My blog has moved!

You should be automatically redirected in 6 seconds. If not, visit
and update your bookmarks.

Wednesday, October 7, 2009

Recruiting Big & Kicking Short: More Fun With Numbers

Here's a common complaint I've heard from fans recently: Georgia has a top-10 recruiting class every year, so shouldn't we always be a top-10 program?

On the surface, it seems logical, so let's look at exactly how the numbers stack up, courtesy of a bit of research done by Randy, one of the blog's fine readers...

Randy went back through the past five years (i.e. any player that would still be on a team's roster) and looked at the average star ratings by class. He then took the top 25 from each year and assigned a point value. The top-rated class each year was given 25 points, the 25th-rated class given one point, and so on. He then tallied the points to see, based on recruits, who should have the most talented teams.

The next step was overlapping those results with the current rankings. Here's the current AP top 15 and their "recruiting score," according to Randy's research.

1 Florida 221
2 Texas 204
3 Alabama 124
4 LSU 197
5 Virginia Tech 24
6 Boise State 0
7 USC 246
8 Cincy 0
9 Ohio State
10 TCU
11 Miami 149
12 Iowa 23
13 Oregon 40
14 Penn State
15 Oklahoma State
NR Georgia 203

Now, a couple of problems with the study, some of which Randy also alluded to in his note to me.

For one, recruiting rankings are not an exact science. They're generally strong indicators of future success for individual players, but when you add them all up for a class, there can be wide variances. A few extra diamonds in the rough mean you can have a stud class with a low socre and a few busts means you can have a great score but a below-average class.

The other thing to take into account is how many players leave early for the NFL. Those big recruits also tend to be the guys who bolt school after three years, too. Matthew Stafford and Knowshon Moreno are helping Georgia's overall score here, but they aren't helping the Bulldogs on Saturdays.

And finally, a minor problem with the math. Randy used both Rivals' and Scout's rankings and added them together. What that means is that if you had the No. 5 class one year, you really got 40 points for that, and if you had the No. 20 class, you got 10 points for it. By adding them together, the variance between the top and bottom was essentially doubled. That's not necessarily a problem in terms of how we compare the rankings of each team, but saying Georgia's score of 203 is five times better than Oregon's score of 40 is probably overstating the difference in total talent by a bit.

Still, there are some obvious things we can take from this research.

For one, the great recruiting schools -- i.e. schools over 200 points -- do quite well.

Second, winning with a low recruiting score isn't likely unless you're in a non-BCS conference or the Big East.

Third, it's absurd how much of an advantage USC has in terms of recruiting compared to other Pac-10 schools.

Wait... you wanted discussion about Georgia? Ah, I figured you might.

Yup, Georgia is by far the highest score not to be ranked in the top 15 (or even the top 25) in the AP poll. The next closest competitors are Michigan (170), Notre Dame (155), Oklahoma (151), Florida State (142) and Tennessee (114). I'll give Oklahoma a bit of a pass since they've gone to a bunch of recent BCS bowls and they'd probably be a top-15 program if it weren't for their Heisman-winning QB sitting on the sidelines with a shoulder injury.

But look at the rest of that list: Michigan and Tennessee are in either Year 1 or 2 of a new coaching regime. Notre Dame is a disaster, and, well, you've read the news on what's going on at Florida State.

It's not exactly a group Georgia wants to be associated with.

Of course, the other side of the coin is this: Georgia is in the same conference with three of those other top schools. It also competes against South Carolina, Georgia Tech, Tennessee, Auburn and Arkansas, which had reasonable recruiting scores and added Oklahoma State to its schedule this year. Meanwhile, as I mentioned, teams like Ohio State and USC are miles ahead of the vast majority of their competitors when it comes to bringing in talent.

But even when you add up compeition, attrition and margin of error in the rankings, it's hard to not come to a relatively similar conclusion. I'll let Randy give you his:

"Tossing out injuries, early departures, strength of schedule – hell just about everything – I have concluded either UGA is being coached down or the recruiting services are as clueless as Mike Bobo."

I'm not saying this is fair, but recruiting rankings vs. results is probably the most analytical measure we have of which coaching staffs get the most out of the talent they have to work with, and by that measure, Georgia really does not stack up well -- at least this season. Of course, it's also probably worth remembering that the Dawgs did finish No. 2 in the country just two years ago, and with those same high recruitings scores, LSU, Florida, Alabama and Miami have all had 4+ loss seasons in the past three years.

ADDENDUM: As many of you have pointed out in the comments, this analysis is little more than a snapshot in time. It's a rough approximation of the talent currently in Athens vs. the current ranking. That ranking was different a week ago and could be different next week. And the talent is certainly not necessarily evenly dispersed. Again, that's why I included so many caveats in discussing the research.

I don't know that we can make any absolute statements about Georgia from these results, but what we can say is that, given the talent that should be expected from consistently strong recruiting classes, Georgia should routinely play at or near the same level as other top programs, including Florida, LSU, Alabama, USC and Ohio State. In terms of sheer wins and losses, that has been the case. But I'd also be willing to guess from the vast majority of the comments I hear from fans, the program isn't necessarily viewed that way.

So then the question is: Are these rankings useless? Maybe. Is the value of a program simply derived from national titles rather than simple Ws and Ls? Maybe. Or is there more talent than results? Maybe.

Again, these numbers aren't a final solution to any questions. They're a piece to the puzzle, and a starting point for a discussion I'm sure will continue throughout the season.

OK, moving on...

We've talked quite a bit about kickoffs and kick coverage this week, for obvious reasons. But we've dealt with specific examples, like last week's debacle at the end of the game. What about more aggregate numbers?

