Friday, February 27, 2015

The Surprised Loser

In May I'm scheduled to participate in an international panel of scholars who will discuss electoral expectations. My own part of the panel is on surprised losers. In fact, as I look at it, I'm up first. Oh great. Anyway, I'm cranking some initial data for the presentation and thought I'd share a little of it here for my tens of readers worldwide.

My presentation will:
  1. Briefly discuss how previous studies establish electoral losers are more negative about government and elections than are electoral winners. Democracy rests on the consent of the losers.
  2. Argue that losers can be divided into two types -- those who expected to lose, and those surprised by the loss.
  3. Further argue that surprised losers may explain the winner-loser differences seen in previous studies. In other words, surprised losers are more pissed by the election results, and thus more negative.
  4. Analyze data from 1952 to 2012 to examine this point.
  5. Provide a deeper analysis of the 2004 and 2012 elections in which incumbents ran for re-election, one from each party.
  6. Look briefly at the news media's role in all this.
  7. Leave and go to the beach (conference is in San Juan, Puerto Rico).
 Okay, so a few data points for your enjoyment. If we pool all the data from 1952 to 2012, we find that:
  • Of those before an election who said they would vote Democratic, 74.5 percent expected a Democrat to win.
  • Of those before an election who said they would vote Republican, 78.4 percent expected a Republican to win.
So it's safe to say people expect their own candidate to win, some years more so than others.  The graph below shows you, over time, the percentage of surprised losers, expected losers, and winners. In close election years, like 1960 or 2000, there are a lot more surprised losers in proportion to expected losers. In runaway election years, the result is more obvious and, therefore, fewer are surprised. Also, note the trend in the last few elections appears to be for more surprised folks. That's interesting, and a good hook.



On another day I'll continue this, breaking down whether surprised losers indeed differ from expected losers in terms of trust in goverment, the fairness of the election, and trust in democracy. Stay tuned.

Polls and the Factually Challenged

The Republican presidential race is the interesting one. I got to wondering how well a candidate was doing in the polls compared to how factually challenged that candidate happens to be. So I took data from PolitiFact's "personality" section and counted up the total number of evaluations and calculated two percentage scores for the key GOP candidates. These scores were:
  • Percentage False -- essentially, the percent of all evaluations that were judged "Mostly False," "False," or the ever-popular "Pants on Fire."
  • Pants on Fire -- the percent of all evaluations that were judged as the worst, the "Pants on Fire."
I compared these to recent polls conducted in Iowa, New Hampshire, and South Carolina (links to the polls here). After all, it's useful to know what states prefer the more factually challenged and, as a consequence, which should be voted off democracy island.

Before we get to the results, a coupla caveats. Two possible candidates, Ben Carson and Donald Trump, were excluded. Carson I excluded because he's never had any statements evaluated by PolitiFact, making it hard to score him. Trump I excluded because he's a clown, but also because his name wasn't used in any polls.  But mostly because he's a clown. And as a final caveat, these are statements judged in large part because they were so out there, so it's not really a measure of a candidate's honestly but more a measure of his or her likelihood to say stupid things.

Basic Results

Rick Perry had the most statements evaluated (158, as of 8 a.m. February 27, 2015), followed by Scott Walker (126). Lindsey Graham had the fewest (9). The table below shows the candidates, the percent of false statements, and the percent of "pants on fire" statements. As you can see, in terms of total falsity Ted Cruz holds a reasonably comfortable lead over Rick Santorum, and Scott Walker and Mike Huckabee are tied in their ability to say false things. Also Huckabee leads the GOP pack in terms of having his pants on fire, followed by Rick Perry. Everyone after that is in single digits.


Name %False %PantsFire
Ted Cruz 64.3 9.5
Rick Santorum 53.8 9.6
Scott Walker 48.4 7.9
Mike Huckabee 48.4 12.9
Rick Perry 46.8 11.4
Marco Rubio 38.6 2.4
Rand Paul 33.3 6.1
Chris Christie 31.5 7.6
Jeb Bush 27.3 4.5
Lindsey Graham 11.1 0

(As an aside, 69.2 percent of all Trump statements were some form of false, and he led by far in terms of percent of statements earning a "Pants on Fire" judgment (30.8 percent ... wow).

