Skip to content

Bill James thinks Sabermetrics has overrated ground ball pitchers

Mar 22, 2013, 8:35 PM EDT

Derek Lowe AP

Rob Neyer expounded on a Bill James piece (here, behind a paywall) at SB Nation in which James says that proponents of Sabermetrics have “horribly overstated” the case for ground ball pitchers. He cites a handful of elite pitchers — Tom Seaver, Bob Gibson, Randy Johnson, Justin Verlander, among others — and points out that they weren’t very good in the ground ball department. He also cites a handful of ground ball pitchers — namely Chien-Ming Wang and Brandon Webb — who have had serious issues with injuries.

What I have never understood about ground ball pitchers, and do not understand now, is why they always get hurt. Show me an extreme ground ball pitcher, a guy with a terrific ground ball rate, and I’ll show you a guy who is going to be good for two years and then get hurt.

[snip]

They’re great for two years, and then they blow up. Always.

The one exception to James’ analysis is, of course, Derek Lowe, who made at least 32 starts in each season from 2002-11.

I don’t think we have the capability to make a strong conclusion one way or another based on the quality of data we have right now. Presently, there is no differentiation between types of batted balls. There is a vast difference between a dribbler down the third base line and a screamer up the middle. FanGraphs differentiates between infield and outfield flies, which has helped us to better appreciate pitchers like Matt Cain and Jonathan Papelbon. No such distinction is made for ground balls.

There is no way of knowing now, of course, but there may be a link between injury risk and the types of ground balls a pitcher induces. Basing analysis on data that utilizes binary qualitative groups — “on the ground” and “not on the ground” — is far too broad.

(Tip of the cap to David Schoenfield at ESPN Sweet Spot for directing me to Neyer and James.)

  1. proudlycanadian - Mar 22, 2013 at 8:51 PM

    Does James like fly ball pitchers instead? Sluggers love fly ball pitchers.

  2. scatterbrian - Mar 22, 2013 at 8:54 PM

    Tim Hudson?

  3. churchoftheperpetuallyoutraged - Mar 22, 2013 at 9:18 PM

    Favorite comment on the Neyer piece:

    Clearly, the foot injury Chien-Ming Wang suffered while running the bases
    is directly related to his sinkerball style of pitching.

    Also, Derek Lowe is mentioned, and Kevin Brown is another one. So doesn’t Bill James’s “theory” of “They’re great for two years, and then they blow up. Always.” not hold water?

    • Lady Loves Pinstripes - Mar 26, 2013 at 3:51 PM

      But Wang had serious shoulder surgery in 2009, which caused him to miss the entire 2010 season.

  4. rk70 - Mar 22, 2013 at 9:35 PM

    How about Greg Maddux …?

    • Detroit Michael - Mar 23, 2013 at 8:06 AM

      If you click through the link, you’ll see that James acknowledges Greg Maddux as an exception.

      • GoneYickitty - Mar 23, 2013 at 2:53 PM

        What he pretty much said was that they *always* get hurt within 2 years, but there are exceptions. Let’s not mention any fly ball pitchers who have gotten hurt (Chris Young comes immediately to mind as a guy who never gets ground balls and is hurt all the time) and let’s excuse away any ground ball pitchers who haven’t gotten hurt because they don’t fit the point we’re trying to make. For someone who got his reputation from the cold hard use of statistics, it’s pretty weak in my opinion.

  5. paperlions - Mar 22, 2013 at 9:54 PM

    Isn’t the point that if you aren’t going to be striking out a lot of guys, it is better to have them hit it on the ground than in the air? Because most of the guys James listed weren’t flyball pitchers either, they were strikeout pitchers. There are weird statements in there intimating that he’s just going to ignore guys that are good at both striking out hitters and getting groundballs….but he’s NOT going to ignore guys that are good at striking out hitters and getting flyballs. I’d love to see the long list of guys that were awesome that couldn’t strike anyone out and that gave up a bunch of flyballs the last 20 years….but yeah, back in the days of huge parks and a dead ball….why not give up fly balls? They weren’t going anywhere. We don’t play baseball in those conditions anymore.

  6. hackerjay - Mar 22, 2013 at 11:20 PM

    I think it’s pretty unfair to say Webb only had two good years. He was a top five pitcher for like six years.

    • yahmule - Mar 24, 2013 at 12:51 PM

      I think the guy blurts out nonsense pretty frequently, but mostly gets a pass because he’s Bill James.

