Production Per Swing in 2012

There are rate stats for just about every kind of opportunity a hitter faces in a game. Batting average tells you how often a player reaches base via a hit. On-base percentage tells you how often a player avoids making an out per plate appearance. But what about swings as opportunities?

Last year, I played around with the idea of production per swing. The idea was to examine what hitters gave the most value when they took a swing. The methodology was pretty simple: calculate the Weighted On-base Average (wOBA) each hitter generated using their swings — instead of plate appearances — as the denominator*.

Of course, there is a healthy correlation between actual wOBA and wOBA per swing (.83 in 2012), but less so Isolated Power (ISO). (wOBA/swing and ISO share only a .53 correlation.) Some of the results may not be all that surprising, but many certainly are.

Let’s first look at the top-25 so far this year:

Top 25 (minimum 100 PAs)

Player Swing% (pfx) Actual wOBA wOBA/swing
Josh Rutledge 50.9% 0.398 0.214
Salvador Perez 49.3% 0.358 0.212
Mike Trout 39.8% 0.422 0.209
Joey Votto 37.4% 0.449 0.205
Jeff Keppinger 42.6% 0.355 0.204
David Ortiz 41.6% 0.422 0.204
Melky Cabrera 47.0% 0.386 0.203
Alex Rios 45.7% 0.357 0.203
Manny Machado 45.6% 0.337 0.202
Carlos Ruiz 47.0% 0.407 0.195
Chris Denorfia 39.0% 0.351 0.194
Billy Butler 42.1% 0.371 0.194
Miguel Cabrera 48.0% 0.412 0.192
Aaron Hill 43.7% 0.366 0.192
Michael Brantley 39.7% 0.319 0.192
Troy Tulowitzki 42.1% 0.360 0.192
Ryan Braun 48.2% 0.417 0.191
David Cooper 44.1% 0.335 0.191
David Wright 40.8% 0.377 0.190
Scott Podsednik 40.4% 0.324 0.190
Allen Craig 43.1% 0.385 0.190
Chipper Jones 43.5% 0.369 0.190
Robinson Cano 48.7% 0.382 0.190
Kevin Frandsen 43.8% 0.347 0.190

Colorado Rockies rookie Josh Rutledge so far has generated the most production per swing for players with at least 100 plate appearances this season. Only 23 years-old, Rutledge has posted a .398 wOBA since being called up to fill in for the injured Troy Tulowitzki. Rutledge has certainly not been shy about swinging the bat, offering at more than half the pitches he’s seen. He’s produced 33 singles, 14 doubles, four triples and seven home runs with his 314 swings. Sure, he plays his home games at Coors Field, but he still managed a 142 wRC+ with only a 2.2% walk rate.

The Royals 22-year-old rookie, Salvador Perez, comes in at No. 2 with a .212 wOBA/swing. Like Rutledge, the young catcher tends to hack — his swing percentage is nearly 50% — but those hacks have lead to a 126 wRC+.

Top 25 (minimum 300 PAs)

Player Swing% (pfx) Actual wOBA wOBA/swing
Mike Trout 39.80% 0.422 0.209
Joey Votto 37.40% 0.449 0.205
Jeff Keppinger 42.60% 0.355 0.204
David Ortiz 41.60% 0.422 0.204
Melky Cabrera 47.00% 0.386 0.203
Alex Rios 45.70% 0.357 0.203
Carlos Ruiz 47.00% 0.407 0.195
Chris Denorfia 39.00% 0.351 0.194
Billy Butler 42.10% 0.371 0.194
Miguel Cabrera 48.00% 0.412 0.192
Aaron Hill 43.70% 0.366 0.192
Michael Brantley 39.70% 0.319 0.192
Ryan Braun 48.20% 0.417 0.191
David Wright 40.80% 0.377 0.190
Allen Craig 43.10% 0.385 0.190
Chipper Jones 43.50% 0.369 0.190
Robinson Cano 48.70% 0.382 0.190
Paul Konerko 44.10% 0.376 0.188
Andrew McCutchen 46.70% 0.401 0.188
Martin Prado 38.00% 0.343 0.188
Prince Fielder 43.30% 0.395 0.186
Nick Markakis 40.90% 0.357 0.186
Jose Altuve 43.00% 0.329 0.184
Joe Mauer 34.60% 0.370 0.184
Yadier Molina 51.50% 0.378 0.184

When we focus on players with at least 300 plate appearances, we see a list mostly of everyday players. Mike Trout and Joey Votto have been one and two in the league for most of the season in wOBA, and both repeat here. Trout and Votto also sport swing rates below 40%, emphasizing how efficient they are when they swing.

