Chip Kelly And the ‘Moneyball’ Theory

There are many theories floating out there about Chip Kelly and his approach to football, most of which are promptly shot down by the coach when presented to him. He would like us all to take our broad brushes and throw them in a vat of turpentine.

One popular thought — brought to life by a Yahoo! Sports article in November — is that Kelly’s decision-making is driven by numbers, that he is using mathematics to guide him, and will spark a “Moneyball revolution” in the NFL.

Like Paul DePodesta and Billy Beane did in baseball, Kelly’s genius comes from exploiting arithmetic that other coaches are too naïve to acknowledge.

“I was told there would be no math,” Kelly joked during his sitdown with reporters at the owners meetings this week. And with that, he began to dismantle the “Moneyball” argument.

“It’s a different game. There are analytics that you study because you can get information. We’re going to study the statistics of our game but to equate it to Moneyball, no. Even Moneyball itself if you really understand it, that’s not what they did. They had three unbelievable rookie pitchers that nobody ever talks about. All they talked about is how those guys are going to take hits or make people walk and do those other things and it made for a great movie, but if you don’t have three great pitchers in baseball it doesn’t matter, you ain’t getting anywhere. Watch the movie, read the book they don’t talk about those pitchers at all, but those pitchers were really, really good. It’s a good story but we’re not modeling ourselves after Moneyball, I’ll tell you that.

“You’re also dealing with a different thing. There is a different amount of money that is spent in Major League Baseball, their payroll is different than what the Yankees’ is. In this league everybody has the same thing.”

When applied literally the idea that “Moneyball” would sweep the league quickly crumbles. The salary cap keeps spending down and  sucks out the middle class (especially with a flat cap). But there are elements of it that are already prevalent both in Philadelphia and elsewhere.

The Bills made headlines by announcing that they will implement a “very robust analytics operation” in Buffalo, and Jaguars owner Shahid Khan recently named his son, Tony, Vice President of Football Technology and Analytics in Jacksonville. The truth is, analytics departments are widespread across the NFL. There is one right in your back yard.

“You’ve got to define what you are looking for from your analytics department,” said general manager Howie Roseman. “If you’re talking about it being part of the process and marrying the subjective to the objective to make really good decisions, that sounds like a good process as you go through this. You want to make sure the tape is really good, the background is really good, and then when you marry the stats to it — marry the measurables — and everything falls into place, that’s when you feel really good about your decisions.

When all of those things come together, as Roseman notes, you are talking about a player that will be selected high in the draft or paid a lot of money in free agency. There are so few players in this league that fall through the cracks. Where analytics are most helpful is as a fail-safe. If statistics indicate that a player with certain measurables rarely succeeds at a given position, you know to probably stay away. Conversely, if a player has great measurables and you missed him during the scouting process, you make sure you round back and check him out.

Kelly, as we have learned, has specific measurables that he wants at each position. Namely, he desires taller, longer players because “big people beat up little people.” That helps refine the search even more.

“We’re lucky to be surrounded by a lot of smart people in this building who are able to tell us what’s important to look for in the numbers by position,” said Roseman. “And it’s also helpful when you have your coaches telling you what they’re looking for at each position — height, weight, speed — and that helps you narrow it down. At least in the past couple years we’ve tried to be a know a lot about a little, not a little about a lot team because there are so many players so as you funnel down the information and get to it, let’s know a lot about those guys.”

The “Moneyball revolution” that Yahoo! predicts is more about game-day coaching decisions. Kelly is known to be an aggressive play-caller, going for it on fourth down where others wouldn’t, trying a two-point conversion or an onsides kick because the mathematical evidence tells him to.

Not so, says Kelly. He cited the number of times he went for it on fourth down — 20 times in 14 games, he thinks — and says that is not out of whack compared to the rest of college ball.

“A lot of our decisions came in the kicking game. If you don’t have a guy that can kick a long field goal what are you going to do when the ball is on the 37 yard line? Will you kick a 52-yarder or are you going to punt it? If it goes in the end zone you have a net of 17 yards. Or do you go for it because you have a good defense and you’re not averse to putting them on the field on the 37 yard line? Those weren’t statistical decisions.”

