Batter Traits That Cause Infield Fly Balls

The infield fly ball is the second worst outcome for a hitter besides a strikeout. With almost 100% of all popups turning into outs, a hitter, who is prone to skying the ball into the infield, will generate more outs and therefor a lower batting average. Several factors make a pitch more likely to be hit as an infield fly ball, but the key factor is the batter’s mechanics.

Background

A few initial facts on infield flies balls (IFFB) can help with understanding them.

1. Infield flies (used interchangeably with pop ups in this article) have a direct effect on a player’s BABIP. Here is a plot of the BABIP and infield fly ball percentage (IFFB%) for all players with more than 300 PA from 2002 to 2012.

Using the equation generated from the best fit line, a player’s BABIP will drop around 3 points for every 1% point increase in IFFB%. A hitter with a 5% IFFB% will normally have a BABIP around .316 while a hitter with a 15% IFFB% will have a BABIP around 0.286.

2. IFFB% stabilizes fairly quickly. Pizzacutter found that it took 500 PAs for pop ups to stabilize. Derek Carty found that it took about half that number of plate appearances. No matter the exact number, a player IFFB% will stabilize within a single season.

3. Besides stabilizing quickly, the IFFB trait correlates from year-to-year. Here is the plot of all players with over 300 PA in 2011 and 2012 and their IFFB%.

Batters who pop up tend to do so the next season.

Research

Note: In the process of writing this article, I ran many queries on IFFB% and found some data irrelevant to this article. I have included the data in an Appendix at the end of the article for those who can’t get enough numbers.

I sorted out some of the conditions that lead to a high number of IFFBs for a hitter. The first bit of useful information is pretty obvious. A hitter is more likely to induce a fly ball on a pitch that doesn’t break downward.

Pitch: IFFB%
Cutter: 9.3%
4-finger fastball: 8.4%
Knuckle: 8.1%
Slider: 7.6%
Change up: 5.9%
Split Finger: 5.3%
2 Finger fastball: 5.0%
Curve: 4.6%
Sinker: 4.0%

The reason the cutter induces so many infield flies is because it moves inside on a hitter and makes contact with the narrow part of the bat. If the ball happens to be a bit on the high side of the thin part of the barrel, a pop up occurs. Or a weak grounder happens if the ball hits on the bottom half. A batter wants to keep the ball on the barrel of the bat.

Besides the pitch type, the pitch’s location is important to inducing pop ups. The key location to induce a pop up is high and inside. Here are two tables representing to the IFFB% for various pitches in the strike zone. The area covers the rule book strike zone and is from the catcher’s perspective.

Right handed Hitters

-1 to -0.5 -0.5 to 0 0 to 0.5 0.5 to 1
3 to 3.5 21% 15% 11% 8% > than 10%
2.5 to 3 14% 10% 7% 5% 6% to 10%
2 to 2.5 8% 6% 5% 4% 5% or less
1.5 to 2 5% 4% 4% 4%

Left Handed Hitters

-1 to -0.5 -0.5 to 0 0 to 0.5 0.5 to 1
3 to 3.5 6% 11% 18% 24% > than 10%
2.5 to 3 4% 7% 12% 17% 6% to 10%
2 to 2.5 3% 4% 7% 11% 5% or less
1.5 to 2 2% 3% 4% 5%

High and inside is by far the best location to induce a fly ball. Just 12% of the all the pitches are in the upper inside zone and the two zones adjacent to it (Note: I will refer to these 3 zones as the IFFB Zone from now on), but these pitches induce 28% of all the pop ups. The problem for hitters is that they just can’t lay off and take a pitch in the IFFB Zone for a ball. While pitches in this area are near the edge of the umpire called strike zone, 73% of all taken pitches in this area are called strikes.

With that background, I looked at what makes a hitter more prone to IFFBs. I thought the answer would be easy. Possibly players who swung at more pitches in the IFFB Zone hit more IFFBs. After running way too many tests, I actually found that hitters that swung at pitches in the IFFB Zone where less likely to hit an IFFB. It seems that hitters who can handle the high inside fastball, seem to swing at it. Those hitters who can’t handle the pitch lay off of it.

Example

At this point, I decided to compare 2 somewhat similar power hitters. One with a high IFFB% and another with a low value. For this comparison, I looked at Buster Posey and Edwin Encarnacion.

Name Team IFFB% BABIP BB% K% HR/FB LD% GB% FB% AVG OBP SLG
Buster Posey Giants 4% 0.368 11% 16% 19% 25% 47% 29% 0.336 0.408 0.549
Edwin Encarnacion Blue Jays 12% 0.266 13% 15% 19% 18% 33% 50% 0.280 0.384 0.557

Truthfully, I don’t like this comparison, but I will live with it. I wanted two hitters on the opposite sides of the IFFB spectrum. I got that. I got similar HR/FB rates. Both are right handed. Both have similar K% and BB%. At that point, the similarities end. To get players at the extreme ends of IFFB%, some compromises where made.