The third-ranked player, however, is the Rays’ Jeff Keppinger. Keppinger is playing for his sixth team in eight years. The 31-year-old is having a career year for Tampa Bay, and has posted a 129 wRC+ due in large part to his .204 wOBA/swing. Keppinger is swinging at fewer than 43% of pitches, but he’s made the most of those swings. That’s great value and production for a player asked to man first, second and third base at times this season.

Braves infielder Martin Prado has a very low 38% swing rate, but he’s managed to produce a .188 wOBA/swing and 84 wRC.

Bottom 25 (minimum 300 PAs)

Player Swing% (pfx) Actual wOBA wOBA/swing
Brendan Ryan 44.50% 0.256 0.098
Rod Barajas 52.60% 0.258 0.102
Carlos Pena 42.90% 0.298 0.106
Clint Barmes 52.10% 0.242 0.106
Brian Bogusevic 46.10% 0.275 0.109
Cliff Pennington 44.40% 0.263 0.110
Jordan Schafer 42.80% 0.274 0.111
Jeff Francoeur 52.30% 0.272 0.113
Kelly Johnson 47.10% 0.301 0.117
John Buck 43.50% 0.286 0.118
Dan Uggla 42.50% 0.320 0.119
Sean Rodriguez 48.20% 0.271 0.119
Justin Smoak 43.30% 0.253 0.120
Daniel Descalso 46.00% 0.276 0.121
Dee Gordon 46.30% 0.262 0.121
Josh Thole 44.90% 0.257 0.122
Nyjer Morgan 46.00% 0.275 0.123
Jemile Weeks 38.40% 0.276 0.124
Gregor Blanco 42.80% 0.309 0.125
Gordon Beckham 48.90% 0.283 0.126
Jose Tabata 47.20% 0.273 0.126
Brandon Crawford 49.40% 0.279 0.126
Elliot Johnson 47.30% 0.291 0.126
Everth Cabrera 40.30% 0.302 0.127
Chris Davis 53.70% 0.326 0.128

With every leader board is the laggard board. And Brendan Ryan (.098) has the worst wOBA/swing so far this year. Ryan has a 60 wRC+ and really doesn’t offer much value to the Mariners, outside of his defense (first among shortstops in Defensive Runs Saved and second in UZR/150).

The Rays’ Carlos Pena has swung the bat nearly 1,000 times this season, but those swings haven’t been very productive (.106 wOBA/swing). Pena has a wOBA below .300 — and even adjusting for park and league, he’s still been 10% worse than the league average at the plate.

Hacker extraordinaire, Jeff Francoeur, doesn’t disappoint with his 52% swing rate. But but he doesn’t do much with those swings (.113 wOBA/swing). After posting a 117 wRC+ last year for the Royals, the 28-year-old came back to earth this season in a rough way. Francoeur currently has a 68 wRC+, fourth-worst in baseball.

Finally, here is the top 25 for the big-time swingers: those swinging above the 45.2% league average:

Top 25 (minimum 300 PAs and >= 45.2% Swing Percentage)

Player Swing% (pfx) Actual wOBA wOBA/swing
Melky Cabrera 47.00% 0.386 0.203
Alex Rios 45.70% 0.357 0.203
Carlos Ruiz 47.00% 0.407 0.195
Miguel Cabrera 48.00% 0.412 0.192
Ryan Braun 48.20% 0.417 0.191
Robinson Cano 48.70% 0.382 0.190
Andrew McCutchen 46.70% 0.401 0.188
Yadier Molina 51.50% 0.378 0.184
Giancarlo Stanton 48.60% 0.397 0.183
Carlos Gonzalez 45.70% 0.384 0.182
Adrian Beltre 51.20% 0.381 0.181
Albert Pujols 46.10% 0.369 0.180
Matt Kemp 45.80% 0.391 0.180
Jordan Pacheco 47.60% 0.330 0.178
Omar Infante 47.30% 0.315 0.177
Wilin Rosario 48.50% 0.339 0.175
Ian Desmond 54.70% 0.362 0.175
Aramis Ramirez 51.30% 0.382 0.172
Willie Bloomquist 49.70% 0.303 0.171
Matt Holliday 47.70% 0.382 0.170
Yoenis Cespedes 48.20% 0.359 0.170
Tyler Colvin 50.60% 0.374 0.170
Jay Bruce 45.90% 0.373 0.170
Scott Hairston 46.30% 0.346 0.169
Erick Aybar 50.20% 0.316 0.168

Everyone’s favorite replacement-level player, Willie Bloomquist, made the top-25 list with a .171 wOBA/swing. Yes, he swings almost 50% of the time. And, yes, he walks less than 4% of the time. Still, Bloomquist manages to produce a solid amount of value with those swings. For example: Bloomquist, who has a mere .096 ISO, actually has a higher wOBA/swing than Mark Trumbo (.159) and Bryce Harper (.143). And both of those guys swing as often as Bloomquist.

