Big Ten Ratings after Early Non-Conference Games

[PurpleWhiteBoy]

Team	Record	Torvik	Miya	KenPom	PWB - Pts	Sched
Illinois	6-1	10	17	14	23	11
Purdue	7-1	12	15	15	27	1
UCLA	6-1	16	16	17	38	14
Ohio State	5-2	22	21	18	30	6
Michigan	6-1	25	14	22	40	15
Oregon	8-0	29	20	19	10	2
Maryland	7-1	30	36	32	56	17
Michigan St	6-2	32	33	36	37	3
Wisconsin	8-0	34	31	29	11	7
Nebraska	6-1	40	37	47	17	8
Indiana	5-2	44	38	53	70	4
Iowa	6-1	47	44	45	49	13
Rutgers	5-3	48	70	70	94	10
Penn State	7-1	49	43	37	46	16
Northwestern	6-2	52	62	64	72	9
Washington	6-1	84	91	78	108	18
USC	5-3	94	102	108	114	5
Minnesota	6-3	97	86	100	125	12

The last column is strength of schedule relative to the Big Ten, using my ratings.
My systems don't place much weight on blowouts of inferior opponents.
Nor do they severely punish teams for not blowing out inferior opponents.
Nor do they care how good a team is supposed to be, based on last year's team or roster moves.
(In other words every season starts with all teams considered equal)
This shows up readily in the ratings differences.

Some examples...
Illinois beat Arkansas-Little Rock 92-34, a 37 point outlier. However, they lost to Alabama 100-87.
Maryland beat Canisius 108-37, a 41 point outlier. Not impressed.
You can see that Torvik and KenPom are fooled by that sort of thing.
On the flipside of that, unbeaten Oregon has an 83-81 win over #13 Alabama, a 78-68 win over #29 San Diego State and an 80-70 win over #41 Texas A&M.
The game that lowers Oregon's rating for Torvik and friends is an 80-70 win over Portland, where they were supposed to win by 40 or so.

Unbeaten Wisconsin beat #7 Pitt by 6, beat #65 UCF by 16 and #43 Arizona by 15. But they have a couple of clunkers, like the 3 point win over Texas-Rio Grande and the 21 point win over Chicago State.

UCLA has played one game against a team they weren't supposed to blow out. They lost that game to #44 New Mexico 72-64. However, their blowouts of Lehigh (85-45) and Southern Utah (88-43) have some systems thinking UCLA is a very good team. To me, they haven't given any serious evidence of that.

As the season plays out, these differences reduce and teams improve or decline, but treatment of outliers is important as it can be misleading (and hidden).
We'll see how things look at the end of December, when conference play begins in earnest.

 

[Hungry Jack]

The good news is that we can make plenty of hay in conference against higher ranked teams

 

[CappyNU]

Torvik has an additional ranking - Wins Above Bubble (WAB) - that focuses more on who you beat then how you beat them. For the conference, it looks like this:

Team	Torvik	WAB
Oregon	29	2
Wisconsin	34	9
Purdue	12	10
Nebraska	40	39
Illinois	10	40
Michigan St.	32	41
Washington	84	43
Michigan	25	47
Penn St.	49	60
Maryland	30	62
Iowa	47	64
Ohio St.	22	65
Indiana	44	66
UCLA	16	77
Northwestern	52	80
Rutgers	48	115
USC	94	144
Minnesota	97	165

Seems like WAB aligns closer to your ratings, PWB. Only real outliers look to be Washington, Nebraska, UCLA and Ohio St.

 

[PurpleWhiteBoy]

Torvik has an additional ranking - Wins Above Bubble (WAB) - that focuses more on who you beat then how you beat them. For the conference, it looks like this:

Team	Torvik	WAB
Oregon	29	2
Wisconsin	34	9
Purdue	12	10
Nebraska	40	39
Illinois	10	40
Michigan St.	32	41
Washington	84	43
Michigan	25	47
Penn St.	49	60
Maryland	30	62
Iowa	47	64
Ohio St.	22	65
Indiana	44	66
UCLA	16	77
Northwestern	52	80
Rutgers	48	115
USC	94	144
Minnesota	97	165

Seems like WAB aligns closer to your ratings, PWB. Only real outliers look to be Washington, Nebraska, UCLA and Ohio St.

Thanks for posting that, Cappy!
Torvik is great - don't get me wrong - he provides a ton of free data and I can only thank him for that.
His website is filled with interesting information and he has coding skills that I can only dream about.
Plus, he's making a living off this stuff and seems like a decent guy.

 

[Purple Pile Driver]

Team	Record	Torvik	Miya	KenPom	PWB - Pts	Sched
Illinois	6-1	10	17	14	23	11
Purdue	7-1	12	15	15	27	1
UCLA	6-1	16	16	17	38	14
Ohio State	5-2	22	21	18	30	6
Michigan	6-1	25	14	22	40	15
Oregon	8-0	29	20	19	10	2
Maryland	7-1	30	36	32	56	17
Michigan St	6-2	32	33	36	37	3
Wisconsin	8-0	34	31	29	11	7
Nebraska	6-1	40	37	47	17	8
Indiana	5-2	44	38	53	70	4
Iowa	6-1	47	44	45	49	13
Rutgers	5-3	48	70	70	94	10
Penn State	7-1	49	43	37	46	16
Northwestern	6-2	52	62	64	72	9
Washington	6-1	84	91	78	108	18
USC	5-3	94	102	108	114	5
Minnesota	6-3	97	86	100	125	12

The last column is strength of schedule relative to the Big Ten, using my ratings.
My systems don't place much weight on blowouts of inferior opponents.
Nor do they severely punish teams for not blowing out inferior opponents.
Nor do they care how good a team is supposed to be, based on last year's team or roster moves.
(In other words every season starts with all teams considered equal)
This shows up readily in the ratings differences.

Some examples...
Illinois beat Arkansas-Little Rock 92-34, a 37 point outlier. However, they lost to Alabama 100-87.
Maryland beat Canisius 108-37, a 41 point outlier. Not impressed.
You can see that Torvik and KenPom are fooled by that sort of thing.
On the flipside of that, unbeaten Oregon has an 83-81 win over #13 Alabama, a 78-68 win over #29 San Diego State and an 80-70 win over #41 Texas A&M.
The game that lowers Oregon's rating for Torvik and friends is an 80-70 win over Portland, where they were supposed to win by 40 or so.

Unbeaten Wisconsin beat #7 Pitt by 6, beat #65 UCF by 16 and #43 Arizona by 15. But they have a couple of clunkers, like the 3 point win over Texas-Rio Grande and the 21 point win over Chicago State.

UCLA has played one game against a team they weren't supposed to blow out. They lost that game to #44 New Mexico 72-64. However, their blowouts of Lehigh (85-45) and Southern Utah (88-43) have some systems thinking UCLA is a very good team. To me, they haven't given any serious evidence of that.

As the season plays out, these differences reduce and teams improve or decline, but treatment of outliers is important as it can be misleading (and hidden).
We'll see how things look at the end of December, when conference play begins in earnest.

I like the other ratings over your’s PWB!

 

[AdamOnFirst]

So how do yours perform vs the spread any moneyline compared to the others?