Minnesota Twins News & Rumors Forum
  • Glen Perkins and the modern pitcher

    Before the opener on Monday morning, Glen Perkins sat at his locker, lacing up a new pair of cleats. His locker’s location – one of four corner spots in the spacious clubhouse reserved for the pillars of the team like Joe Mauer and Justin Morneau, a spot that once was home to Joe Nathan’s garb – is standing as a reminder of his status with the team.

    It is hard to envision now but not long ago it appear that his time in the Twins organization was nearing an end.

    Ineffectiveness, followed by a shoulder injury and topped with a semi-contentious battle regarding his arbitration status left him seeming like a pariah with the team. To observers, the left-hander who had grown up on the eastern edges of the Twin Cities and developed in the heart of Minneapolis for the Gophers looked to be on borrowed time. However, a switch to the bullpen combined with good health and a shift in his mental approach resuscitated his career with the Twins.

    And now? Now, he is one of the best relievers in the game. He won the organization’s Good Guy Award this past winter. He has a multi-year contract. He has been asked to play with Team USA in the World Baseball Classic. He has a 5K which he and his wife organize, Fifteen’s 5K, which raises money to benefit Cystic Fibrosis Foundation. He’s got nearly 16,000 followers on Twitter where his interactions range from telling someone the best way to prepare meat for the smoker to joshing that he switching from his current agent to Dwight Yoakam, in response to Robinson Cano’s switch to CAA that is partnered with Jay-Z. Plus - oh man - that coveted corner locker.

    There’s no question he is embraced in Minnesota.

    Over the past few years, the pitcher had become more analytical and thinking more about his craft, using sabermetric theories and tools as he wields the ball. Perkins says he has digested the information and ideas to add another level his game.

    Approaching a player and raising some timely statistical analysis is not a new shtick. In 2010, ESPN broadcaster Jon Sciambi wrote about his rather humorous encounter with Chipper Jones. The gist of it is Sciambi had presented Jones with some new data that the surefire Hall of Fame third baseman was not aware of: Jones had seen the fewest first-pitch strikes in the league. Sciambi wanted to know why the great hitter still swung at a high amount of first pitches seen. The data surprised Jones to the point where he circulated the findings around the clubhouse. The rest of the team, meanwhile, appeared well aware of this phenomenon.

    In a similar fashion Perkins, who was lacing up his fresh field kicks and readying himself for a new season of baseball, was asked if he was familiar with a recent study on Fangraphs.com. The piece studied Pitchf/x data for pitchers who had the highest percentage of pitches taken in the strike zone but was called balls. As the game’s foremost victim of squeezing, Cleveland’s Justin Masterson received the primary attention but Perkins was informed that he could find his name within the top ten as well.

    “So that obviously means I’m getting screwed,” Perkins replied a bit quizzically after being presented with the study’s synopsis.

    Well, yeah, by some measurements.

    The bulk of the data for the research was collected while Perkins was a starting pitcher, over 44 starts between 2007 and 2010. Still, that carried over into his first year as a full-time reliever. According to StatCorner.com’s data, in 2011, when opponents did not swing at in-zone pitches – those that are inside the strike zone as measured by Pitchf/x cameras – were called for a ball a whopping 25.7% of the time. By comparison, the rest of the league averaged 16.7%.

    A quarter of all pitches that should have been called strikes not being called strikes two years ago? Did the Perkins notice any of this?

    “To be honest with you, I feel like I get a pretty good zone. I mean, I throw in and I feel like there are some times when I get the pitches off the plate. I don’t feel like they are tight on me,” said Perkins. “I feel like my slider gets squeezed. Like I throw some on the inner half to righties that a catcher and it will be an inch or two in off the plate and that those catch the corner, the front corner of the plate, and sometimes I just don’t get those. That’s mainly it.”

    The speculation regarding his slider probably has some traction. After all, that cuts across the zone rather quickly, often making a catcher move and a hitter flinch. If you look at his strike zone map from TexasLeaguers.com which dates back to before the 2011 season, you can see quite a few of his sliders on the inside portion of the plate to righties were called for a strike. Conversely, there’s a significant grouping of sliders at the lower portion of the zone – right where Perkins senses he may be getting squeezed – that are literally borderline pitches that were not called for a strike.