They were personnel decisions. If he did not have faith in his defense to hold up, he would not gamble.

Numbers don’t play the game.

 

Be respectful of our online community and contribute to an engaging conversation. We reserve the right to ban impersonators and remove comments that contain personal attacks, threats, or profanity, or are flat-out offensive. By posting here, you are permitting Philadelphia magazine and Metro Corp. to edit and republish your comment in all media.

  • peteike

    I think if he can manage to get more out of average players and make the coaching and schemes be the success as much as the athletic ability of the players he will be loved by Philly fans. Similar to Belichick with a Troy Smith or Vrabel or many of the other players who failed when they left NE. Based on the FA signings, he seems to be headed in that direction in a way. Like the Pats seem to be able to fit anyone into their plans, we hope that can happen here. You still need a QB which can take up to a decade for some teams to find, in terms of top 5 talent, franchise player. Exciting to see how it plays out either way.

    • peteike

      Troy Brown, not smith

  • CJ

    I used the moneyball phrase with this free agent class, and I still stand by it. If you know the whole story, it’s obvious to all you can’t win with 100% of those types of players. But if you can look beyond the raw numbers to see who’s getting undervalued in the market based on projected performance based on scheme (Casey), limited time due to injuries (Phillips) or whatever (Fletcher), you increase your chances of hitting it big in terms of value. You can’t win in this league unless you have stars for one, but also some solid contributors who out perform their contracts. This year, Jacoby Jones, Paul Kruger, and Darnell Ellerbe come to mind for the Ravens.

    While that philosophy in and of itself can’t and won’t ever win a super bowl, applying concepts from it like the Eagles have can deepen the middle and back end of a roster, increase depth and competition throughout, and help build a winning team in the future.

    • hillbillybirdsfan

      It really does help with the middle and bottom third of your roster. That way you don’t have to reach for guys off the street when somebody goes down.

  • jp

    Wow, this is a lot more refreshing than “I have to do a better job at that”, “I have to put players in position to make plays” and “Time’s yours.” Can’t wait for the season to start!

    • Kaycee

      We’ll see how refreshing it is if he doesn’t have the Eagles in the NFC companionship game within 3 years like Reid did. Don’t be naive with your selective memories of Reid; he was always obnoxious with the press, but no one cared when he was winning. Belichick treats the press like crap but no one cares because he wins. It’s all about winning, not the press’ feelings.

      • GoBirds1

        Spy cam aside, you can’t compare Reid and Belichick. One wins SBs as in plural and one can’t. One can maximize a players value and potential by fitting a game plan and scheme to his roster, the other still trying to put the proverbial square into the round hole. One can make adjustments, the other is incapable. The head to head record speaks for itself. Reid’s record since 2004 is barley above .500, that screams mediocre.

        • A_T_G

          Don’t look now, but you just compared Reid and Belichick.

          • Essell

            It wasn’t as much a comparison as it was a contrasting of capabilities.

          • GoBirds1

            Thank you, I am sure ATG does not know the difference between compare and contrast.

  • Andy

    I actually think the salary cap makes using a “moneyball” type of mentality even more important. Everyone has the exact same set of scarce resources, so if you can gain an edge by expending yours in a more productive way, there is no way for other, less intelligent, allocators to offset their shortcomings like the Yankee do (did).

    • GoBirds1

      That’s why drafting well and quickly developing players is the main way you can create value by players way outperforming their rookie contract. The current rookie slotting makes rounds 2-4 so important. Also where Reid and company struggled over the years.

    • JSP42

      Exactly. I think Chip Kelly has to be playing dumb here. When everyone has the same amount of money to spend like the NFL, finding inefficencies in the market is MORE important.

      Look at the Oakland A’s for example. They had a superior way of doing things, and in the end all it got them was overachieving (based on their budget), playoffs and a neat book written about them.

      In the NFL if you have a superior way of doing things, and get lucky at the QB position you will win multiple titles.