Well, Bulldogs Blog MVP Jim F. looked at each kickoff this season and came up with some numbers.

The one problem with analyzing kickoff numbers from the stat sheet is that it doesn't tell us whether a kickoff was directional or deep or how the coverage unit looked or any of that. We get two numbers: How deep was the kickoff and where was the kick returned to? If someone wants to watch the film for all five games and do some better math, I'm all for it. Sadly, I don't have that kind of time right now.

But, thankfully for us, Jim applied a loose interpretation of the numbers that does a good enough job of approximating Georgia's kickoff philosophy on individual kicks. If a kick was 65 yards or farther, Jim considered it a "deep" kick. If it was 60 yards or less, he considered it a "directional" kick. (And, for good measure, there was also one poorly executed squib kick in there, too.)

Here's what he found (total 28 kickoffs):

Type of kick
Avg.Field Pos.
Deep w/touchback Opp 20
Deep w/o touchback Opp 35
Directional Opp 34
Squib Opp 37
All Opp 32

So what does this tell us?

Well, for one, there is some evidence to support Jon Fabris' claim that directional kicking has merits, since essentially it does save Georgia an average of 1 yard per returned kick in field position.

But my guess is that one yard is not enough to overcome the likelihood that a deep kick for a touchback offers. Even if you figure that three-quarters of all of the "deep" kicks will be returned, the statistical difference between the average field position the opposition would get on that 75 percent of kicks would be just about four yards per game more than the directional kicks. Yet on that one out of four that went for a touchback, Georgia is gaining 14 yards of field position over a directional kick.

And how much of a difference does that make? A big difference, actually. Here's Jim's breakdown of field position:

Starting field pos.
Opp. Drives
Opp. Scores
Percent scores
UGA territory
4 4100%
Opp. 40 or better 7 6 86%
Opp. 30 or less
16 4 25%
Touchback (i.e. opp 20)
6 1 17%

*Note: Two of the drives that started inside the 30 ended in turnovers.

So, what do we learn from this?

The obvious answer is, if Georgia forces its opponent to march 70 yards or more for a touchdown, the defense is going to be pretty successful. If they don't, there's a good chance points are going on the board.

So if we know that a.) your best chance at having the opposition start behind its own 30 comes from kicking deep, and b.) having the opposition start behind its own 30 drastically reduces its scoring efficiency, then shouldn't a + b = c.) kick the ball deep for cripe's sake!

But these are just numbers. They don't measure the heart of a walk-on or the excitement of a good challenge or the intensity of the defense when faced with adversity. Those are the intangibles you just can't account for with stats.


Anonymous said...

Dave, next time you're on 960 in the afternoon or the next time you see them, talk to Brahm or Dantzler about our recruiting. They discussed it yesterday afternoon - specifically OL recruiting and the number of complete busts from the 02-05 classes. There is not one 5th year senior on our OL this year that was recruited out of HS. And, if memory serves, there won't be next year either. We've experienced bust after bust at OL. Meanwhile, Arkansas, Tennessee and Va Tech has All American OLmen from the state of GA on their rosters.

Jeff and Chris concluded that we've done what we can to rectify our OL issues and I'm sure there are others who would critique the way that solution came about, but most will agree that Searles has been a worthwhile addition to the staff.

Anonymous said...

Brame, I mean. Sorry.

Anonymous said...

I recently watched an NFL game where the announcers said (and showed) that there had been a rule change on kickoffs to where a return team could only have a 2 man wall now. Is that the rule in college as well? If so, isn't that another reason to kick deep, less likely to set up the pickett fense compared to years gone by?...Just a question

Brian said...

The recruiting rankings should weigh the class of each recruit, as well. Seniors still on the team should weigh more than freshmen recruited just last year who might not even be playing. That would take Staff and Moreno out of the equation, as well as others who didn't stay with the team for one reason or another. It would give players actually seeing the field and/or with more actual experience in D1 football a higher weight.

Billy D said...

One quick stat that destroys the entire conversation.

If the ref didn't call A.J. Green for celebration we are probably ranked in the top 10 right now.

David Hale said...

Agree Billy... again, I included so many caveats for a reason. It's not an indictment of anyone, it's more of a snapshot of the current situation. That dynamic will change -- for better or worse -- each week. Really what the numbers should show is simply that the talent level in Georgia should be expected to be at a similar level as Florida, LSU, etc. In terms of simple wins and losses, you'd probably say the results have been roughly equal, too. But I'd bet most fans don't necessarily say that.

I don't know what the answer is, but I think the numbers here are a good bit of information we can at least bring into a continuing conversation.

Shoaib Ashraf said...


I love the analysis. Can I suggest a post on how subjective rankings are especially at this time of the year? I mean, Georgia loses to two Top 10 teams and we're not ranked. Life would be very different if just a few plays went our way in both games.

By the way, I loved the interview with Thomas Davis!

Castleberry said...

I'd be curious what the numbers look like with all deep kicks lumped together. My assumption is that when we kick deep we're aiming for a touchback.

It seems like that would drop the average starting position lower on a deep kick and give the argument that much more credibility.

Anonymous said...

Those recruiting vs rankings numbers are interesting, but I have two points:

1. It seems to weight all classes equally. A top 3 class that is all seniors now will make for a better team than a top 3 class that is all freshman.

2. You can argue that we're "coaching down" our talent, but the coaches deserve credit for getting that talent on campus to begin with.

Fabris said...

But where is the challenge in logic? Man up, turn your back on frivilous numbers and such.