Polls and Falsity

So how does this compare to the individual state poll numbers? Not well. First, for the statistically inclined, some correlations -- basically the measure of how good a relationship exists, beweeen -1.0 (perfectly negative) to 1.0 (perfectly positive). For example, In Iowa, the correlation between the poll rankings and percent of false statements is a paltry .04, which is damned close to being zero. Luckily the percent of "pants on fire" judgements comes to the rescue, with a correlation of .30. What's that mean? The more likely a candidate was to make really really factually incorrect statements, the better he did in the Iowa poll. Huckabee drives this relationship as he leads in "pants on fire" and polls in first place in Iowa.

We don't see much in New Hampshire in terms of relationships, so let's skip to South Carolina. In South Carolina, there is a -.73 relationship between percent of false statements and how well a candidate is doing in the polls, and a -49 correlation on the "pants on fire" measure. In other words, the better you did on the polls, the lower your percentage of factually wrong statements.
 
So, are South Carolinians less forgiving of the factually challenged? Perhaps. More likely it's the favorite son status of Lindsey Graham skewing the data. To test this, I excluded him from the analysis. The relationships remained negative, but not as strong (-.55 on all false, -.18 on "pants on fire").

The graphic below gives you a visual display.

Percentage False (x-axis) by Support in South Carolina (y-axis)



So what can we take from all this, other than I need a better hobby? The PolitiFact data seems to be a lousy (so far) predictor of popularity among early GOP voters. I frankly expected more. I figured the candidates more likely to toss red meat out to the early voters would be more popular, and that would be reflected in the "pants on fire" or false evaluations. Of course the PolitiFact data relies on which statements the PolitiFact staff choose to examine, and it's still very early in the primary season, so this is probably an analysis better done later in the year.




Thursday, February 26, 2015

Priming, Anyone?

So there's this meaningless story on a TV station site, which is really a story about how an expert says you shouldn't pay your brat for chores.  And then they ask people, so do you do thing we just said sucks?

Are you surprised, then, that 71.43 percent said "no" and only 28.57 percent said "yes"?

And what the hell is up with statistics to the hundredth to the right of a decimal point? Talk about false precision given this is one of those BS, non-scientific polls that are really more about audience engagement than measuring opinion.  My rant here is really about priming. You have a "financial expert" saying it's a bad thing, then you ask whether people are doing that bad thing. Duh. Of course more are gonna say "no" than "yes." You've primed them to do so. Cooked the data. Skewed the results.

Finally -- and this is the best part -- that 71.43 percent? Based on five of seven respondents. Lemme say that again ... based on seven total respondents. 

Okay, now eight total respondents. I voted twice. Vote early and often, they say, usually in Chicago, or on "news" sites.





Tuesday, February 24, 2015

How To Report a SLOP

I hate SLOPs, as regular readers know, those self-selected opinion polls. But sometimes I come across someone who, while using such a device, actually do it right. Here's an example, and below I include the opening graphs.
Business Journal readers have given new Oregon Gov. Kate Brown a little bit of support as she completes her first week on the job.

However, more respondents to a non-scientific poll asking whether Brown "will be good for Oregon" either don't think she'll do right by the state or simply don't know much about her.

See how right off in the lede "...readers have given..."? That's good. Makes it clear that we're not talking about a random sample generalizable to the public at large. No, it's readers. And in the second graf, even more important, is "...a non-scientific poll...". Again, good. Makes it clear this is not a real poll.

So, while I hate SLOPs, points here for making it clear to the reader that this isn't a traditional, scientific poll. Now if only other journalists would get the message.  

Monday, February 23, 2015

Fox vs Deer

The real political battle in Georgia, the one that really matters, is whether the gray fox or the whitetail deer should be named the state's official mammal.

As this story explains, some kids at a school realized Georgia was one of the few states without an official mammal. Now you'd think maybe humans could be the official state mammal, but that's not the direction the kids took. They went fox, but the state leaned on them to go with deer instead.

In an "only in Georgia" moment, we have this:

First off, only in Georgia would an elected official think being the state (fill in the blank, such as official state reptile, or legislator) provides some legal protection. Sheesh.

So, isn't the fox superior? It certainly is if you check out how often "fox" characters appear in film and literature (scroll down here to see "fox" in popular culture). Other than Bambi, what cool deer characters come immediately to mind?  And here's another vote against deer -- they're apparently popular in hipster culture? Big. Vote. Heck, even Disney recognizes the superiority of foxes. One got to be Robin Hood (below).

But let's explore the "nuisance" argument. Deer certainly win. I've hit one deer in Georgia. They're like really large rodents that can destroy cars. By making them the official mammal, we're declaring war on all those late-night drivers who end up with Bambi splattered across their grill.

The only mammal more annoying than deer? People. I would put legislators, but there's some debate about whether they're mammals.










