James Shields: Becoming an Ace

This might seem like a weird time to be writing an article about James Shields experiencing a breakout season, given that his 3.95 ERA this season is over a run higher than the 2.82 mark he posted last year. From a traditional standpoint, if Shields made an ace like leap, it was last season, and this year he’s simply regressing back to something less than ace-worthiness.

But, of course, I’m not exactly a big fan of ERA, especially when it comes to evaluating changes in pitcher performance. There are so many variables in ERA that a pitcher has little or no control over, and evaluating a pitcher by the amount of “earned runs” (whatever that means) he allows often causes us to miss real changes that do tell us something about what we should expect in the future. That looks to be the case with James Shields right now.

If you hadn’t looked at the batted ball leaderboards lately, you might not have noticed that Shields is currently fourth in the Majors in ground ball rate, coming in right between Jake Westbrook and A.J. Burnett. Shields has never been much of a GB pitcher, brining a career 43.8 percent ground ball rate into the 2012 season. This year, though, Shields has been one of the most extreme ground ball pitchers in all of baseball, and perhaps more amazingly, he’s done it while slightly increasing his overall strikeout rate from last season.

Strikeouts and ground balls generally don’t go together, as pitchers tend to specialize in getting one or the other. In fact, the average strikeout rate of the other nine qualified starting pitchers in the top 10 in GB% is just 13.8 percent, well below the league average and 10 percentage points below Shields’ mark. A.J. Burnett is the only starter in baseball this year with a top GB rate over 56 percent who is also posting an above average strikeout rate. The pitcher who can simultaneously get ground balls and strikeouts is a rare bird indeed.

A pitcher who combines very high ground ball and strikeout rates in the same season generally produce results that would put them in Cy Young contention in most years. In order to most accurately show the effectiveness of these types of pitchers, I went through and compared every qualified starting pitcher season since 2002 to that year’s league average ground ball and strikeout rates, and then determined how many standard deviations away from the mean that they were in those two categories. For instance, Derek Lowe is the most extreme ground ball pitcher we’ve seen in the last decade, while Randy Johnson is the modern day strikeout king, and both put up seasons that were three and a half standard deviations from the mean.

However, the king of combining these two factors was indisputably Brandon Webb. In 2003, his ground ball rate was 3.0 standard deviations above the norm, while his strikeout rate was 1.4 standard deviations from the mean. Randy Johnson in 2002 (3.5 StDevs in K%, 0.6 StDevs in GB%) was the only other pitcher in the last 11 years to put up GB and K numbers that were a combined four standard deviations from the norm. In fact, only 88 of the 995 (8.8%) qualified pitcher seasons since 2002 have resulted in a pitcher posting GB/K numbers that combined to be more than two standard deviations from the league norm in that year.

Right now, James Shields combination of GB rate (1.9 StDevs) and K rate (0.9 StDevs) is the 24th best mark since 2002 by when you look at ground ball and strikeout rates together. And pitchers who put up these kinds of GB/K rates at the same time are generally dominant forces on the mound.

There have been 26 pitchers who have posted a combined StDev of between +2.5 and +3.1, which gives us a decent sample of pitchers who as a group have the same 2.8 StDev total as Shields. In those 26 seasons, these pitchers combined for an ERA-/FIP-/xFIP- line of 74/73/72. To put that in perspective, Roy Halladay‘s career line is 72/75/74. Basically, the group of pitchers who combined for something like the amount of GBs and Ks that Shields is getting now put up results that would fit right into Halladay’s career.

Shields’ home run problem may prevent him from ever reaching Halladay-esque levels, but his HR/FB rate isn’t going to stay over 19 percent for much longer, and the fact that he’s added a significant increase in ground balls while keeping his BB/K rates stable suggests that Shields is simply getting better as a pitcher. It might not have shown up in the results so far, but Tampa Bay should be pretty thrilled about what they’re seeing from him so far this season.