As seen in the above stats, Posey only had a 4% IFFB% compared to Encarnacion’s 12% rate in 2012. Besides their overall stats, here are the results of the balls that each hitter made contact with in the IFFB Zone during this past season:

Batted Ball Type: Number (Percentage)

Encarnacion
Grounder: 14 (29%)
Liner: 6 (12%)
Fly Ball: 11 (22%)
Pop Up: 18(37%)

Posey
Grounder: 16(32%)
Liner: 16 (32%)
Fly Ball: 18 (36%)
Pop Up: 0 (0%)

Posey did not hit one high inside pitch for a pop up. On the other hand, Encarnacion popped up the ball over 37% of the time. The difference in how the pair handles inside pitches can be seen in this run value heat map (Posey’s values compared to Encarnacion). The heat map is from the catcher’s perspective.

The contrast is easy to spot with Posey getting better results on the inside part of the plate compared to Encarnacion.

To find the difference in how the pair handles inside pitches can be achieved by looking at just few swings each batter made at pitches in the IFFB Zone. Here is the pair swinging at a pitch high and inside that resulted in a fly out to the outfield.

Besides the two videos, here are 6 images of each hitter when they swung at a pitch in the IFFB Zone on a 0-0 counts. I only wanted examples when the hitters were not trying to protect the plate, so I limited the data to 0-0 counts. (embiggen)

Several traits can be seen that allow Posey to pop up less often.

1. First, Posey does not drop his back elbow as much as Encarnacion and therefor keeps his swing more level. Encarnacion has a natural upper cut to his normal swing, but it gets exaggerated when dealing with pitches high and inside. Since Posey doesn’t drop his back shoulder as much, he is less likely to pop the ball up.

2. Most of Posey’s power comes from his hips (which are facing the pitcher when the ball and bat make contact) and his wrists, that he keeps in. By keeping his wrist close to his body, Posey is able to make contact with the solid part of the bat. Encarnacion keeps his bat away from his body. Since he has the bat further away, he makes less solid contact. A couple of the images show the ball hitting Encarnacion’s bat on the handle.

3. Finally, Posey repeats his swing exactly the same each time. His feet, hips, head and wrists are all in the same location in each image. Encarnacion is all over the place with his mechanics. For example, his feet are in a different position in each image. Posey has found a swing that works for him on inside pitches and he repeats in constantly. Encarnacion just can’t seem to get comfortable on the inside swing and it shows with his results.

Not all hitters who hit a high number of IFFB will have the same traits that Encarnacion exhibit or vise versa for hitter who don’t hit a high number of IFFB like Posey. All hitters are different, but the preceding comparison is an example of why two hitters get different results on high and inside pitches.

The infield fly ball is the worst batted ball result for a hitter since it almost always results in an out. Infield fly balls are usually induced by high-and-inside pitches that don’t have any downward breaking action. The only way to determine why a hitter has a propensity for hitting weak pop ups is to break down their mechanics.

H/t to Mike Newman for helping with the swing evaluations.
.
.
.
.
Appendix
Count

Count IFFB%
0–1 7.0%
2–1 7.1%
1–1 7.2%
1–2 7.3%
2–2 7.4%
0–2 7.5%
3–2 7.5%
3–1 7.6%
1–0 7.7%
0–0 7.8%
2–0 8.2%
3–0 9.1%

The big surprise to me was the the higher number of pop ups on hitter counts. With hitters probably looking for a specific pitch at a specific part of the plate, the hitter bails out the pitcher by popping up the ball.

Inning

Inning IFFB%
1 7.3%
2 7.5%
3 6.9%
4 7.3%
5 7.3%
6 7.4%
7 7.5%
8 7.6%
9 8.2%
10 8.3%

When looking at pop ups per innings, there is a drop from the beginning of the game through the next 3 innings. Then the rate slowly increases back up. Not much of a conclusion can be drawn of hitters from the data because the IFFB% increases the most when teams are bringing in their relief pitchers.




Print This Post



Jeff writes for FanGraphs, The Hardball Times and Royals Review, as well as his own website, Baseball Heat Maps with his brother Darrell. In tandem with Bill Petti, he won the 2013 SABR Analytics Research Award for Contemporary Analysis. Follow him on Twitter @jeffwzimmerman.

35 Responses to “Batter Traits That Cause Infield Fly Balls”

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

    “The infield fly ball is the second worst outcome for a hitter besides a strikeout.”

    GIDP says hello!

    And GITP pushes them both aside.

    Though this gets to me wondering how (negatively) correlated IFFB% and GIDP% are, and whether that would mean that a high IFFB% isn’t quite as big of a negative as it seems to be.