  7. spudchukar - Mar 23, 2013 at 12:37 AM

    Thanks so much for the acknowledgement that all ground ball, or fly balls for that matter are not the same. This seems so obvious, and yet in Sabermetricland if it bounces it is identical.

    Of course power guys, Gibson, Koufax, Seaver etc are preferable, but they are rare. Pitch to contact isn’t ideal, just practical. If you flounder trying to be Gibson, then try something else, and for some guys, like Kyle Lohse, yes I am still flogging that long deceased equine, have found another method that can be successful.

    James’ mistake, is forced categorization, thus labels.

  8. mrfloydpink - Mar 23, 2013 at 12:40 AM

    Bill James was once a brilliant visionary. In his older years he’s become…considerably less so, let’s say.

    • scoocha - Mar 23, 2013 at 2:58 PM

      He’s just an idiot now. Wrong more than 75% of the time now.

  9. Mark - Mar 23, 2013 at 12:41 AM

    Without doing too much research, my gut instinct would be that groundball pitchers might be more likely to get injured because they throw a slider. You hear broadcasters all the time say it’s a “sinker slider type”, and sliders generally put more strain on the arm. Granted Webb didn’t throw a slider, but Wang did.

    I think James is wrong that groundball pitchers are overrated. Obviously a guy who has a high strikeout rate is better than someone with just a high GB rate, but that doesn’t make a GB pitcher overrated. We know for a fact that pitchers who get groundballs are on average more successful than their flyball counterparts. The Baseball Analysts studied it for several years, and every year the GB pitcher had a better ERA than a flyball guy (everything else being equal) –> http://baseballanalysts.com/archives/2009/02/post_6.php.

    I’m too tired to look, but they have studies for 2009 and include walk rates in addition to K’s and GB, but the bottom line is that groundball pitchers are more successful than flyball guys with everything else being equal. Of course an elite strikeout pitcher like Verlander is more valuable, but that has more to do with strikeouts being so important and not because of the lack of ground balls.

    • spudchukar - Mar 23, 2013 at 9:33 AM

      Some valid points, except for the importance of strikeouts. Strikeouts aren’t important, outs are.

      • Kevin S. - Mar 23, 2013 at 10:06 AM

        Strikeouts are important because they’re outs far more often than any type of batted ball except pop-ups.

      • spudchukar - Mar 23, 2013 at 10:17 AM

        This is exactly what it troublesome about James article. Strikeouts are a by-product, not a goal, and if a pitcher can get an out with one pitch, that is superior to the 3,4,5 or more necessary to record a strikeout.

        There are only a handful of pitchers who get the best results via strikeouts, and they are the ones with the great power stuff. Unfortunately, the less talented strive to pitch like the elite, and suffer because of that desire. The best strategy is to make quality pitches, especially on strike one, and let the chips fall where they may. This really is the pitch to contact philosophy. Of course strikeouts carry a higher percentage of outs, Duh. But it shouldn’t be a goal, just a comfortable result of excellent pitches.

      • spudchukar - Mar 23, 2013 at 10:23 AM

        In the end, it can be summed up this way. Great pitchers are not great because they get a lot of strikeouts. Great pitchers are great because they make a lot of quality pitches, with the net result being they record a lot of strikeouts. But a pitcher can be great if he records few strikeouts. And a pitcher can also suck and still record a lot of strikeouts.

      • Kevin S. - Mar 23, 2013 at 10:31 AM

        I wouldn’t call them a by-product, I’d call them a process. And the process is what a pitcher controls, not the outcome. A strikeout is more certain than a batted ball. If a pitcher can rack up strikeouts without giving up too many walks or home runs, he should. I think you underestimate how many quality strikeout pitchers there are. Sure, pitchers need to stay within their abilities, and groundball/control is a formula for success, but it’s still rare to be a good pitcher without having at least some swing-and-miss stuff.

      • spudchukar - Mar 23, 2013 at 11:35 AM

        But the error in thinking comes from the swing-and-miss approach. If a pitcher strives to make pitches a hitter will swing and miss he is doomed. This is what good pitching coaches constantly preach. This is the definition of a pitcher not a thrower.

        What is essential is that on strike one, a pitcher must make a quality pitch, but not being the aim of a pitch that cannot be hit. Almost never is a pitcher good enough to throw three pitches to a hitter that he will swing and miss, with the obvious exceptions, and assuming the hitter is at least mediocre.