Thursday, February 19, 2015

UGA Student

Interesting as to whether or not (as officials, including the University president, have said, and the local hospital has said as well) a UGA student died of bacterial meningitis. I'm guessing not. But I'm a bit baffled now.


Tuesday, February 17, 2015

AJC Gets SLOPpy

It's an old complaint of mine -- SLOPs.

That stands for self-selected opinion polls. They're entertaining, but they're bogus. And the AJC has one up and running for your journalistic enjoyment. I saw it via Twitter.
To be fair, this is a blog, not a news story, so I'm not gonna complain too much. The choices are
  • Shorter wait times
  • Routes closer to my home
  • Extended hours of service
  • More transit cops on board
  • Better connections to trains
You can only vote once, which is good. I voted for routes closer to my home because, well, I live in Athens. Fairly sure MARTA does not have a route close to my home.

Oh, and on Twitter there were a few funny responses to the question of what MARTA can do. My favorite?  Not smell like pee

Monday, February 16, 2015

Polling (Un)Surprise

A poll by a Georgia transportation group found support for a tax to support Georgia transportation, according to this AJC story.

Here's a key graf:
The poll, conducted for the Georgia Transportation Alliance, shows strong support among Georgians for lawmakers to boost funding for transportation, according to an advance copy obtained by The Atlanta Journal-Constitution. The alliance is an arm of the Georgia Chamber of Commerce, which has made increased transportation funding a priority.
Beware when a poll coincidentally happens to support the sponsor's point of view.

It's a survey of 603 "likely voters" earlier this month. But what I can't tell, at least at first blush, is how the survey was conducted. The Chamber's vice president, however, defended the poll. “It’s scientifically accurate." The AJC notes, however, the results differ from one it conducted in January. That poll did use serious methodology that included both landlines and cell phones. This new poll, I can't tell if it's a robo-poll, whether it included cell phones, or anything at all of the methodology except they do provide the questions. On the site of the group that did the polling: "What separates us from the pack is the use of our innovative techniques both in conducting the survey and analyzing the results." I don't care from innovative. I care from the quality of the sample.

So let's look at the questions. At first look they seem straightforward, but think about the order here. I can't tell if the questions were rotated in any way, and the order in which they are presented could certainly prime a respondent to be more responsive to raising taxes for roads.

First Question:  How important are roads, bridges and public transportation to economic development and job growth?
 
You ask that one first, you've set up the rest of the answers to be more positive even if you say the choice is between killing kittens and taxes to improve roads (which, by the way, I support). Go to the bottom of this page and see the subsequent questions. They are designed, quite frankly, to elicit a positive answer to increased taxes for roads.








Friday, February 13, 2015

UGA Jobs That Pay The Same

In my never ending quest to analyze UGA's byzantine salary system, I've discovered many unrelated jobs are in the same pay scale. In other words, the position of fry cook may have the same minimum and maximum salary as a mechanic. Or journalism professor. No, wait -- this excludes faculty positions. So, what are some of the fun jobs that have the same pay range? Glad you asked. I only take two or three jobs in each category. Many have 10 or more listed in a particular pay range.
  • herdskeeper II, snack bar supervisor, and TV programming specialist are all listed as "pay range 45" with an annual minimum of $22,750 and an annual maximum of $43,831. Herdskeeper. Neat.
  • A fire extinguisher inspector, a video editor, and (I am not making this up) "post conviction relief assistant II" are all at range #47, from $22,750 to $46,051.
  • A welder, a lab animal health technician, and a sheetmetal worker are all #53 ($25,383 to $54,977).
  • My favorite: "first mate" is #68, along with business manager II and parking operations manager and police corporal and, of course, others. Range is $36,760 to $79,620. First mate? Ahoy, skipper.
  • A police sergeant, a project architect, and a TV news director are all #74, ranging from $42,629 to $92,334. I should insert some snark here about TV news directors. Insert your own.
  • At the very tip top end of the scale, at range #123, there's only one job -- "senior associate CIO & Technology Officer" with a range from $85,000 to $195,500. My guess is, only one person is exactly that. And UGA being UGA, I wonder if it's not near the top of that scale.

What does this mean in the greater scheme of things? That I need to get the hell out of the office on a Friday afternoon, that's what. Or at least walk away from the spreadsheet.

Thursday, February 12, 2015

Rush Limbaugh Reveals J-School Secret

I'm a huge Rush Limbaugh fan. Talk radio, with Limbaugh as its longtime star, helped me get tenure. For a while I was (modestly) among the top scholars in the country researching the effects of talk radio. It didn't get me rich, but it did get me enough publications in top academic journals to win promotion and tenure.

