(Far Too Early, Sure To Change A Thousand Times) 2012 F/+ Projections

by Bill C. on Feb 3, 2012 2:55 PM EST in 2012 College Football Season

Perhaps I was underestimating Joseph Randle and Oklahoma State (and overestimating the hell out of Boise State) in my Far Too Early Top 25 a month ago?

Today at the Mothership, I took a look at the way recent recruiting rankings might actually have an impact on the field in 2012. To further elaborate on the impact two-year recruiting rankings and returning starter data might have on F/+ projections, I felt that going all out to create a full Top 120 (yes, 120 -- not ready to address the new teams yet) might be the best thing to do. Even in saying that Ohio State, Indiana and Vanderbilt might improve a solid amount (or that Arizona and Penn State could regress) doesn't really give you a good glimpse of the big picture.

Below are some rough F/+ projections for 2012, taking into account only three factors: 2011 F/+ ratings, two-year weighted recruiting data and lost starters. These are, after all, the components to anybody's predictions/projections, math-based or not: how were they last year, who are they bringing back, and who are they bringing in?

What we will see as the offseason progresses are more refined ways of determining these pieces. Instead of simply looking at last year, we will see more recent history taken into account. We've determined that four- and five-year history are typically more predictive than simply looking at last year. Plus, there will be adjustments made to the Returning Starters figure based on the magnitude of the players lost/returning. For instance, Oklahoma State loses only six starters, but that list includes Brandon Weeden and Justin Blackmon, both of whom were far better than the average starter. Plus, we might find there is an even better way to take recruiting into account. (In other words, I cannot stress enough how "rough" these projections are. Plus, and this is also quite important: I haven't run any of this by Brian Fremeau yet.) But I'm a visual learner, and I thought the best way to look at the impact of incoming and outgoing talent would be to isolate these three variables.

(And yes, projections see plenty to like about 20 returning starters; I doubt the final projections have either Tennessee or Texas Tech ranked this high, but it certainly does emphasize the role of experience. At least, I hope it changes, as my alma mater's new place of residence -- the SEC East -- currently has six teams in the Top 33.)

Proj. Rk	Team	Conference	2011 F/+	Rk	2012 Recruiting F/+	Rk	2012 Returning Starters	2012 Projected F/+
1	Alabama	SEC	35.8%	1	33.0%	1	13	32.9%
2	LSU	SEC	34.1%	2	28.2%	8	14	32.6%
3	Oklahoma State	Big 12	27.4%	3	17.6%	28	16	28.5%
4	Oklahoma	Big 12	21.9%	7	24.8%	16	17	27.3%
5	USC	Pac-12	16.2%	11	30.9%	4	17	24.0%
6	Florida State	ACC	18.4%	8	31.0%	3	15	22.8%
7	Oregon	Pac-12	23.8%	5	26.8%	11	13	22.5%
8	Michigan	Big Ten	16.0%	12	24.9%	15	16	21.0%
9	Texas	Big 12	12.1%	19	31.5%	2	17	20.9%
10	Georgia	SEC	14.9%	15	29.0%	6	15	19.6%
11	Stanford	Pac-12	17.7%	9	25.7%	12	14	19.4%
12	Notre Dame	Independent	15.5%	13	25.1%	14	15	19.2%
13	Texas A&M	SEC	14.5%	16	21.1%	22	16	19.0%
14	Florida	SEC	7.8%	30	28.1%	9	18	17.9%
15	Arkansas	SEC	15.0%	14	18.3%	26	15	17.3%
16	Ohio State	Big Ten	5.9%	38	29.1%	5	18	16.5%
17	Michigan State	Big Ten	16.8%	10	15.0%	34	14	16.3%
18	Nebraska	Big Ten	10.1%	26	21.5%	21	16	15.6%
19	South Carolina	SEC	11.1%	21	24.6%	17	14	14.1%
20	Tennessee	SEC	1.0%	57	25.4%	13	20	13.9%
21	Auburn	SEC	2.5%	52	28.4%	7	18	13.6%
22	TCU	Big 12	14.2%	17	16.7%	32	13	12.9%
23	West Virginia	Big 12	10.6%	23	7.8%	45	16	12.9%
24	Baylor	Big 12	8.4%	29	9.4%	40	16	11.5%
25	Rutgers	Big East	7.2%	34	18.6%	25	15	11.2%

As always, full list after the jump.

