Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52
keyboard_arrow_uptop

The Oakland Athletics finished 2013 with baseball's fourth-lowest payroll, fourth-best offense, and best clubhouse chemistry. Debate has centered on whether the latter two are related. There’s nothing objectionable about “good guy” genes—it’s a solid organizational goal to have. But chemistry alone doesn’t put runs on the board, and if a team is missing the talent, they better find the runs elsewhere. The 2002 Athletics discovered them in walk deities and college arms; once those methods pervaded front offices, the A’s slipped back into losing. Was chemistry the only undervalued commodity of their recent resurgence?

As the baseball community obtains more knowledge, roster construction strategies evolve. Previously undervalued talents like walks and defense are now accepted constructs. The A’s are Hollywood-infamous for adopting them before their competitors while prices were low. After a 74-win 2011, they cheaply signed Brandon Inge and Jonny Gomes, who Brandon McCarthy claimed bolstered the clubhouse DNA to the tune of 20 wins. But Inge and Gomes were two of several players who also bolstered a less-visible statistic: fly ball-to-ground ball ratio.

League

Oakland

Year

FB%

GB%

FB/GB

FB%

GB%

FB/GB

Rank

2010

28.2

45.6

0.62

27.2

43.1

0.63

11

2011

28.0

46.1

0.61

29.0

42.5

0.68

5

2012

27.2

46.6

0.58

30.4

40.4

0.75

1

2013

23.6

46.1

0.51

27.1

38.5

0.70

1

Since 2009, fly ball rates (which exclude popups here) have gradually decreased. Oakland has defied this trend, ranking first from 2012-13. What caused the A’s jump? Let’s inspect their transaction logs, beginning with their acquisitions for 2012:

Name

Pos.

Method

FB% Prior to OAK

PA with OAK

TAv

Josh Reddick

RF

Trade from BOS

33.6

1114

.272

Seth Smith

LF

Trade from COL

33.1

851

.274

Brandon Moss

1B

FA (minor-league)

29.3

801

.331

Jonny Gomes

DH

FA (1 yr., $1m)

37.9

333

.324

Brandon Inge

3B

FA (minimum)

32.7

311

.244

Stephen Drew

SS

Trade from ARI

33.8

172

.274

Kila Ka’aihue

1B

Trade from KCA

33.6

139

.252

Is it a coincidence these seven batters had above-average fly ball rates when Billy Beane and his front office acquired them? Maybe. It’s less believable when reviewing his 2013 shopping cart:

Name

Pos.

Acquired for

FB% Prior to OAK

PA with OAK

TAv

Jed Lowrie

SS

Chris Carter

43.8

662

.290

Chris Young

CF

Cliff Pennington

33.9

375

.246

John Jaso

C

A.J. Cole

27.8

249

.289

Alberto Callaspo

3B

Grant Green

31.4

180

.285

Stephen Vogt

C

PTBNL/cash

56.5 (27 PA)

148

.254

This time, Beane spent more to fill premium defensive positions1. Their commonality is unmistakable: while fly balls around the league grew rarer, Beane stocked his lineup with air-inclined hitters. As a result, 60 percent of Oakland's plate appearances last season were taken by “fly-ball hitters” (defined as a hitter whose ground ball rate is one standard deviation below the league mean)2. See how that compares to the rest of the league:

Let’s contextualize Oakland’s outlier ways: 60 percent of their plate appearances were taken by fly-ball hitters, who by definition compose 16 percent of the league. No other team in the past nine years has touched 45 percent. Beane’s roster was so ground-allergic that only 0.8 percent of their plate appearances were taken by “ground-ball hitters.” That’s not just a concentrated effort to target fly balls. That’s a mission statement.

The other platoon advantage
The Book: Playing the Percentages in Baseball devotes a chapter to platoon effects. Five pages discuss handedness, a mainstay of baseball analysis today. Two pages cover a less visible effect: batted-ball tendencies. Authors Tango, Lichtman, and Dolphin found that fly-ball hitters had an advantage over ground-ball hitters, simply because they are better hitters—you can’t homer on grounders, after all. They also found that fly-ball hitters are especially good against ground-ball pitchers, because the former tend to swing under the ball while the latter want the hitter to swing over the ball.

However, Tango et al. noted that this platoon advantage is hard to exploit because players tend to be neutral rather than lean to either extreme. Also, the advantage itself is very small, and hence overshadowed by the handedness platoon. Such a minimal advantage would (theoretically) require being multiplied through several hitters to become meaningful.