So Limbaugh the other day revealed one of the deep dark secrets of j-schools. I've been teaching at a j-school for 23 years and, to be honest, I'd never heard this. Someone's been keeping it a secret from me, I suppose.  So here's the text of the bit I'm focusing on. It's a caller to his radio show talking about Hillary Clinton. The caller has worked at a TV news station and they agree about how the news media create hooks or narratives, in this case about Clinton and her candidacy. The full text is here if you want context, but this is the bit I'm interested in:

RUSH:  No.  What's new is, the journalism industry acknowledging that there is such a thing as the hook to the consumer, or acknowledging that there is such a thing as the narrative to the consumer.  This is one of the deep, dark secrets that they taught in J-School, that the news consumer was never to be told and never to know.  Anyway, Jackie, I gotta run.  I'm a little long.  I appreciate the call.  

Let's give Limbaugh the benefit of the doubt. Maybe he's quite knowledgeable about what's taught in journalism schools. Maybe he has a mole in a j-school somewhere, learning the "deep, dark secrets." Or maybe he's just making shit up.  In a way, I think he's on to something. In storytelling, you generally don't tell people "oh, this is my hook, here it comes, wait for it." That gets in the way of the story, unless of course you're into some postmodern stuff in which structure overwhelms narrative.

So maybe he's right -- in a way -- though for all the wrong reasons. It's not a "deep, dark" secret that you don't tell the audience "hey, this is my hook" because, well, it'd be damned stupid and it would get in the way of the news and interfere with the storytelling. That's what media critics are for.

Best Paid UGA Faculty

And in my final post on UGA salaries (see previous days for other stuff), I look at the best paid faculty who are not also administrators (deans, departmeht heads, etc.).

Now keep in mind that the top salaries belong to the UGA president, a few coaches, and you have to get to #11 for the first professor who is not in charge of a department or college. In that case, he holds an endowed chair -- in the business school -- which explains the hefty salary. As is the second highest, also a chair, also in the biz school. Sigh. And the third. You see a trend here? And yes, the fourth highest salaried professor is also a chair in ... wait for it ... the business school. And the fifth name? Yup, she's in the Terry College as well. It goes on that way for a while. I got bored checking but you have to go down the list a ways to find, yes, someone from the law school where, if you publish two law review articles, that'll get you tenure.


Professor Name
Department
Salary 2014
Jason Colquitt
Management
$332.980.09
Stephen Baginski
Accounting    
$320,673.50
Sundar Bharadwaj
Marketing
$310,999.70
John Hulland
Sales Management
$297,999.80
Annett Poulsen
Banking and Finance
$297,830.79

Where's the data from? The State of Georgia makes it available online. Look for yourself.  And in fairness, if we had a medical college tied to the university, like UF, odds are those salaries would top the list. 

Oh, and I am the 1,240th best paid employee at UGA. Fear me.

Wednesday, February 11, 2015

More Fun with UGA Salaries

I've been playing the last couple of days with UGA salary data (see here and here). In this analysis I look at the salaries by category. For example, the 2014 data lists 18 deans. The salary ranges from $55,643 to $360,395. That low salary is probably someone who was here only part of a year. We spent just over $4 million in dean salaries last year.

Oh, but how about vice presidents, you ask? I show 12 vice presidents, which is probably about 10 too many, with a salary range from $14,187 to $318,600. Again, assume that small number is someone here for only a month or so. So how much in total did UGA spend on vice presidents? Just over $2.3 million.

Department heads? My data says there are 115 of them, ranging in salary from $21,444 to $377,585. As you'd expect, with so many, the total spent was steep, almost $18 million. But don't forget directors, which are kinda like deans and department heads. There are 210 of them in this data ranging in salary from $10,448 to $264,388 at a total cost of over $21 million.

At UGA we also have a lot of deanlets. You know, those associate and assistant deans. How many is hard to say because they go by different titles, but best I can tell there are at least 48 of them with sslaries ranging from $51,667 to $315,909 at a total cost to UGA of over $7 million.

So what do we have the most of? According to the salary data, that would be "student assistant" (5,536 of them), followed by "graduate assistant" (3,434 of them). Those student assistants rack up a cost of over $13 million and those grad students, $55 million. Wow. I would not have guessed that. Then again, if you multiply even crappy pay by the thousands, the dollars add up.