    Vote -1 Vote +1

    • Eminor3rd says:

      That’s apples to oranges. IFFB and GIDP are on different levels. You have to compare IFFB to ‘ground ball,’ and ‘IFFB pop out’ to GIDP.

      IFFB/GB refer to outcomes of contact, but Pop-out/GIDP refers to outcomes of the play.

      Vote -1 Vote +1

      • ralph says:

        Of course. It’s just that the way he wrote the intro took me out of the article for a second while I re-interpreted it to mean what Jeff meant.

        That isn’t an ideal characteristic for an article’s intro. Maybe I’m the only one to have that reaction, but I suspect not.

        Basically, I think the article would have been improved by making the statement intro more precise, maybe something like “A pop-up guarantees an out almost as surely as a strikeout guarantees an out. Therefore, a hitter…”

        Vote -1 Vote +1

    • jim says:

      batters don’t have GIDP rates, they have GB rates. GIDP and GITP depend very heavily on having runners on in front of you, which is not something an individual hitter controls

      Vote -1 Vote +1

      • ralph says:

        I was just cutting to the heart of the matter. I’d expect more IFFBs mean fewer GBs which should mean fewer GIDP if we were able to run a perfectly controlled-experiment.

        Though your comment does make me wonder whether there might be an observable effect as to whether certain players might ground into more or less GIDPs than would be expected by their GB%.

        With terrribly-small sample sizes, I’m not sure we could ever really be certain of anything on this front, but I suspect there are hitters who can reduce GB% in GIDP situations. Though it’s an open question whether they should attempt such a thing if they’re capable of it.

        Vote -1 Vote +1

    • Gareth says:

      GIDP isn’t always available though. The IFFB is the second worst outcome that always has a chance of happening.

      Vote -1 Vote +1

      • ralph says:

        Yep, see above. Maybe another way of saying it is that with no men on base, the IFFB is the second-worst outcome for a hitter.

        Vote -1 Vote +1

      • Kris says:

        I think IFFB is probably worse than a strikeout. Runners have zero chance of advancing, the batter (if runners are on base) has zero chance of reaching, and truly miniscule chance with bases empty. His chances of reaching are probably better by strikeout, and the chance of runners advancing is also better by strikeout.

        Vote -1 Vote +1

  2. ralph says:

    “A hitter with a 5% IFFB% will normally have a BABIP around .316 while a hitter with a 15% IFFB% will have a BABIP around 0.286.”

    This is really useful info, by the way. It seems like when trying to estimate AVG potential, K% and IFFB% could be considered together as something like Hopeless Outcome% (HO% – the acronym could use some work, perhaps). IFFB% seems like it’s at least important enough to be up in the dashboard view alongside BB% and K% on the default FanGraphs player page.

    Vote -1 Vote +1

  3. Rob says:

    Umpire Traits that Cause Infield Fly Balls:

    1. Being Sam Holbrook

    Vote -1 Vote +1

  4. DJ says:

    I think it’s the back shoulder that is the true measure of the swing plane (which I guess, in turn, drops the elbow, but the shoulder is the key piece).

    RE: The appendices. Both of these findings actually make sense to me.

    If you ask most hitters (especially RHH), they will tell you that when they are “zoned up,” such as in 2-0 and 3-0 counts, they are probably looking for a fastball on the upper/inside quadrant of the zone. This is, for obvious reasons, the easiest pitch to turn on and pull for a HR. There’s also a fine line between really crushing one of these pitches and jamming yourself enough that you pop it up. In these cases, the ball probably runs in more than you expected and you get it off the label/handle of the bat. Or, in some 2-0 cases, you’re early on a change up and mis-hit the ball. (I’m willing to bet that there are more IFFBs from same-handed matchups i.e. RHP/RHH and LHP/LHH.) Many LHH who aren’t looking up and in are looking down and in, so if they get under it too much and/or way out in front, they’ll pop that pitch up as well.

    As for why the rates are what they are regarding the innings, my guess is that it has to do more with the pitchers and pitch selection during these times. Most pitchers generally work off FBs the first time through the order (2+ innings) and their velocities are typically higher early in the game (unless your name is Justin Verlander). And set-up men/closers are more likely to go for Ks with high and/or inside FBs than anybody else. (Easier to do with you’re gassing up 97-100 for one inning, and plenty of late inning guys don’t have much other than a FB.) Maybe the dip in inning three has to do with NL pitchers batting there? Seems like they would be on the low end of the IFFB% spectrum, and the prevalence of pinch hitters later would cancel that out some.