So what happens when a determined, resourceful general manager decides to overhaul his lineup with fly-ball hitters, capitalizing on a league-wide trend toward ground-ball pitchers?

First, updating The Book’s findings using the above hitter definition (and applying the same for pitchers), here’s how each type of batter performs against each type of pitcher in True Average, from 2007-13:

TAv

Pitcher Tendency

2007-13

FB

Neutral

GB

All P

Batter Tendency

FB

.271

.276

.278

.275

Neutral

.270

.267

.258

.266

GB

.240

.229

.221

.230

Fly ball batters are nine points of TAv better than neutral batters. That’s the difference between Cleveland’s sixth-ranked offense and the Dodgers' 10th-ranked offense. That’s a significant advantage! Here’s how often each matchup occurs:

% of PAs

Pitcher Tendency

2013 LG

FB

Neutral

GB

All P

Batter Tendency

FB

3.4

9.6

2.6

15.5

Neutral

14.0

40.4

11.3

65.8

GB

3.7

11.7

3.2

18.7

All B

21.1

61.7

17.1

100.0

This table doesn’t differ much from The Book’s version. Approximately two-thirds of batters and pitchers have neutral batted ball tendencies, and thus the majority of batter-pitcher matchups will not have this platoon effect in play. That is, unless you hit for the 2013 Athletics:

% of PAs

Pitcher Tendency

2013 OAK

FB

Neutral

GB

All P

Batter Tendency

FB

13.4

37.6

8.7

59.8

Neutral

8.9

24.8

5.7

39.4

GB

0.3

0.4

0.1

0.8

All OAK

22.6

62.8

14.4

100.0

The 60 percent found earlier, segmented into pitchers faced. From above, fly-ball hitters have a .275 TAv—15 points above league average, and the mark of a top-five offense. If a team distributed their plate appearances this way, they would position themselves for offensive prosperity. How much did it help Oakland?

TAv

Pitcher Tendency

2013 OAK

FB

Neutral

GB

All P

Batter Tendency

FB

.256

.282

.302

.279

Neutral

.285

.284

.279

.283

GB

.160

.130

.297

.161

The 38 percent of their fly ball hitters’ plate appearances against neutral pitchers resulted in a .282 True Average. That’s better than the solid league TAv in that matchup (.276)—and it occurred for the Athletics four times as often!

Moreover, Oakland fly ball hitters hit .302 against GB pitchers, a matchup occurring nine percent of the time. Another way of putting that: In 547 plate appearances against ground-ballers, fly ball-hitting Athletics (such as low-salary acquisitions like Jed Lowrie and Brandon Moss) hit like $16-million Matt Holliday. The rest of the time—over 90 percent of PAs—they hit like Chase Headley.

The Book claimed that managers weren’t using this platoon advantage enough. It appears that Billy Beane has, effectively transforming his batting roster into 12 Chase Headleys and a Matt Holliday.

But doesn’t O.co Coliseum stifle fly balls?
Yes, it still does. In fact, the Athletics hit .299 on fly balls—the 10th-worst TAv—and that’s park-adjusted. Their slash line on fly balls was a pitiful .197/.191/.615. The Boston Red Sox, who had the next-highest percentage of plate appearances taken by air hitters at 39 percent, had a .409 TAv on fly balls (.313/.302/.855)3. Compounding the problem further is the A’s hitting 1,198 fly balls, tops in the league. It’s even worse if popups are included—no team came close to their 420 popups.

But an air-happy team means a ground-averting one: The A’s had the fewest grounders in the league by far (not a single team broke .200 TAv on ground balls)4. More importantly, the A’s had the league’s fourth-highest line drive rate—and a .634 TAv on liners, second best in the league:

FB

TAv

GB

TAv

LINEDR

TAv

POPUP

TAv

OAK

1198

0.299

1701

0.183

1094

0.634

420

0.026

LG_AVG

1045

0.320

2043

0.182

1027

0.578

312

0.028

The purpose of acquiring fly ball hitters isn’t necessarily to hit fly balls—it's also to hit line drives. Fly balls lie at the mercy of HR:FB rates and park effects. Those constants hardly affect line drives, though, and line drives are the best kind of batted ball, because they become hits two-thirds of the time. And when are hitters more likely to hit a line drive? When they have a matchup advantage!