He probably won’t keep getting ground balls at a Jake Westbrook type level, but he doesn’t need to sustain that kind of rate in order to become a legitimate ace. If he can keep his ground ball rate between 50-55% while also keeping his walk rate between 6-8% and his strikeout rate between 22-24%, he’d put himself in this group of pitchers who have put up those marks over the last 10 years: Felix Hernandez (2009, 2010, and 2011), Adam Wainwright (2010), and Josh Johnson (2009).

Shields’ ERA is deceiving – the way he’s pitching he’s right now suggests that he’s taken another step forward, and is on the verge of becoming one of the very best pitchers in all of baseball.

Print This Post

Dave is a co-founder of USSMariner.com and contributes to the Wall Street Journal.

34 Responses to “James Shields: Becoming an Ace”

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

    The one thing that is concerning me is that he is giving up more hits this season. In light of your info, though, I expect that to regress a bit. Probably luck driven.

    Vote -1 Vote +1

    • pudieron89 says:

      More groundballs generally means more singles. Many of us have always groaned at Shields’ “hittability” but one thing to consider this season is the patchwork defense he’s been playing in front of as opposed to the more consistent diamond of previous years. Until Longoria gets back it’s a revolving door of Sean Rodrigues, Elliot Johnson, and defensive scrubs at 3B/SS/2B.

      Vote -1 Vote +1

      • ThePartyBird says:

        Both of those guys are pretty good defenders (and Rhymes is also just a hair below average probably). You’re basing their defense on what their UZRs look like right now – while that /may/ explain a bit of Shields’ luck, it’s certainly not enough to conclude that he’s only getting hit because of “patchwork defense”.

        Shields does have a pretty high career BABIP for a guy who was a flyball pitcher until this season. He’s still a great pitcher, but he does seem to be the type who will consistently give up more hits than average, especially as a groundball pitcher.

        Vote -1 Vote +1

  2. pudieron89 says:

    Great insights Dave, thanks.

    Vote -1 Vote +1

  3. baty says:

    His Pitch Type / Pitchf/x Pitch Type charts are all over the board the last few years. He’s either in a constant state of evolution, or it’s just really hard to tell what and when he’s pitching something.

    Vote -1 Vote +1

    • Dave Cameron says:

      If you look at the movement plots, it looks like his two-seam/four-seam/cutter/slider are all part of one big blob, so it’s pretty tough to definitively say that he’s throwing x% cutters and y% sliders. Basically, he has a distinct curve, a distinct change, and then four four pitches that are pretty similar. Probably contributes to how hard it is to hit him since he has so many options.

      Vote -1 Vote +1

      • Erik Hahmann says:

        Pitch F/x says he’s throwing a slider, but it’s really the cutter. He never speaks of throwing a slider.

        Vote -1 Vote +1

      • monkey business says:

        If your two-seam/four-seam/cutter/slider are apparently the same pitch, in what sense is it not just one pitch?

        Vote -1 Vote +1

      • nolan says:

        Monk, you can grip the ball differently and put pressure on it in a slightly different way to make it break more or less or in a different direction. Buerhle talked about it in a Laurila interview I believe.

        Vote -1 Vote +1

  4. Sam says:

    I know that hitters generally experience an increase in power and walk rates as they age, but what trends can we expect for pitchers? Shields is 30 is it reasonable to expect for the ground ball trend to continue?

    Vote -1 Vote +1

  5. Sandy Kazmir says:

    Sun Sports ran a great graphic during his last start showing that he’s mixing pitches thusly:

    CH: 30%
    FF: 25%
    FC: 25%
    CU: 20%

    Shields fell in love with his cutter before he had honed it, but now it’s proving to be a capable and effective pitch. The change is still his meal ticket, but batters not being able to key on the fastball has been splendid, and the curve is a great pitch to keep batters off balance even if it’s only a 60 pitch.

    Vote -1 Vote +1

  6. RC says:

    “There are so many variables in ERA that a pitcher has little or no control over, and evaluating a pitcher by the amount of “earned runs” (whatever that means) he allows often causes us to miss real changes that do tell us something about what we should expect in the future.”