    Vote -1 Vote +1

  5. Ben says:

    “Just 12% of the all the pitches are in the upper inside zone and the two zones adjacent to it ”

    I don’t think you want to say “just” there, do you? Not trying to be a dick about simple grammar mistakes like people often are here, but I’m genuinely not sure I understand what you’re saying here

    Vote -1 Vote +1

    • Paul Clarke says:

      The “just” is fine (though there’s an extra “the” in “the all the pitches”). He’s saying that only 12% of all pitches are in that zone but they still manage to generate 28% of all popups – i.e. pitches in that zone are much better than average at generating popups.

      Vote -1 Vote +1

  6. Ben Hall says:

    What are the r squareds for the two lines of best fit?

    Vote -1 Vote +1

  7. Well-Beered Englishman says:

    This looks fascinating, and I’m looking forward to diving in, but I scrolled down quickly to ask: who got through 300 plate appearances with a BABIP below .200?

    Vote -1 Vote +1

  8. Jeff T says:

    Couldn’t this article have been truncated to just say:”Being Jimmy Rollins?”

    Vote -1 Vote +1

  9. Nathaniel Dawson says:

    The only way to determine why a hitter has a propensity for hitting weak pop ups is to break down their mechanics.

    You somehow got to this conclusion how? This sounds absurd, really.

    Vote -1 Vote +1

  10. Ivan Grushenko says:

    So high and inside pitches are most likely to be popped up and batters who have the most trouble with inside pitches are most likely to hit pop ups. Sort of makes sense, but why do batters who have trouble with high pitches not also have lots of pop ups? Or do they?

    Vote -1 Vote +1

  11. Paul Clarke says:

    You seem to be using IFFB% to mean at least a couple of different things here, and also maybe using different sources:

    1) The year-to-year correlation and IFFB-vs-BABIP graphs look like they’re using IFFB% as a percentage of all fly balls, just like Fangraphs’ batted-ball data.

    2) Derek Carty reports on IFFB as a fraction of contacted balls, using Retrosheet and MLBAM data. I can’t check Pizzacutter’s study, but I think he used IFFB as a fraction of PAs.

    3) The per-pitch and per-location data looks also looks like it’s using IFFB as a fraction of contacted balls, but I can’t tell what the data source is. Pitch F/X (hence MLBAM)? BIS?

    Other queries:

    A hitter is more likely to induce a fly ball on a pitch that doesn’t break downward.

    Seems to be generally true, but two of the bottom three pitch types for generating fly balls are sinkers and two-seamers, which have “rise”. Presumably this is because they’re usually located in the bottom of the zone? This makes me wonder how much of the pitch-type effects you can explain just by where the different types of pitches are usually located.

    The reason the cutter induces so many infield flies is because it moves inside on a hitter and makes contact with the narrow part of the bat.

    To a same-handed hitter, cutters move in less than other fastballs. Is there a big platoon split in the infield flies generated by cutters? If not you might need a different explanation (location again?).

    Vote -1 Vote +1

    • payroll says:

      To your last question, a pitcher who is same-handed pitcher will generally stay away from cutters. The vast majority of c utters that a hitter sees will be from opposite-handed pitchers. There will be a platoon split, but I don’t think it will carry much if any statistical significance.

      Vote -1 Vote +1

      • Paul Clarke says:

        A quick check of a few pitchers (Halladay, Hamels, Peavy, Vargas, Haren) using Brooks Baseball’s classifications didn’t show them avoiding cutters to same-handed hitters – Vargas and Haren, for example, threw them more often to same-handed hitters this year. Do you have league-wide data?

        Vote -1 Vote +1

  12. Brad Johnson says:

    Jeff and others,

    This is a minor nitpick, but anytime you use the concept of “stabilizing” can you please define explicitly what it means. This concept is way too frequently misused, misunderstood, or taken out of context.

    Vote -1 Vote +1

  13. DD says:

    My first thought was that players who wrap their fingers/palm around the bat knob will pop up more (not sure if E-5 does this) since it is harder to keep the barrel on an inside pitch.

    Vote -1 Vote +1

  14. ArodinmyPujols says:

    Being Vernon Wells

    Vote -1 Vote +1

  15. GMH says:

    What are four-finger and two-finger fastballs?

    Vote -1 Vote +1

  16. Dreamin says:

    Being Chris Young

    Vote -1 Vote +1

  17. Steve S says:

    As Paul Clarke said earlier, IFFB% on Fangraphs is the percentage of fly balls hit that don’t make it past the infield. That makes the stat not especially meaningful without the context of the hitter’s FB%. That’s why in my article on explaining BABIP for pitchers, I use FB%*IFFB% as the factor (rather than IFFB%), as that’s how you represent the total percentage of batted balls that are infield popups.

    With that in mind, the popup frequency comparison between Posey and Encarnacion isn’t 4% vs. 12% — it’s about 1.2% vs. 6.0%.

    I really like the location breakdowns, though.

    Vote -1 Vote +1

  18. Sonny says:

    Good stuff. Thanks

    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>

Current day month ye@r *