Weirdest Data

* The minimum salary of a "service/maintenance worker" is $8. Sometimes the data get messy. That may be a per hour pay slipping into the annual pay column. It's hard to say. "Laboratory assistant" is similar, ranging from $17 to $39,520.

* There are only eight jobs listed in which only one person holds it. Obviously president, but a few others including "interior design professional."

* UGA has three "dental professional" folks. No idea where. We also have four "dental technical/paraprofessional." Without looking them up, I'd guess either in Pharmacy or Vet or Public Health.

* The data list 45 as "police officer" with a total salary of just over $1.3 million, ranging from $160 to $46,228.

* All in all, the data list 24,524 different employees or former employees with a total salary of $657,129,237.  Wow, $657 million. There were 118 different job categories. What's interesting is the salary data job titles do not match the official UGA job titles, which is much longer, more finely tuned. The salary data lumps a lot of people together.




Travel at UGA

An interesting aspect to the 2014 UGA salary data is a column of data on how much was spent on travel by each employee. So who gets the most travel money? The results make sense. I'm going to save myself some time here and not name names, but I did look the top five up to make sure of who they are.

1. A College of Ag and Environmental Science professor, $48.989.84 in travel. No surprise there. Probably grant-funded travel.

2. A research assistant/technician, $40,161.30 in travel. Best I can tell, this person is part of a big grant and the job is to collect data. Get this -- the salary for this person is almost half of the travel money received ($26,720).

3. Another professor, this one in physics, at $37,416.26. Maybe the person travels to other planets.

4. A head of a major place on campus (how's that for vague), and the travel in this case makes perfectly good sense, $36,335.01. If I named the place, it could only be one person.

5. A "public service professional" who received $36,097.31 in travel money. Huh? It makes sense if you look at it, a person who focuses on outreach and training off campus nearly most of the year.

The lowest travel money? Someone had $3.28 in travel. A few people had five bucks. A grad assistant had $9.93.A sandwich, maybe?

And in case you're wondering, there's a pretty good correlation between how much you make (salary) and how much travel money ya get. For you statistically minded out there, it's r = .46. That's a pretty good correlation (1.0 or -1.0 would be perfect, 0.0 would be no relationship).

Yeah yeah, but what was the total of all travel money? Glad you asked. Excel tells me, and this can't be right, more than 15 million bucks was spent on travel last year. Weird. That's gotta be wrong. I've tried the formula two different ways and I get the same unbelievable number. In fairness we're talking 6,842 different people at UGA who received at least some travel money (a whole bunch are zero). If you take all the people who got at least some travel money, the average for 2014 is $2,291.51. I find this amusing, by the way. We're lucky at Grady to scare up $1,000 a year in travel money to present research, etc. Well, some get a lot more. Squeaky wheels, mostly. That's a whole different post.

And yes, I have the names for everyone. I'm just in a rare nice mood. If you really want the names, lemme know. I do also plan on updating this with a deeper dive into the data. Just trying to get ready for class at the same time.

Update

What titles have the average highest travel money? No surprises here, really. It's all about the suits, all about the admins. Money truly does trickle downhill.

1. Deans ($12,090)
2. Vice Presidents ($10,189)
3. Department Heads ($6,216)
4. Assoc/Assist. Deans ($5,863)
5. Directors of Units ($5,037)
6. Alumni Relations ($5,010)
7. Assoc/Asst Dept. Head ($4,465)
8. Assoc/Asst Director ($3,937)
9. Assoc/Asst Provost ($3,616)
10. Professor (finally, dammit, $3,461)

Oh, and I love #22, categorized as "former employee" at an average of $2,257, about double what I get as a current employee in travel money.


Tuesday, February 10, 2015

Salaries at UGA

I always find salaries good for potential stories, and just plain fun to mess around with. There's this pdf that includes all of UGA's (where I teach) codes for various jobs and the minimum and maximum salaries. Me being me, I turned this into an Excel spreadsheet and started playing. A first blush:

Highest "Minimum" Salary

This is $91,800 for something called "senior licensing manager II."  The maximum this person can make is $207,000. What a burden. Next is "senior associate CIO and tech officer" at $85,000.This unlucky person's maximum is $195,500.

Lowest "Minimum" Salary

This is tougher to do. there are several at $15,138 annually (county secretary, camp counselor, laboratory helper, etc.). Oddly, this also includes "mental health professional" and "TV account executive." Weird.