    Pitchf/x, a system of cameras designed to capture and catalogue the release point and where the pitch crosses the plate – while measuring velocity, spin and other data in-between – is not without its flaws. For starters, across baseball each stadium’s cameras are slightly different and therefore pump out slightly different results – some minuscule, some noticeable.

    Additionally, strike zones are not all the same for all hitters. That needs to be taken into account. So while Pitchf/x data suggests Perkins was hosed significantly in 2011, Inside Edge’s data refutes that. Their data – which is used by many major league teams including the Minnesota Twins and is compiled by video scouts – says that between 2011 and 2012, 90.9% (290 of 319 in-zone pitches taken) were called strikes compared to the 90.6% league average.

    Then there is the categorization of pitches.

    “I don’t ever throw changeups and it classifies them as changeups,” said Perkins. “If I go look at Brooks Baseball [BrooksBaseball.net] after a game and I see it shows I threw two changeups, obviously I don’t throw that pitch so I know none of them are changeups.”

    While that aspect of the game has improved substantially, there are misses – such as Perkins’ slider being called a changeup. In 2011, Fangraphs.com’s Pitchf/x database says that 17.7% of his offerings were changeups. But even the human element gets fooled too. Fangraphs.com’s other pitch database, one provided by Baseball Info Solutions and their team of video scouts, says that he threw 1.4% changeups in 2011.

    Rather than being satisfied with the numbers stating he was getting screwed, Perkins continues to ask questions about the study. He wants to understand why instead of taking it at face value. Did the study break it down by pitch type? Is that available? He offers up several hypotheses as to why he may be getting squeezed, why Pitchf/x is not categorizing his pitch correctly and deliberates on the meaning of it all.

    This brought him full circle to his career’s rebirth which coincided with his exploration of a more analytical approach to the game.

    “I didn’t get into any advanced stats until 2010 when I was in Triple-A and struggling,” said Perkins. “It was a combination of [Ryan] Vogelsong and when Brandon McCarthy came back and I started to hear about those guys who were injured or unsuccessful and then they adjusted their pitch selections and focused doing different things. That got me thinking, maybe it’s something I should do.”

    It doesn’t hurt, too, that Perkins returned equipped with a mid-90s fastball whereas the other two starters barely get out of the 80’s -- quite the advantage before even considering the added layer of sabermetric insight.

    “Part of it is that I throw harder,” Perkins acknowledges, completely ignoring the fact that his slider bends through space and time. “But I know what I don’t want to do. I know what results I don’t want. You can’t obviously get ideal results across the board all the time, but I know that I don’t want to get fly balls and I don’t want to walk guys.”

    Of course, Perkins knows that there is more to his success than just knowing the sabermetric basics – like getting ground balls and not walking people. There needs to be an ability to execute pitches. Consider Brian Bannister. Prior to the emergence of McCarthy and Vogelsong, Bannister’s foray into the stat community drew attention but, ultimately, because of his natural ability limits Bannister could not practice what he preached as well as the other three pitchers.

    “It was: This is what I need to do. I knew early last year I had a higher ERA but I knew that I had given up base hits, I had given up a two-run homer to Hamilton, I knew that over time that if I continued to execute my pitches, my ERA would end up matching up with my FIP and that I do not need to change what I’m doing because I have an eight [ERA] 4 innings into the season. Over the course of the season – and it did -- it will even out.”

    Now, with his shoes fastened and prepared to join the rest of his team for warm-ups, Perkins concluded the advanced stats discussion with a refreshing simplification of the art that has inspired numerous websites to generate charts, graphs, theorems, percentages and raw numbers to better understand a position that teams pay millions upon millions for:

    “It is obvious stuff in hindsight, but keep the ball down, try to get ground balls, don’t give up fly balls and don’t walk guys.”
    This article was originally published in blog: Glen Perkins and the modern pitcher started by Parker Hageman
    Comments 14 Comments
    1. Oldgoat_MN's Avatar
      Oldgoat_MN -
      Thank you Parker. Fun to read that we aren't the only ones obsessing about stats.
    1. Seth Stohs's Avatar
      Seth Stohs -
      That was a great article. I think that Bannister was kind of the first one to talk about advanced stats and thinking through what he did on the mound. I know that Max Scherzer is another one who acknowledges that he jumps into the stats. I'm really glad that you got some time one-on-one with Perkins and were able to answer these kinds of questions.
    1. spycake's Avatar
      spycake -
      Nice work, Parker! Fun read.
    1. Brock Beauchamp's Avatar
      Brock Beauchamp -
      Fantastic article, Parker.
    1. ashburyjohn's Avatar
      ashburyjohn -
      Great to see athletes not being immediately dismissive of this. I wonder who most influenced him, in accepting an analytic POV.