    So, basically, its exactly like FIP.

    Vote -1 Vote +1

  7. chuckb says:

    If the change in Shields’s GB rate are here to stay, isn’t it reasonable to conclude his HR% will regress to the mean. It’s very high this year, much higher than league average, so the likely reduction in his HR/9 should improve his numbers even more.

    Vote -1 Vote +1

    • Nathaniel Dawson says:

      Actually, his home run rate is pretty much exactly league-average this year. .029 per batter faced versus the league average of .028. Or if you prefer frequency per 9 innings, 1.11 versus league average of 1.08. Splitting hairs if you want to call him anything other than league average in home run rate.

      Vote -1 Vote +1

  8. rotofan says:

    Since groundball rates for pitchers tend to stabilize fairly quickly, I think you have made a convincing case that Shields has taken another step forward. I’m more apt to comment on posts on which I disagree or can offer a different perspective, but in this case I’ll make an exception: Well done.

    Vote -1 Vote +1

  9. Juan Chapa says:

    An Ace is a STOPPER! That is, he is supposed to stop losing streaks.
    I think people make too much about statistics, and too little on
    analysis and interpretation of the stats. For example, if a pitcher
    starts out with an average of less than two runs for his first ten
    games in a season, and then gives up 8 – 10 earned runs in the
    11th game, that is going to skew the stats sample. In addition,
    if there are weak links on the rotation (injuries etc.), opposing
    teams will try to match their ace against yours, and take their
    chances with the rest of their rotation and bullpen, if they’re
    strong. This will also skew an ace’s stats. But, it should not
    affect the ERA much. Look at King Felix’s ERA. Also, Kevin
    Brown was a power pitcher, but had one of the most wicked
    sinkers in baseball. At one time, he was the best paid player
    which goes to show you that a fast ball and sinker work well
    together. Halladay is the best pitcher in baseball. He has 4
    pitches, uses different speeds, throws all of them for
    strikes, and throws a fast game. He does not allow batters
    much time to think, or make adjustments. In business,
    successful people copy SUCCESSFUL persons. Its dog eat
    dog. Young pitchers should do the same, instead of trying
    to reinvent the wheel! There is no such thing as a pitcher’s
    duel. Its the pitcher vs. batter. And, if batters are too
    succesful against you, it would not hurt to copy cat
    Halladay. It just migh prevent a return tour to the minors.

    Vote -1 Vote +1

    • JDanger says:


      whoa, dude.

      Vote -1 Vote +1

    • Nathaniel Dawson says:

      Can’t say much about the content, but I can at least give you props for your grammar and spelling.

      One grammar mistake (should have had a comma in this sentence: “At one time, he was the best paid player(,) which goes to show you that a fast ball and sinker work well together.”

      A spelling error here: “…opposing teams will try to match their ace against your(‘)s…”

      One more spelling error here: “It just migh(t) prevent a return tour to the minors.”

      Altogether, not a bad effort. A few slight errors, but for the most part, well within the bounds of reasonable internet commenting. Certainly, content is important, but you at least have a good grasp of proper grammar and spelling.

      Vote -1 Vote +1

    • Clancy W says:

      Hepped up on goofballs everybody! High as a kite…

      Vote -1 Vote +1

  10. monkey business says:

    Looking at this type of analysis appears like an argument for tERA, but his tERA says that his ERA is dead on (well, off by 0.2).

    But even if we do give you FIP but peek at GB%, the we still have the problem that GB pitchers are LD pitchers, and his LD % has stayed relatively flat from previous years to 2012. I’d be willing to bet that:

    1) What is getting scored as GBs includes several very well hit balls that are flying (though not literally) through the infield and unlikely to get caught and not bad fielding luck
    2) His LD% goes up while his GB% goes down as the season goes on.