Mind the Gap

Looking at the gap between the lowest or minimum salary versus the ""annual entry maximum" is simply seeing how low a salary you could have against how much you can possibly make just starting out in the same job class. Our friend the CIO person mentioned above has the biggest gap. The lowest you could make in that job is $85,000, but the most you could make just starting out is $127,500, for a $42,500 difference. Next biggest gap is for IT executive director ($39,625). Some of the gaps are small by comparison, as you'd expect -- $3,413 for lots of different jobs, many similar to those listed above under "minimum" salary. That's your "wiggle room" when negotiating for a salary, I suppose.

I could of course find the actual people who hold those titles and see what they actually make. It'd be interesting to see how many at UGA are near the maximum of their specific job title and class code.

Of course salaries are also depressing. For example, I'm a full professor, yet I make significantly less than all the assistant professors in the business school. WTF.

Monday, February 9, 2015

Those Wacky Political Science Profs

This past weekend lots of folks have been playing with this site, which uses Rate My Professor data to look at gender differences based on keywords.

So I put in the two ideological terms of "liberal" and "conservative" to see if any gender differences emerge (do your own terms, it's fun ... I'll probably do really weird ones later and blog about it. Just because).

First, political science is the top of the heap, which makes sense given it's a field devoted to politics. You'd expect more ideological labeling there than in, say, chemistry. So below is "conservative." Blue is male, so clearly political science is not only more likely to have faculty described as either "liberal" or "conservative," but males are more likely to be identified as "conservative."


Okay, but how about female political science profs? There's a difference, as seen in the graphic below. The scores are flipped, with females more than males likely to be identified as "liberal."

The lesson? Chemists apparently have no ideology.







f

Thursday, February 5, 2015

Rent in Athens (Geogia)

I was messing with some Census data today and got curious about how much rent actually is in Athens, Georgia. The results are kinda sorta interesting. According to this page, the median rent in Athens is $774. What's hard to tell is whether UGA dorms are folded into these results, which would seriously skew the numbers.

Okay, but let's look at the fun stuff, the most expensive. According to the data, there are 1,440 households that pay more than $1,500 a month rent. That seems high, but then again we see apartments popping up all over town now that advertise "luxury living" for college students. I'm sorry, if you're in college, you should not be living luxuriously. You should be in a dump. In a student slum.


Wednesday, February 4, 2015

Recruiting Classes

Today is recruiting day. I'm not an expert in this stuff, but I got curious about previous recruiting classes and how well they lined up with how teams finished this past college football season. There's a table below. As you can see, looking at the 2012 and 2013 classes as ranked by Yahoo Sports, the predictive power is at best ... okay. TCU performs far better than you'd expect based on its class rankings (in 2012 it was #37 and in 2013 it was #30). Bama is often the #1 in class, but 2014 didn't turn out as well as Tide fans might have liked. Notre Dame had a top class in 2013 but looked awful this past season. If I had time and the interest, I'd dump a larger data set and try to massively correlate recruiting rankings with final football rankings. Way too much work.


2012 Recruiting Ranking
2013 Recruiting Ranking
2014 Final Ranking
1. Alabama
1. Alabama
1. Ohio State
2. Texas
2. Ohio State
2. Oregon
3. Florida
3. Notre Dame
3. TCU
4. Ohio State
4. Florida
4. Alabama
5. Stanford
5. Michigan
5. Florida State
6. Florida State
6. LSU
5. Mich. State
7. Michigan
7. Mississippi
7. Baylor
8. USC
8. Auburn
8. Georgia Tech
9. Miami (Fl)
8. UCLA
9. Georgia
10. Auburn
10. Florida State
10. UCLA

Let's look at Georgia, given I work at UGA. The dawgs finished 9th this past season. They finished, according to Yahoo, with the #12 class in both 2012 and 2013, so I guess you could argue a #9 ranking means they slightly overperformed. The 2014 class was #7, for what that's worth. Call it even.

Using ESPN's data, you can see below how Georgia has done over time, as high as the #4 class in 2006, as low as the #13 class in 2007.



Finally, using a different site, I wondered who had the worst recruiting class this year. The winner (or loser)? UAB, which may be killing its football program. It's ranked #137, just below Western Illinois.

UPDATE

I ran a correlation of the 2013 class rankings (Yahoo Data) versus the 2014-2015 season final rankings. It's not great, r = .42.  I had to cook the data a bit as some schools with top classes didn't finish in the top 38 teams. In those four cases (Florida, Michigan, South Carolina, and Vanderbilt), I ranked them all 39th. A scatterplot of the data is below. I didn't bother flipping the data, so low = better.