Because Athletic batters had the platoon advantage so often, they hit many line drives, and “quality line drives” at that (only Tampa Bay, also a fly ball-inclined team, bested Oakland’s line drive TAv). No matter how expansive O.co’s dimensions are, it won’t suppress the offensive gold that is a line drive.

Hasn’t this advantage been around forever?
Definitely. Power is a cornerstone of top offenses, since there’s a home run for every 10 fly balls. Even Billy Beane, bless his secretive heart, will divulge that home runs are the key to winning. Everyone wants to cultivate power hitters. Yet no team has sniffed the Athletics’ 60 percent of plate appearances by fly ballers in the past nine years:

Year

Team

FB_PA%

Wins

2013

OAK

59.8

96

2010

ARI

42.0

65

2011

TOR

40.0

81

2013

BOS

39.5

97

2008

ARI

38.7

82

2013

SEA

38.1

71

2009

PHI

37.3

93

2011

ARI

36.9

94

2010

NYA

36.3

95

2010

TOR

36.2

85

2005

TEX

36.0

79

If other teams have attempted this approach, they weren’t nearly as dedicated to it. So how has it remained an “inefficiency?”

Perhaps, as The Book posited, the advantage was too small to be considered. A team could’ve developed a plan to target inexpensive fly ball hitters, but along the way realized that established edges like favorable handedness matchups, walks, or defense had a larger impact. Worse, the team might’ve disregarded those run sources, losing any advantage gained by the batted ball platoon.

The A’s deftly avoided this trap: not only have they benefited from the batted-ball platoon effect, but they had the second-highest percentage of favorable-handedness matchups. Thanks to aggressive platooning, 70.4 percent of their plate appearances came against opposite-handed pitchers. They also had the third-highest walk rate, which isn’t surprising—fly-ball tendencies go hand-in-hand with patience.

Where the A’s really kept on course was defense: Reddick and Young are not only fly ball hitters, they’re established outfield pros. Cespedes improved considerably as a sophomore. With their starters funneling balls to the outfield5, the A’s posted the league’s second-best defensive efficiency. So despite an apparent re-calibration of team strategy, Beane remained loyal to proven sabermetric principles.

A related theory on why no other team has capitalized on the A's air-ball plan: the advantage wasn’t a net positive until now, after the growth of ground-ball pitchers. This is less verifiable because labeling a pitcher isn’t an easy exercise. My one-deviation-above-the-mean method is always scaled to the league—it doesn’t declare pitchers as a type based on an actual rate threshold. Given the rising ground ball to fly ball ratio, though, it’s not unreasonable to suggest that this platoon advantage has increased enough to be potent.

Ultimately, I’d postulate that in some cases, implementing a team-wide fly-ball initiative might’ve compromised in-place strategies. Because the advantage is small, it would’ve taken a roster revamp to legitimately capitalize on it. The A’s did exactly that—they currently look nothing like their 2011 roster—but it couldn’t have been simple. Budget, owner expectations, landing very specific players, competition, and environment are major influences on the process and result.

If you’re into endings, this generation of the Moneyball A’s, like the 2002 team, lost the ALDS in five games. The irony lies in how: in 24 total innings against Justin Verlander and Max Scherzer, the Athletics scored three runs. Fly-ball pitchers limited Oakland’s fly-ball batters to a .256 TAv during the season, and as you know, Verlander and Scherzer are extreme fly-ballers. Good process, bad playoff opponent.