        Hitters who fall behind in the count have their respective averages drop 100 points. That is a huge deal, arguably the most important one of all statistics. Ideally, that first quality pitch will result in a hitter getting himself out, bad hitters do this consistently, good hitters do not. (ex.)
        Albert Pujols almost never swings at the first pitch recognizing the futility of doing so.

        If a pitcher gains the strike one advantage, then he can become more precise. And precision pitches do indeed result in more swing-and-misses but it is still not the goal, the goal is to make a pitch that is even less likely to be hit well, but either is a strike, or close enough for the hitter to believe it is, not wanting to fall behind 2 strikes creating a situation where he is even less likely to hit the ball solidly.

        And when a pitcher gets a two-strike advantage, he is blessed with the option of making pitches, that are even less likely to be hit, often resulting in a strikeout. Here even pitch to contact guys get the KOs.

        So while the argument may appear to be merely one of semantics, it isn’t really. It is a thought process, one that is hard to learn, but essential to succeed.

      • Mark - Mar 23, 2013 at 2:44 PM

        Pitchers with above average strikeout rates have the most success. While it would be ideal for a pitcher to get a one pitch out, you’re more likely to get an out by avoiding contact altogether.

        The pitch to contact philosophy sounds great in theory, but in reality what ends up happening is you get the Minnesota Twins pitching staff. They had the lowest K rate in the majors last season (5.90), and not surprisingly one of the worst ERAs in the league.

        The bottom line is that pitchers who get an above average number of K’s prevent runs better than those who pitch to contact.

        http://baseballanalysts.com/archives/2009/02/categorizing_pi_4.php

        Pitchers with above average K rates in 2008 had a 3.84 ERA while pitchers with below average K rates had a 4.73 ERA. The below average K rates would be your pitch to contact guys. Also from that study we can see the most successful pitcher has a high K rate, low walk rate, and high GB rate – the ERA of 3.32 was the lowest possible ERA of all the different combinations. Your pitch to contact types did significantly worse for instance, as those with below average K rates and above average walk/GB rates had a 4.28 ERA – nearly 1 run higher on average. Even the pitch to contact types that had good walk rates and were flyball pitchers had a 4.66 ERA – showing once again that below average strikeout rates and failing to keep the fall on the ground hurts a pitcher in the long run.

        If you think that was a fluke, then feel free to look at this site – http://baseballanalysts.com/archives/2010/03/categorizing_st.php.

        Here, they looked at pitchers from 07-09 and found that the most successful pitchers, once again, had high strike out rates, good control, and high GB%. The pitch to contact guys, or low K rate pitchers, were the ones who struggled.

        It’s easy to say a pitcher’s job is simply to get outs. But what’s important is how the pitcher gets their outs. And if they have above average strikeout rates, and avoid pitching to contact, they’ll be more successful in the long run.

        I’m sure you can give me examples of how with guys on pitching to contact is important because you can get a double play. I wouldn’t disagree there. But the majority of the time you want your pitchers to get their outs via the K, because that’s the best way for a pitcher to prevent runs from crossing the plate.

      • GoneYickitty - Mar 23, 2013 at 2:58 PM

        Different kinds of outs or more or less likely to move runners. Some types of outs force the defense to do something that they will sometimes fail to do. K’s at the pro level are basically worthless. You won’t move the runners and you’ll virtually never get on base due to a defensive mistake.

      • flavadave10 - Mar 24, 2013 at 7:56 AM

        Considering strikeouts are the only way a pitcher can take the defense out of the equation, I’d say they’re pretty darn important. Not a necessity, but dang they sure do help.

  10. Ben - Mar 23, 2013 at 1:10 AM

    This would be really, really easy to test, and I’m sure someone at Fangraphs, Baseball Prospectus or the Hardball Times will do it in the coming weeks.
    But just for shits and giggles, here are the top 25 groundball pitchers from 2010.