Proj. Rk	Team	Conference	2011 F/+	Rk	2012 Recruiting F/+	Rk	2012 Returning Starters	2012 Projected F/+
26	Clemson	ACC	6.3%	36	27.6%	10	14	11.0%
27	Virginia Tech	ACC	11.1%	22	17.5%	29	13	10.7%
28	Utah	Pac-12	2.4%	54	15.5%	33	18	10.6%
29	Missouri	SEC	10.2%	25	11.5%	38	14	10.4%
30	Wisconsin	Big Ten	22.3%	6	8.0%	44	10	10.4%
31	South Florida	Big East	5.4%	40	3.2%	55	18	10.3%
32	Kansas State	Big 12	7.3%	33	-3.9%	66	18	10.2%
33	Vanderbilt	SEC	4.4%	46	6.5%	47	18	10.1%
34	Houston	C-USA	12.4%	18	-1.9%	62	14	9.0%
35	Georgia Tech	ACC	3.9%	48	8.4%	41	17	8.9%
36	Texas Tech	Big 12	-4.7%	76	22.2%	19	20	8.3%
37	North Carolina	ACC	3.7%	49	18.1%	27	15	8.3%
38	BYU	Independent	6.3%	37	-6.6%	73	17	7.5%
39	Louisville	Big East	4.7%	44	13.9%	36	14	6.7%
40	Miami	ACC	7.4%	32	19.1%	24	12	6.5%
41	Pittsburgh	Big East	4.3%	47	4.6%	51	15	5.8%
42	Cincinnati	Big East	9.6%	27	8.3%	43	12	5.6%
43	Mississippi State	SEC	5.1%	43	13.5%	37	13	5.3%
44	Florida International	Sun Belt	0.0%	60	-5.6%	70	19	4.9%
45	Boston College	ACC	-3.0%	69	5.3%	50	19	4.9%
46	Penn State	Big Ten	7.5%	31	10.8%	39	12	4.7%
47	Southern Miss	C-USA	11.4%	20	-4.7%	69	12	4.1%
48	Washington	Pac-12	-1.0%	64	21.8%	20	14	4.1%
49	California	Pac-12	4.7%	45	23.7%	18	11	3.7%
50	Iowa	Big Ten	5.1%	42	13.9%	35	12	3.7%
Proj. Rk	Team	Conference	2011 F/+	Rk	2012 Recruiting F/+	Rk	2012 Returning Starters	2012 Projected F/+
51	Louisiana Tech	WAC	5.5%	39	-16.5%	89	16	3.3%
52	Central Florida	C-USA	2.5%	53	-3.9%	67	15	2.4%
53	Tulsa	C-USA	6.8%	35	-13.3%	85	14	2.2%
54	UCLA	Pac-12	-6.7%	83	17.2%	31	16	1.2%
55	Ole Miss	SEC	-9.9%	95	17.3%	30	17	-0.2%
56	Illinois	Big Ten	-0.1%	62	-2.3%	63	14	-0.8%
57	Connecticut	Big East	-0.1%	61	-21.3%	98	17	-1.0%
58	Ohio	MAC	1.8%	56	-17.5%	91	15	-1.3%
59	Arizona State	Pac-12	5.2%	41	8.3%	42	10	-1.5%
60	N.C. State	ACC	-2.6%	67	-9.3%	75	16	-1.6%
61	Boise State	Big East	26.9%	4	-0.5%	60	6	-1.7%
62	Virginia	ACC	-1.4%	65	19.2%	23	11	-1.7%
63	Northern Illinois	MAC	3.0%	50	-12.9%	84	13	-2.3%
64	SMU	C-USA	1.8%	55	-3.4%	65	12	-2.8%
65	Duke	ACC	-5.2%	77	-5.7%	71	16	-2.8%
66	Maryland	ACC	-10.2%	96	6.0%	48	17	-3.0%
67	Purdue	Big Ten	-4.6%	73	-9.7%	78	16	-3.2%
68	Bowling Green	MAC	-4.7%	75	-24.3%	108	19	-3.2%
69	Oregon State	Pac-12	-7.4%	86	7.5%	46	14	-4.1%
70	Iowa State	Big 12	-3.9%	72	1.6%	57	13	-4.2%
71	Syracuse	Big East	-3.5%	71	-6.6%	72	14	-4.3%
72	Western Michigan	MAC	2.8%	51	-21.3%	99	13	-4.4%
73	Northwestern	Big Ten	0.6%	59	-7.0%	74	12	-4.4%
74	Wake Forest	ACC	-1.6%	66	-2.8%	64	12	-5.1%