  1. The outgoing players in Beane's 2012 trades have accumulated just 2.5 WARP, while 2013 discards included A.J. Cole and some everyday players. In 2012, Beane splurged on Yoenis Cespedes, whose own ground-ball rate fell significantly this year.
  2. “Ground-ball” and “fly-ball” hitter labeling will have a tiny number of misattributions due to small samples. For example, Lou Marson has demonstrated himself to be a trustworthy ground ball hitter in his career, but had only five plate appearances in 2013 and a two-to-one fly ball-to-ground ball ratio, which would make him mislabeled as a fly-ball hitter. This matters little because these plate appearances hardly move the needle, and batted-ball tendencies stabilize very quickly.
  3. Is it a coincidence that Boston scooped up Stephen Drew and Jonny Gomes after they left Oakland?
  4. Fewer ground balls also means fewer double plays: Oakland had the second-lowest number of them.
  5. In a clever twist, Athletics starters—Bartolo Colon, Jarrod Parker, A.J. Griffin, Tommy Milone, and Dan Straily—all had above-average fly-ball rates, and the A’s as a team led the league in fly-ball rate allowed. As discussed, this would lead to home runs allowed in many cases (see Griffin), but the A’s play 81 games in O.co and their outfielders have flashy leather. A’s pitchers had the sixth-lowest slugging percentage allowed last season.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
apilgrim
12/18
Great article. I tried to look up Chili Davis' batted ball stats (I remember him being a fly ball hitter) but no luck. Also, does this mean batted ball stats will be coming to BP? I had to go to another site to get them. Brooks Baseball has them but a little bit harder to interpret than the simplicity of the other site.
AndrewKoo
12/18
BP has multiple BIP (Balls In Play) reports in the sortables: http://www.baseballprospectus.com/sortable
jfmoguls
12/18
Wow, that's great detective work. Ben and Sam were talking on the podcast about what stats they wanted on a game broadcast. Now I want fly-ball rate! Rethinking my strat-o-matic draft too.
delatopia
12/18
Unfortunately I don't think sim games at all capture this kind of granular statistical data but are more of a brute force approach based on the basic rate stats we're all used to. It seems to me that a card rated .300, whether a flyball or groundball hitter, will still hit .300 against a league average pitcher, whether that's a flyball or groundball guy.

If you're looking for undervalued Moneyball-type areas in baseball sims, look at CERA (component ERA), real-life quality of defense behind a pitcher, and switching ballparks. I'll leave it to you to assess their value.
pguimaraes
12/18
Ths article alone is worth my entire year's subscription, damn that is good work
nberlove
12/18
Fantastic article. This is why I subscribe.
msloftus
12/18
Great work -- really interesting article.
PeterCollery
12/18
Excellent, excellent analysis.
bkirkman
12/18
Good work!
dianagramr
12/18
Very very cool work ..... I love reading about people uncovering little nuggets like this.

One question, how does the increasing K% across MLB impact this "strategy". Every team seemingly has a plethora of K+/inning guys now. Does it help, hinder, or have no effect?
AndrewKoo
12/18
Thanks! That's a good question I hadn't thought of. My initial guess is that it has minimal effect; a strikeout pitcher might allow less balls in play but at the same time, a platoon affect, if it exists, might help the batter make contact. I'll look into it.
travisbecker
12/18
This is a great article! But it also illustrates the difficulty in being an Oakland A's fan: you don't know at the start of the year what the overall team strategy is, and so it's hard to "follow along at home". For the 2014 season, it will probably be the same as for 2013, since the team is mostly the same. But after that? Or if other teams have already adjusted, and the A's are tanking by June? Beane and his staff are probably already working on the next "juke" they need to make to stay competitive.
ErikFanClubPres
12/19
I think that's what makes it so fun though. It makes seasons like 2012 possible
aschauer
12/18
Hey Andrew, you mentioned in the article that FB% and plate discipline are strongly correlated, I was just wondering if you had the numbers. Just curious! Great article though, fascinating observation.
AndrewKoo
12/18
Thank you. I wouldn't stay strongly correlated, but it is positively correlated. Bill James has a discussion on this in his new handbook. Fly-ball hitters tend to be power hitters, so pitchers might pitch around them, or they'll wait for good pitches to drive. More aggressive hitters will hit the ball into the ground. For a quick measure, in the past 10 years of player seasons with over 250 PA, FB% and BB% have a +0.31 correlation (N=3099).

aschauer
12/20
Thank you!
nyyfaninlaaland
12/18
Really great - one of the most enlightening things I've read this year - guess I have to go back to The Book.
bline24
12/18
More great BP analysis. Great work!
lloydecole
12/18
To quote nberlove, "this is why I subscribe". (Well, this and Professor Parks.) Tremendous work.
mattidell
12/18
This is fascinating.
newsense
12/18
If you weight the TAv in the first table by the matchup frequencies in the second table, you get an average TAV of .260 (league average). If you use the frequencies in the third (A's) table, you get .271. That's a huge advantage just from matchups!
gecko1
12/18
Fantastic article!
Kreylix
12/18
"Fewer ground balls also means fewer double plays: Oakland had the second-lowest number of them." - I think this matters more than you stated.

Great article!
AndrewKoo
12/19
Thank you all for your reading and your comments!
NathanAderhold
12/19
So cool. Great work, Andrew.
pjbenedict
12/19
Wow indeed.