  • knighn

    Whatever math or science Chip Kelly plans to use, it makes no sense for him to give away all of the answers right now, especially if he wants to win in the NFL.

    • limodriver27

      Trust me, based on his current mindset, he didn’t tap the surface of what is yet to come. It appears to us that he revealed a lot of info, but keep in mind that this is the first lengthly face-to-face he’s had with the Philly press, so ANYthing he revealed is fresh news to us. I have to admit that I loved reading his transcript. He said more in 72 minutes about his plans than Reid did in 14 years.
      CK knows this media game quite well and knows how to play it (and us). It’ll be up to us to wean ourselves from Reid’s style of media contact, which was embarrassing to the organization.

  • Dave

    Add Chip Kelly to the list of people who have no idea what Moneyball is.

    • NickS1

      Good, Oakland never won the World Series with it, anyway.

      • Dave

        Add another.

      • slorck

        But, the point is, they were able to remain competitive and put themselves in a position to succeed which is all you can ask for. By applying the right sabermetrics, you can have a very successful team and maybe, just maybe win a Super Bowl with those mathematical principles. That’s why Howie talked recently of paying more attention to in-game performances rather than relying strictly on the NFL Combine performance of perspective selectees when deciding whom to select.

        • NickS1

          Andy already remained competitive for years, and almost won a SB, so you don’t need “Moneyball” to do that.

          • slorck

            Yeah but he was never able to take advantage of his teams position and get them “over the top”. Furthermore, Reid was horribly flawed and way over rated.

          • NickS1

            Just like the guys in ‘Moneyball’

      • JSP42

        That’s irrelevant. It’s like saying a book detailing how to buy everything (cars, groceries, clothes) very cheaply is flawed because the author isn’t a millionaire.

        It’s the process not the results. Championships are a arbitrary thing anyway from a statistics standpont. Playoffs are basically a lottery (esp with the parity of the NFL + the fact they play one game and not a 5 or 7 game series like the NBA or MLB).

        • NickS1

          Your analogy is flawed. The person writing the book should own cheap things, and the book would be flawed if such details were never put to practice… Your second paragraph basically proves the point that its nothing like moneyball.

    • slorck

      Sounds to me like he just explained it in his answer to the question when he answered that there are “analytics that you (they) may study because you can get (objective) information”. But he didn’t want to compare it with Moneyball but that’s not entirely true. Moneyball is defined partly as a team’s analytical, evidence-based, sabermetric approach to assembling a competitive team. which Roseman further alluded to in his elaboration of that philosophy. This is a science that the Oakland A’s used to procure undervalued players using statistics such as on-base percentages and slugging percentages as indicators of success that they reasoned were cheaper to obtain on the open market rather than speed and contact.

      They took advantage of more analytical gauges of player performance to field a team that could compete successfully against richer competitors in major league baseball such as the New York Yankees. Because of their smaller revenue ($41m) compared to the Yankees who spent over US$125 million in payroll that same season, Oakland was forced to find players undervalued by the market, and their system for finding value in undervalued players has proven itself. There’s more money sharing in the NFL and thus more parity but those concepts are still useful to analytically gauge which players will be more useful in certain systems than others.

      (see Wikipedia for a more detailed explanation)

      • NickS1

        Odd that you failed to mention the main point that CK brought up: They had not one, not two, but three stud rookie pitchers. Maybe if we get three stud rookies in this draft that lead us to some postseason glory we can start with the real analogies comparing the Eagles to ‘Moneyball’.

    • JSP42

      He does know. Hence he spent that whole paragraph dismantling it. Like his argument that the Oakland A’s won because of their pitchers not their cheap hitters.

  • Max Lightfoot

    Well, who knows what he’s actually going to do? Kelly seems like a coach who tries to get the most out of what he’s been dealt, and he wasn’t dealt the greatest hand by Reid’s decade of lousy drafting. Let’s face it – Chip has to clean out the dead wood, find FAs who have something left and share his philosophy, and draft for the near future. So far he’s on track. I DO like that he considers the running game to be a foundation, even though the NFL is a high-scoring passing league. Because that means that he sees the need for a killer offensive line. My hope is that he gets Joeckel or Fisher or even Warmack in the first round, rather than reaching for Geno Smith at No. 4. No matter what, though, it’s probably going to be a rebuilding year, and if they get to 8-8 it will be a tremendous success!