    Vote -1 Vote +1

    • Antonio Banana says:

      yea, why don’t we look at tERA more? Let’s pretend you give up a home run with the wind blowing out hard, that’s sort of luck is it not? So instead let’s also include LD, GB, and FB. I think that paints a much more complete picture. If a guy gives up 4 straight doubles drilled in the gap, FIP doesn’t recognize it. Says it was luck. The only luck would be that they weren’t HR or hit right at someone.

      Second thing, can we not measure the speed of the ball off the bat? How about we further divide BIP into how hard they are hit off the bat? A GB is good unless it’s smoked up the line. You could quantify that if you knew the speed of the ball off the bat.

      Vote -1 Vote +1

  11. Jeffrey says:

    Ace means, literally the best pitcher on the teams pitching staff. I don’t see the title relating to the article written. I thought this was going to be an article comparing shields to price, Moore, etc.

    An ace isn’t just a top pitcher in baseball. James McDonald is an ace, and Cliff Lee isn’t. Curt Schilling wasn’t an ace in 01 and 02 either.

    Just the wrong word to use.

    -12 Vote -1 Vote +1

    • GorillaKilla says:

      And the best hitter is the one with the most RBI, right? This was just a dumb attempt at criticism. I hope Cole Hamels has fun trying to land a huge contract this offseason since he’s only a # 3 starter.

      Vote -1 Vote +1

      • Jeffrey says:

        Christ, look up the definition. I was being literal, not facetious or saying anything even close to what you said. An ‘ace’ by definition is the best pitcher on a teams pitching staff. Literally. If you want to say James Shields is becoming a legitimate front line starter, or that he is becoming a top pitcher in baseball, then this article would be correct. But the use of the word ‘ace’ is thrown around incorrectly all the time in baseball and on forums.

        Not trying to be a smart-ass. It’s simply the wrong word to use for this posting.

        Vote -1 Vote +1

    • Johnny Big Potatoes says:

      Actually, his use of “ace” is how people in the industry use the word.
      An “ace” is generally considered to have several above-average pitches, an elite level of command and control, durability, and an extended track record of success. Obviously this still leaves quite a bit of gray area, but right now there are only roughly ten or twelve or so true aces in baseball, and I assure you, James McDonald is not one of them.
      Whether a guy is the best pitcher on any one staff at any given time is irrelevant. It is not how the word is used in baseball among scouts and front offices.

      Vote -1 Vote +1

      • Jeffrey says:

        I didn’t write the definition. Maybe baseballs front offices should let Merriam-Webster in on this, or maybe they should change their usage of the word ;)

        Vote -1 Vote +1

      • Simon says:

        Jeffrey – if everyone apart from you uses a word one way, it’s probably going to be you that has to change, even if you can find a dictionary that supports your argument. If you’re actually not bothered and are just trolling, feel free to bash on.

        Vote -1 Vote +1

  12. Jeffrey says:

    When did we in society decide to ignore the dictionary and make up our own meanings to words? It is a mis-interpreted word.

    Vote -1 Vote +1

  13. Will says:

    Great article about Shields! Do you think that with all these factors – increase GB rate coupled with the K rate – that much of the blame for a high ERA can be attributed back to the defense that is supporting him? Last year’s Rays defense was tops in the league and while it still ranks relatively high in 2012, do you think he’s suffering a bit from the loss of Longoria and the reshuffling of Sean Rodriguez from short to third?

    Vote -1 Vote +1

  14. Skin Blues says:

    So, in 8 starts since this article he’s had a 5.15 ERA and 1.72 WHIP. 72 hits in 50 IP… at what point does it stop being attributed to bad luck? K/BB is still respectable but for whatever reason he gives up a lot of solid contact. It would be really nice to have batted ball data include average speed off the bat.

    Vote -1 Vote +1

    • The Truth says:

      So, in the 7 starts since this comment he’s had a 2.98 ERA and 0.82 WHIP. 32 hits in 51.1 IP… at what point does it start being attributed to good talent? K/BB is still respectable but for whatever reason people don’t want to admit that Shields just might be as good as some of the bigger names.

      Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>