On the one hand, kudos... on the other hand, shame on you for making it harder for the underdog ;-)
mwright
12/19
This piece is a throw back to an earlier era of BP, which is meant as a compliment.
cbf1985
12/19
Great article. Although it also shows how much smaller the gains from incremental information gains are getting. 15 years ago teams overlooked walks were valuable...now we are down to admittedly gains based on batted ball types.
Michael
12/19
I think there are more groundballers now because (1) umpires are calling more low strikes, (2) pitchers are therefore throwing more low pitches, and (3) low pitches are more likely to produce groundballs. See Jon Roegele's article in the latest Hardball Times annual book.

Oh, and excellent article, adding my voice to the chorus.
jroegele
12/19
Great read Andrew - very well researched and presented.

Yes I have to wonder if Beane timed this partly in conjunction with seeing the growing focus on pitchers keeping the ball down around the league. I think the split I found for above 2.5' off the ground/below 2.5' off the ground changed from 46%:54% at the start of the PITCHf/x era to 41%:59% last year. Combined with the strike zone expansion to the bottom of the zone, and pitchers are generally trending toward more ground balls. So as you cited, with the smart platooning/switch-hitting that they have collected, and FB hitters, they've hit on the benefits that you noted and were in The Book. Seems possible anyway!
rawagman
12/19
Great article. Complex stuff and easy to read presentation. You pulled off the Daily Double!
HawaiiFO
12/19
That clue on the clever twist got me looking at the A's entire pitching staff including new pitching acquisitions. It seems that the A's may be trying to be even more extreme with outliner high FB% pitchers of all types than even the high FB% batters.

I only see two or three total pitchers below average ( Gray & Johnson & Otero if he makes the team) but nine pitchers in the 40% range which would seem to be around the standard deviation mark.

New pitchers Kazmir, Abad and Lindblom also fit the 40% mark and Pomeranz and Gregerson are average. So A's hitters are 60% deviant FBers but A's pitchers could be 75% deviant FBers.

However Johnson (25%) replacing Balfour (45%) does not fit the pattern. Can there be more here that meets the eye? Perhaps Ninth inning leads are better protected with extreme GB% but ( 1st-8th, either team is ahead) is better otherwise. Great job. I imagine this article will spawn many more on the subject.
AndrewKoo
12/19
I avoided putting a number on their fly-ball pitching distribution, since one starter can move the needle quit a bit, and there's no magic number on what qualifies as pitcher as high-FB. (We'd also want "true" FB%, by regressing over their career.) And there's a certain amount of error too, as BIS and Retrosheet/MLBAM differ slightly on classifications. But qualitatively, I'm comfortable calling them a FB-inclined staff.

I think there's always more that meets the eye. It would be narrow thinking to assume this is Oakland's only (or even primary) acquisition tactic.

And thanks - I really hope this does spawn more writing.
HawaiiFO
12/19
Athletics Nation already has a post up analyzing your analysis. It's already spreading.....
markpadden
3/04
I could be missing something here, but I don't really follow how stocking up on flyball hitters qualifies as exploiting an inefficiency. I understand that flyball hitters are far better than groundball hitters, on average, but unless this not properly reflected in industry-standard projection systems, that fact by itself is not going be an exploitable inefficiency. Unless you are saying teams are overly reluctant to assemble an extreme flyball team, leading to individual flyballers being slightly undervalued -- but that does not appear to be the point of this article.

The thing that struck me about the MLB-wide data presented is that, after adjusting it (for the fact that high FB hitters and low FB pitchers are better than avg. overall), the so-called platoon edge for flyball hitters appears symmetrical. That is, flyball hitters perform approx. 9 points of TAv better than "expected" (based on the intersections of overall skill levels of each bucket) when facing GB pitchers and 9 points worse than expected when facing FB pitchers. The only way to exploit this is make sure your FB batters have an excessive number of PA (than expected) vs. GB pitchers. But last year, Oakland's FB hitters faced FB pitchers (bad) 13.4% of the time, while they faced GB pitchers (good) only 8.7% of the time -- for a ratio of 1.54. The average MLB team's FB hitters faced a FB/GB pitcher type ratio of 1.30 -- significantly better than the A's. Part of this is that they play in a division with fewer GB pitchers than average, but that should be part of Beane's calculus, no? If he plays a schedule low in GB pitchers, why assemble an extreme team designed to exploit them?

To summarize, I'm not seeing any conclusive evidence the A's are creating more net beneficial matchups than the average team is. A more volatile team? Definitely. I'm wondering if game theory (if you have mediocre talent, you want to maximize volatility) isn't as big or bigger a factor here.