  • http://twitter.com/BaseballSpencer Spencer Schneier

    Yeah, figuring out the best way to use the information available to you is groundbreaking stuff.

    Also- “moneyball,” which I will assume is meant as the use of advanced analytics in sports (but that’s not what the book was about), does work in the NFL (ask the Ravens), and obviously has won a Super Bowl.

  • BrickSquadMonopoly

    Whatever he knows hes already said too much. Bottom line he aint tellin us anything does Belichek ever? Chip Kelly is a mastermind and an overthinker. He analyzes the shit outta everything and thats the kind of neurotic obsessed coach I want coaching my team. He deserves ATLEAST 3 years to implement his system.

  • hotcakes33

    Too bad we didn’t have this to save us from Mike Mamula.

  • Patrick Herron

    It’s pretty clear Kelly’s quotes undermine the central claim of this article. If anything, Kelly has an advanced and nuanced understanding of what Moneyball is supposed to represent as well as the particulars around it. Further, he’s trying to create a Philly-happy PR image, and in doing so, he’s distancing himself from any perception that he’s a numbers geek. But he understands the details about the set of A’s teams that narrowly define Moneyball and he exploits your lack of a similar understanding to help him establish the PR persona he so desperately wants. That difference in knowledge that he exploits, however, gives away his Moneyball-esque approach. I mean, he isn’t arguing with taking a quantitative approach. Instead, he reveals hints of his expertise. Kelly tells you how quantitative analysis is different in football, how the salary structure is different and therefore not identical to the A’s. This is nothing but a sort of sleight of hand that makes an unwitting journalist think Chip’s not into numbers. But to any seasoned analytics expert it’s clear Kelly cannot but help being precise in his analysis and how that analysis informs his decision-making. Because, really what Moneyball is about, is about breaking with a tradition of scouting personnel by who “looks like a player” or “feels like a player,” avoiding the gut instincts of scouts, and finding personnel, schemes, matchups and other dimensions of a game by finding aggregated data that cannot be directly perceived by any scout or expert except through the numbers.

  • JSP42

    I think Chip Kelly is flat out lying. He does make statistical decisions he’s just talking like a traditional football guy. Why? Because no sane coach would want the label of “Moneyball” or “revolutionize the game”. Look at the horrible stuff (in addition to the good) Billy Beane went through. Imagine Philly sports fans wanting to talk about what a genius Chip Kelly and his stats are to the sports radio station in the inevitable 3 game losing streak.

    Go read the book “Scorecasting”, they say that going for 4th down is (by a almost laughably large margin) the right call almost all the time. They even profile Kevin Kelley, HC of Pulaski Academy who NEVER punts the ball.

    http://sportsillustrated.cnn.com/2011/writers/scorecasting/09/15/kelley.pulaski/index.html

    Kelly go for a lot of 4th down conversions, many people think this is Kelly being “ballsy” but anyone who read the Scorecasting recognizes that Kelly is doing the statistically right thing.

    But as you say numbers don’t play the game. There’s a key psychological component. Fans, sportswriters, bloggers, owners, they all exert a pressure on the HC. When you have only 2, maybe 3 years top to win a lot of games and make some noise in the playoffs you’re hesitant to put the mathmatically sound strategy at work, esp since that strategy only works if you employ it for a long time.

    It’s like flipping a coin. In the long run you’ll break even with heads and tails but if in your first 20 flips you get a bunch of heads and few tails and you get fired it doesn’t matter if your strategy of getting a even number of heads and tails is correct. People, since the beginning of time, only focus on the results and not the process.

    So I don’t expect Chip Kelly to go for a bunch of 4th down conversions in his first few years, he’ll do things the traditional way, build some tenure then when he builds up some job security THEN you’ll see the real Chip Kelly. Will he get to that point? Stay tuned.