554

-





3

https://a.espncdn.com/combiner/i?img=/i/teamlogos/ncaa/500/2250.png&w=40&h=40 (http://www.espn.com/mens-college-basketball/team/_/id/2250/gonzaga-bulldogs)
Gonzaga (http://www.espn.com/mens-college-basketball/team/_/id/2250/gonzaga-bulldogs)
23-2
1457

1





4

https://a.espncdn.com/combiner/i?img=/i/teamlogos/ncaa/500/258.png&w=40&h=40 (http://www.espn.com/mens-college-basketball/team/_/id/258/virginia-cavaliers)
Virginia (http://www.espn.com/mens-college-basketball/team/_/id/258/virginia-cavaliers)
20-2
1406

1





5

https://a.espncdn.com/combiner/i?img=/i/teamlogos/ncaa/500/96.png&w=40&h=40 (http://www.espn.com/mens-college-basketball/team/_/id/96/kentucky-wildcats)
Kentucky (http://www.espn.com/mens-college-basketball/team/_/id/96/kentucky-wildcats)
20-3
1330

-





6

https://a.espncdn.com/combiner/i?img=/i/teamlogos/ncaa/500/130.png&w=40&h=40 (http://www.espn.com/mens-college-basketball/team/_/id/130/michigan-wolverines)
Michigan (http://www.espn.com/mens-college-basketball/team/_/id/130/michigan-wolverines)
22-2
1253

1





7

https://a.espncdn.com/combiner/i?img=/i/teamlogos/ncaa/500/2440.png&w=40&h=40 (http://www.espn.com/mens-college-basketball/team/_/id/2440/nevada-wolf-pack)
Nevada (http://www.espn.com/mens-college-basketball/team/_/id/2440/nevada-wolf-pack)
23-1
1230

1





8

https://a.espncdn.com/combiner/i?img=/i/teamlogos/ncaa/500/153.png&w=40&h=40 (http://www.espn.com/mens-college-basketball/team/_/id/153/north-carolina-tar-heels)
North Carolina (http://www.espn.com/mens-college-basketball/team/_/id/153/north-carolina-tar-heels)
19-4
1178

-






















BPI




RK
TEAM
CONF
W-L (http://www.espn.com/mens-college-basketball/bpi/_/dir/desc/view/overview/sort/winpct)
BPI RK (http://www.espn.com/mens-college-basketball/bpi/_/dir/desc/view/overview/sort/bpirank)
SOS RK (http://www.espn.com/mens-college-basketball/bpi/_/dir/asc/view/overview/sort/sospastrank)
SOR RK (http://www.espn.com/mens-college-basketball/bpi/_/dir/asc/view/overview/sort/sorrank)
RPI RK (http://www.espn.com/mens-college-basketball/bpi/_/dir/asc/view/overview/sort/rpirank)


1
Gonzaga (http://www.espn.com/mens-college-basketball/team/_/id/2250/gonzaga-bulldogs)
WCC
23-2
1
77
7
--


2
Duke (http://www.espn.com/mens-college-basketball/team/_/id/150/duke-blue-devils)
ACC
21-2
2
14
1
--


3
Virginia (http://www.espn.com/mens-college-basketball/team/_/id/258/virginia-cavaliers)
ACC
20-2
3
23
2
--


4
Michigan State (http://www.espn.com/mens-college-basketball/team/_/id/127/michigan-state-spartans)
Big Ten
19-5
4
5
8
--


5
Tennessee (http://www.espn.com/mens-college-basketball/team/_/id/2633/tennessee-volunteers)
SEC
22-1
5
65
4
-

Zagceo
02-11-2019, 09:25 AM
NET breakdown with Quadrants Gonzaga has 1 remaining Q1 game and Duke has 5

https://i.imgur.com/SlLD1ZO.png

https://i.imgur.com/fVsZPgz.png

gonstu
02-11-2019, 09:32 AM
NET breakdown with Quadrants

https://i.imgur.com/SlLD1ZO.png

https://i.imgur.com/fVsZPgz.png

First thought: That's a lot of Quad 4 games. Thanks for the chart, it's awesome.