In terms of examining the value of wOBA/swing, part of it is just amusement. But I think there’s some value in the metric when looking at young players with limited playing time. Or in Bloomquist’s case, understanding what value replacement-level players may provide that other metrics might overlook. No, this doesn’t mean Bloomquist is much more valuable than what other metrics suggest, just that the value he has can be understood in a different way.


Data and wOBA coefficients are current as of Sept. 10, 2012.

*wOBA here was calculated using the following equation: [(1B*.880)+(2B*1.250)+(3B*1.583)+(HR*2.042)]/Swings

Print This Post

Bill works as a consultant by day. In his free time, he writes for The Hardball Times, speaks about baseball research and analytics, consults for a Major League Baseball team, and has appeared on MLB Network's Clubhouse Confidential as well as several MLB-produced documentaries. Along with Jeff Zimmerman, he won the 2013 SABR Analytics Research Award for Contemporary Analysis. Follow him on Tumblr or Twitter @BillPetti.

9 Responses to “Production Per Swing in 2012”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. odditie says:

    Too early in the morning apparently…someone explain this to me.

    How does Keppinger have a higher swing %, lower wOBA, but the same wOBA/swing as D. Ortiz?

    Wouldn’t that mean he is swinging at more pitches per PA and producing less wOBA per PA?

    Like I said…just trying to wrap my head around this…just not clicking for me.

    Vote -1 Vote +1

  2. philosofool says:

    Is this anything other than a novelty? I just don’t understand why we would use swings as a basis of evaluation. Notice that if we do wOBA per pitch taken, we should get a reverse order of the standings, but why is being at the top of either list better or worse? Maybe I’m totally missing something. Also, should you do wRC/swing, not a rate stat divided by a rate stat?

    Vote -1 Vote +1

    • Adam says:

      I think you are misunderstanding how wOBA/swing was calculated. It isn’t wOBA divided by Swing%. Swing% was only included as a point of reference. It wasn’t used in the calculation at all. From the last line of the article: “*wOBA here was calculated using the following equation: [(1B*.880)+(2B*1.250)+(3B*1.583)+(HR*2.042)]/Swings”

      wOBA per pitch taken would not be a reverse order, because presumably you would only include outcomes possible when the batter doesn’t swing (i.e. walking, being hit by a pitch, and striking out looking).

      Vote -1 Vote +1

  3. Nick says:

    This is pretty dumb. There is no value here.

    Vote -1 Vote +1

  4. maskierter muchacho says:

    Of course there is value. This pretty much shows the quality of a swing taken by player x. Think of best mechanics. If player x has very good value per swing but that does not reflect in his actual woba it could be an approach problem or bad patience. Maybe even player x should swing more often if his walk rate is low but his value per swing is good.

    There are tons of inte

    Vote -1 Vote +1

    • philosofool says:

      I’m perfectly happy to deny that this shows anything about the “quality of swing” taken by a player. The value of a swing is highly relative to the count and swing-% doesn’t account of this at all. Furthermore, since “quality of swing” is just the inverse of “quality of takes” by this metric, I don’t see why it’s measuring something interesting.

      On the whole, I have never found a single strong or interesting correlation between batter plate discipline and batter performance. I think it turns out that hitters are varied in their skills and no one set of skills is the one that works (think Dunn, Votto, Hamilton–all great hitters, completely different in their approach and skills). Plate discipline, IMO, is best seen as interesting desciption of what a batter does, not a measure of his quality.

      Vote -1 Vote +1

  5. maskierter muchacho says:

    *There is tons of interesting information one could draw from this.

    Vote -1 Vote +1

  6. Paul says:

    So wOBA is supposed to be a single all-encompassing stat for production. Of course they would correlate, since wOBA is based mostly on balls in play. I feel like there are major bias issues here, maybe double discounting walks, but don’t have time to think very hard about it and I’m not a statistician.

    Vote -1 Vote +1