      I remember Glenn Gostick going over his hand-typed sheets of stats at the local SABR confab in the '80s. In every generation, there are always at least a few trying to do this. Maybe it's gaining more traction.
    1. mike wants wins's Avatar
      mike wants wins -
      Nicely written, great interview.
    1. YourHouseIsMyHouse's Avatar
      YourHouseIsMyHouse -
      Glen Perkins is going to be my new favorite Twin since Denard was traded. Let's see a sub 3 ERA, lots of Ks, and saves!!
    1. LaBombo's Avatar
      LaBombo -
      Quote Originally Posted by YourHouseIsMyHouse View Post
      Glen Perkins is going to be my new favorite Twin since Denard was traded. Let's see a sub 3 ERA, lots of Ks, and saves!!
      Settle for a win instead?
    1. jokin's Avatar
      jokin -
      Quote Originally Posted by YourHouseIsMyHouse View Post
      Glen Perkins is going to be my new favorite Twin since Denard was traded. Let's see a sub 3 ERA, lots of Ks, and saves!!
      Sorry to rain on the parade, but if he sustains a pitching-level like today's outing he will be gone by July.
    1. Jim H's Avatar
      Jim H -
      I think that the big problem with f/x data and grading umps based on this, is that the plate has depth. Like Perkins said, where umps miss his strikes is when they catch the front part of the plate and the catcher doesn't end up catching the ball behind the plate. I wonder if f/x data really reflects the depth of the plate. I know that the ball/strike stuff that annoucers use doesn't always seem to reflect that. I know that some people here talk about framing pitches, but where the ball is caught doesn't always reflect where the ball was when/or if it crossed the plate.

      I know umpires are being graded on their ball and strike calls. I hope that the data used isn't based on where the ball is caught.
    1. YourHouseIsMyHouse's Avatar
      YourHouseIsMyHouse -
      Quote Originally Posted by LaBombo View Post
      Settle for a win instead?
      I was happy with that
    1. Shane Wahl's Avatar
      Shane Wahl -
      This was a fantastic read!
    1. Mr. Brooks's Avatar
      Mr. Brooks -
      Quote Originally Posted by Jim H View Post
      I think that the big problem with f/x data and grading umps based on this, is that the plate has depth. Like Perkins said, where umps miss his strikes is when they catch the front part of the plate and the catcher doesn't end up catching the ball behind the plate. I wonder if f/x data really reflects the depth of the plate. I know that the ball/strike stuff that annoucers use doesn't always seem to reflect that. I know that some people here talk about framing pitches, but where the ball is caught doesn't always reflect where the ball was when/or if it crossed the plate.

      I know umpires are being graded on their ball and strike calls. I hope that the data used isn't based on where the ball is caught.
      Pitch F/x tracks where the ball crosses the plate.
      It's incredibly accurate. Within 1/2", and 1 mph at most ballparks.
      Here is a good link with data measuring it's accuracy at each ballpark. (And this report is almost 2 years old, they are constantly tweaking it for accuracy, so it's safe to assume that it's even more accurate today.)

      http://www.baseballprospectus.com/ar...rticleid=13109
    1. Mr. Brooks's Avatar
      Mr. Brooks -
      Quote Originally Posted by Jim H View Post
      I think that the big problem with f/x data and grading umps based on this, is that the plate has depth. Like Perkins said, where umps miss his strikes is when they catch the front part of the plate and the catcher doesn't end up catching the ball behind the plate. I wonder if f/x data really reflects the depth of the plate. I know that the ball/strike stuff that annoucers use doesn't always seem to reflect that. I know that some people here talk about framing pitches, but where the ball is caught doesn't always reflect where the ball was when/or if it crossed the plate.

      I know umpires are being graded on their ball and strike calls. I hope that the data used isn't based on where the ball is caught.
      Yes it does. Pitch F/X rules a pitch a strike when the ball crosses any part of the plate in the zone.
©2014 TwinsCentric, LLC. All Rights Reserved.
Interested in advertising with Twins Daily? Click here.