75	Toledo	MAC	8.8%	28	-9.6%	77	9	-5.4%
Proj. Rk	Team	Conference	2011 F/+	Rk	2012 Recruiting F/+	Rk	2012 Returning Starters	2012 Projected F/+
76	Arizona	Pac-12	-5.5%	79	5.5%	49	12	-6.1%
77	Miami (Ohio)	MAC	-6.5%	82	-26.6%	116	18	-6.4%
78	Temple	MAC	10.3%	24	-20.5%	96	9	-6.9%
79	Utah State	WAC	-3.1%	70	-20.9%	97	14	-7.2%
80	Kentucky	SEC	-8.9%	89	2.5%	56	13	-7.8%
81	San Diego State	MWC	-2.9%	68	-10.3%	79	12	-7.8%
82	Western Kentucky	Sun Belt	-7.0%	84	-23.1%	105	16	-8.2%
83	Nevada	MWC	-0.8%	63	-19.1%	95	12	-8.2%
84	Washington State	Pac-12	-12.1%	100	-0.1%	59	15	-8.2%
85	Arkansas State	Sun Belt	0.8%	58	-9.6%	76	10	-8.7%
86	Indiana	Big Ten	-18.3%	112	1.3%	58	19	-8.8%
87	Minnesota	Big Ten	-11.0%	99	3.6%	54	13	-9.1%
88	Eastern Michigan	MAC	-10.4%	97	-25.1%	110	18	-9.2%
89	Marshall	C-USA	-7.9%	87	-0.8%	61	12	-9.3%
90	East Carolina	C-USA	-8.9%	90	-10.9%	80	14	-9.5%
91	Navy	Independent	-5.8%	81	-17.4%	90	13	-10.0%
92	Central Michigan	MAC	-12.4%	101	-22.7%	102	18	-10.3%
93	UL-Lafayette	Sun Belt	-5.5%	78	-13.6%	86	12	-10.4%
94	Hawaii	MWC	-4.6%	74	-17.5%	92	12	-10.7%
95	Ball State	MAC	-9.4%	93	-27.8%	118	16	-11.2%
96	Colorado State	MWC	-15.4%	105	-15.6%	87	18	-11.2%
97	Fresno State	MWC	-8.5%	88	-31.5%	120	16	-11.3%
98	North Texas	Sun Belt	-13.6%	102	-22.9%	103	18	-11.3%
99	Kent State	MAC	-9.4%	92	-25.3%	111	15	-11.8%
100	UL-Monroe	Sun Belt	-9.7%	94	-26.0%	114	15	-12.2%
Proj. Rk	Team	Conference	2011 F/+	Rk	2012 Recruiting F/+	Rk	2012 Returning Starters	2012 Projected F/+
101	Colorado	Pac-12	-17.1%	108	3.7%	52	14	-12.5%
102	Wyoming	MWC	-13.7%	103	-11.6%	83	14	-13.3%
103	Kansas	Big 12	-19.8%	113	3.7%	53	15	-13.4%
104	UTEP	C-USA	-9.0%	91	-25.0%	109	13	-14.1%
105	Army	Independent	-15.9%	106	-23.3%	107	16	-15.3%
106	Rice	C-USA	-10.8%	98	-11.3%	82	11	-15.5%
107	San Jose State	WAC	-7.1%	85	-25.4%	113	11	-16.0%
108	Troy	Sun Belt	-16.2%	107	-22.6%	101	15	-16.6%
109	Tulane	C-USA	-20.9%	114	-18.1%	93	17	-17.2%
110	Memphis	C-USA	-24.1%	117	-4.6%	68	15	-18.6%
111	Buffalo	MAC	-13.7%	104	-31.3%	119	13	-19.1%
112	Air Force	MWC	-5.7%	80	-11.1%	81	7	-20.5%
113	UAB	C-USA	-18.0%	111	-16.2%	88	11	-21.8%
114	Idaho	WAC	-17.4%	110	-25.3%	111	12	-21.9%
115	Florida Atlantic	Sun Belt	-24.6%	118	-18.2%	94	15	-22.2%
116	UNLV	MWC	-23.5%	116	-22.3%	100	15	-22.2%
117	Middle Tennessee	Sun Belt	-21.9%	115	-23.2%	106	13	-23.5%
118	New Mexico State	WAC	-17.2%	109	-26.7%	117	10	-25.2%
119	Akron	MAC	-27.5%	119	-26.1%	115	15	-26.3%
120	New Mexico	MWC	-35.5%	120	-23.0%	104	15	-31.8%