    1 Tim Hudson–TJ before this period
    2 Justin Masterson–Nope
    3 Derek Lowe–Nada
    4 Jake Westbrook–TJ before this period
    5 Trevor Cahill–Nothin’
    6 Jaime Garcia–Labrum and Partial Rotator Cuff Tear, previous TJ
    7 Roberto Hernandez–لا
    8 Ricky Romero–לֹא
    9 R.A. Dickey–Are you kidding? It’s R.A. fucking DIckey.
    10 Felix Hernandez–Nope
    11 Francisco Liriano–previous TJ
    12 Jon Lester–I doubt ground balls give you lymphoma
    13 Jon Garland–2011, Debridement Labrum & Bursa & Rotator Cuff
    14 Adam Wainwright–Yikes
    15 Carl Pavano–Nope
    16 Roy Halladay–Well, I guess we’ll see? But no
    17 Hiroki Kuroda–Nope
    18 Chris Carpenter–Thoracic Outlet Syndrome. Also, kicked in a brawl?
    19 Paul Maholm–mild shoulder strain in 2011
    20 Clay Buchholz–Stress Fracture in 2011
    21 CC Sabathia-nein
    22 Rick Porcello-nada
    23 Gavin Floyd-nothing of note
    24 Chad Billingsley–Partial UCL Ligament Tear and Flexor Tendon in 2012
    25 Edwin Jackson–is getting traded a lot an injury?

    You’d obviously have to compare that, and get a much bigger sample, but pitchers get injured. And there’s no common injury there other than TJ, but really, who doesn’t get TJ these days?

    • jeffbbf - Mar 23, 2013 at 3:42 PM

      is that a “nope” next to Carl Pavano’s name?

  11. onbucky96 - Mar 23, 2013 at 6:58 AM

    Lame. Bill James, step off the soapbox.

  12. Detroit Michael - Mar 23, 2013 at 8:22 AM

    Both this post and Rob Neyer’s summary fail to adequately describe James’ article very well.

    The reason why James comes to an unorthodox conclusion is because he defined “groundball pitcher” in an unorthodox way. Most of us think of a groundball pitcher as someone with a high percentage of groundballs divided by balls in play excluding bunts. James defines a groundball pitcher as someone with a high percentage of groundball outs divided by total outs.

    Well, duh. That’s going to lead to different conclusions. Take someone like a young Felix Hernandez. Under the common defintion of a groundball pitcher, Felix was an extreme groundballer but under James’ definition Felix recorded so many strikeouts that he would not be an extreme groundout generator. Strikeouts are so important to a pitcher’s success that someone who James considers to be an extreme groundball pitcher tends to have few strikeouts, which of course leads to a subpar career.

    James’ research in this article was not illuniminating. He still has unique insights often, but he tends to spend fairly little time reading others’ research so he can be very idiosyncratic.

    • churchoftheperpetuallyoutraged - Mar 23, 2013 at 8:43 AM

      his problem is he’s also basing information on anecdotes. I noticed the way he was describing the pitchers when he excluded guys like Brown due to the strikeouts. The problem is that we don’t have batted ball data going back that far, so we have no statistical idea about people in the 90s and further back.

      It’s a bad article based on zero data, especially when multiple examples disproving his “theory” are brought up.

  13. sabathiawouldbegoodattheeighthtoo - Mar 23, 2013 at 8:46 AM

    Well, find me the type of pitcher who doesn’t get hurt and I’ll give you a cookie.

    • Kevin S. - Mar 23, 2013 at 9:01 AM

      Knuckleballers, but that could just as easily be from microscopic sample size.

  14. Just a Fan - Mar 23, 2013 at 11:44 AM

    Sabermetrics is garbage….leave the game to your eyes.

    • 18thstreet - Mar 23, 2013 at 12:03 PM

      2,600ish MLB baseball games will take place this year. How many of them will you see with your own eyes?

      In 2007, Joey Votto hit 4 homers. How many of them did you see?

  15. marcinhouston - Mar 23, 2013 at 2:57 PM

    I know nothing about pitching mechanics but in a way this could make sense. A groundball hits the bottom half of the bat, a fly ball the top. So its about putting the ball lower or higher than expected. Pitchers naturally pitch at downward angle, releasing the ball during the downward forward rotation of their shoulder with a downward jerk of their elbow and flick of their wrist. If you are trying to get a ground ball by pitching deceptively low, you might put a little extra on one of these downward motions, but to get a fly ball you might hold off on part of the motion, and maybe get some back spin and higher trajectory in the process. If the ground ball is more of an art of extending akward motions and the fly ball is more of an art of moderating them, it is easy to see how the ground ball pitcher could be injured more frequently.

  16. Uncle Charlie - Mar 23, 2013 at 8:50 PM

    http://www.fangraphs.com/blogs/index.php/is-bill-james-right-about-ground-ball-pitchers-and-injuries/

Leave Comment

You must be logged in to leave a comment. Not a member? Register now!

Featured video

Red Sox shopping Lester and Lackey
Top 10 MLB Player Searches