Tweet Comment 19 comments | 0 recs |

Do you like this story?

Comments

Ugh.

Vols ranked 20th in projections…and 8th in the conference. Mercy.

If I hit a hole-in-one on this grand slam the dominoes will fall like a house of cards. Checkmate.
Follow @bates_jason

by jasonkylebates on Feb 3, 2012 3:05 PM EST reply actions

At least you're not Kentucky?

Follow me at @SBN_BillC!

SB Nation
Rock M Nation
Football Outsiders
Football Study Hall

by Bill C. on Feb 3, 2012 3:12 PM EST up reply actions

But they beat us last year!

With a WR playing QB!

If I hit a hole-in-one on this grand slam the dominoes will fall like a house of cards. Checkmate.
Follow @bates_jason

by jasonkylebates on Feb 3, 2012 3:52 PM EST up reply actions

the Vols

still recruit very well, which makes an impact on any projection system. That said, after last year I’m officially off the Dooley bandwagon. Arguably the worst AQ coach in America. I have zero faith in that program under him.

by cfn_ms on Feb 6, 2012 12:02 PM EST up reply actions

Are you using Steele for returning starter data?

Follow @stilts19

by BenDNole on Feb 3, 2012 3:20 PM EST via Android app reply actions

For now, yeah.

Obviously those figures will change significantly over the coming months.

Follow me at @SBN_BillC!

SB Nation
Rock M Nation
Football Outsiders
Football Study Hall

by Bill C. on Feb 3, 2012 3:44 PM EST up reply actions

Returning starter data

Is there any utility/possibility of incorporating additional data as to whether a team has a returning starting QB? It would seem that that would have an outsized impact.

Also – I’m curious if you could give a hint as to how you’ve been incorporating Rivals’s recruiting rankings. As in, is there any reason to think that Rivals is properly weighting the impact of a 5 star recruit vs. two 4 stars? You mention in the linked piece that you had to guesstimate about the teams beyond 50, but I’m curious if you did this in a way to best fit to Rivals’s own scoring system, or if you’d done your own research as to the proper weighting system?

Using the likelihood of a recruit becoming an All American done by Dr. Saturday here, the top 50 classes become:

Alabama
Texas
Ohio State
Florida
Stanford
Miami (FL)
USC
Michigan
Oklahoma
Florida State
UCLA
Auburn
Oregon
Tennessee
LSU
South Carolina
Clemson
Virginia Tech
Texas A&M
Notre Dame
Washington
Georgia
Rutgers
Utah
California
Mississippi State
Nebraska
Texas Tech
Virginia
Iowa
Baylor
North Carolina
Vanderbilt
Missouri
Oklahoma State
TCU
Purdue
Colorado
Louisville
Arkansas
Arizona State
West Virginia
Pittsburgh
Arizona
Michigan State
Cincinnati
Oregon State
South Florida
Mississippi
Penn State

Now that’s a pretty similar list obviously, but there are some notable shifts, with the biggest movers being teams like Iowa, Baylor, and UNC.

Finally, there’s bit a bit of research done at MGoBlog which might also be helpful, in that it looks at what positions have the strongest correlations to recruiting rankings. Defensive star rankings are more predictive than offensive star rankings, QB star rankings are more predictive than other offensive positions, etc…

Anyways – this is great work. This is just meant as a few other things that might be helpful to incorporate.

by bibigon on Feb 3, 2012 4:25 PM EST reply actions

In the past....

…we’ve done just that with the quarterback position. I’ve tinkered with more heavily weighting returning RBs and WRs depending on run-pass ratio, too. Like I said, this version is overly simple, but I thought it was the very best place to start.

Follow me at @SBN_BillC!

SB Nation
Rock M Nation
Football Outsiders
Football Study Hall

by Bill C. on Feb 4, 2012 7:51 AM EST up reply actions

QB does

have an outsize impact. It’s arguably the ONLY starting position where it matters much more than others, though if you also factor in stuff like % yards returning, % tackles returning etc. the impact gets dulled (i.e. a QB is responsible for a big chunk of yards returning, so there isn’t much additional impact IF you’re already counting % yards returning).

by cfn_ms on Feb 6, 2012 12:00 PM EST up reply actions

meh...

QB does have a larger weight than everywhere else typically, but it’s dependent on the system. In some cases a RB or even a pass catcher can be significantly more important than say an OLB or SS if they are highly skilled.

by Caban on Feb 7, 2012 1:07 PM EST up reply actions

i haven't noticed

in general a big discrepancy between various types of positions and their impact, though admittedly I haven’t really investigated. I’d agree that in some systems position X has an especially high impact compared to position Y in other systems, though I don’t have the data to quantify it or figure out which system/position combos it applies to.

QB almost always has an outsize impact in part because there are limits to how much you can hide a bad one. If you have a crappy RT you can run more to the left and/or put a pass blocking FB in right side of backfield. If you have crappy RB’s you can pass more; crappy WR’s you can run more and/or go jumbo and pass to TE’s (and if you have one good WR and one crappy one then it’s MUCH less of a problem than if all are crappy; this is part of why the Vols struggled after Hunter got hurt); crappy TE’s you can go put them on the bench and run spread or just put in FB’s instead. etc.

If you have a crappy QB though, you really can’t hide him. You CAN’T run every play, and if he fumbles a lot too, even running doesn’t help much.

by cfn_ms on Feb 7, 2012 1:41 PM EST up reply actions

I figured Boise State would be low...

what with their low recruiting numbers and losing so many starters, but 60? That’s right in the middle of the FBS, but their projected F/+ value is -1.7%, which I believe means “below average”. Surprising. Obviously just rough projections using only the numbers at hand. Perhaps a reasonable adjustment would be to include previous recruiting data and determine, based on how it affected their 2011 season, how much of a factor each team’s recruiting performance is expected to be on its 2012 season.

"Shiloh Keo got a MOUTHFUL of Doug Martin!" - Joe Tessitore
Follow @D_Summit

by D_Summit on Feb 3, 2012 5:00 PM EST reply actions

One thing I did a couple of years ago...

…was set up two completely different sets of data — one for major conference teams and one for mid-majors. The idea was that different pieces of data would matter more or less for each. It didn’t end up making that big a difference, really, but it’s a long offseason, so I’ll probably tinker with that. And yeah, I couldn’t make myself drop Boise State below about 15th in my manual Top 25 a month or so ago, so we’ll see. They have to replace a LOT, but in the final projections I’m going to assume that a really good five-year history will help them out more than it is in this set.

Follow me at @SBN_BillC!

SB Nation
Rock M Nation
Football Outsiders
Football Study Hall

by Bill C. on Feb 4, 2012 7:55 AM EST up reply actions

It'd be interesting to see a season simulation using these numbers.

There are a few very intriguing teams on here. If your F/+ projections are accurate, Rutgers has a pretty good shot of going 12-0 (unless that RU/Oklahoma game ends up happening.) FIU is also a possible undefeated.

by hamdenhusky on Feb 3, 2012 10:28 PM EST reply actions