Remember me

Who is benefiting from pulling fly balls more often?

May 25, 2018
 We’re in the midst of a fly ball revolution, with more major league hitters taking the approach that it’s best to swing for the fences than play small ball. The past success of players like Daniel Murphy in changing their swing and approach has begotten many followers.

One way to look at these changes is to check out who has increased the frequency with which they pull their fly balls, as nine of the top 11 players who have increased their fly ball pull frequency have increased their slugging percentage on flies. Who are some of those players? 

Miguel Rojas, Marlins SS

Miguel Rojas might be the most unlikely player to launch his way up the home run leaderboard. Rojas hit one home run in each of his previous four seasons and was averaging one home run every 189 at-bats for his career entering 2018. He has seven home runs in 2018 and has joked that he wants to enter the Home Run Derby.

This season, he’s upped the pull rate on his fly balls by 20 percentage points – the fifth-highest rate in the majors. In other words, he’s pulled 16 of 46 fly balls, when by 2017’s pull rate he would have been expected to pull seven. Rojas has had similar pull rates before, but he seems to be consciously trying to drive the ball. He has a 33 percent hard-contact rate on his fly balls. He’s never been above 25 percent on them previously.

The power has made Rojas more of a complete player. He already rates as a solid defender and is among the shortstop leaders in Defensive Runs Saved.

Jed Lowrie, Athletics 2B

Veteran second baseman Jed Lowrie is swinging like someone who wants to crack the 20-homer mark for the first time in his career. Lowrie, a switch-hitter, is doing major damage from the left side.

On that side, he’s hit eight of his nine home runs and has increased the rate at which he pulls fly balls by 16 percentage points (he’s pulled 17 of 42 from the left side). His slugging percentage on fly balls has gone up by 456 points from last season. Lowrie still has the challenge of playing 81 games in Oakland where even heavy pull tendencies may not help him. He’s hit seven of his nine home runs on the road.

Xander Bogaerts, Red Sox SS

Xander Bogaerts had a drop in power production from 2016 to 2017, going from 21 home runs to 10, but has regained some of what he lost early in 2018. He’s hit six home runs already. Early on, Bogaerts' pull percentage on fly balls is 11 points higher than it was in 2017.

And though that could be due to chance (his line drive pull rate is up too), it would behoove Bogaerts to continue to try to take aim at the Green Monster. He has as many home runs over it this season as he did in 2017 (4).

Spencer Harrison also contributed to this article


COMMENTS (9 Comments, most recent shown first)

KL: Now you're doing me too much credit!!
I take shortcuts all the time! But indeed that's not a kind that I would take -- unless I didn't see that it's based on 'impure' data, when purer data are right there too; unless I just didn't see what it was based on, which sometimes happens.
In short: There was every reason to wonder what I did.

You did a great job fleshing this out further.

It is unfortunate that Mark Simon and apparently others from his organization don't reply to comments here. But hopefully they take heed from them.

When they make mistakes like the one at the top up there, it seems they just don't know the history, and didn't think to take a quick look to confirm it, which wouldn't have been hard. Even taking shortcuts. :-)
11:37 AM Jun 1st
I figured you had done that, from the way your numbers followed the same trend as mine. I was careless, though, because reading my post again, I can see how what I meant to be general warning about the potential confusion from BBR's inconsistency would have easily sounded like I was specifically lecturing you, considering that you had just finished using BBR data. In my defense, I've seen enough of your work to know you don't take shortcuts on precision, so it really didn't occur to me that the confusion would have tripped you up.

So it never crossed my mind to rule out what didn't occur to me as a possibility, just as it never before has crossed my mind to say everyone going outdoors should beware of flying circular saws — except that now I see the difference is that the absence of flying circular saws is much more universally accepted than those things so obvious to me that I haven't expressed them.

Looking back, I suppose I was asking too much of visitors to this page to perceive my confidence in your process by reading my mind. :-) ;-)
6:11 AM Jun 1st
Let me clarify, in case you might have thought otherwise, which it seems you might:

I didn't just use whatever ratios the site might show for flyballs vs. grounders, and I didn't use those sets of data that mish-moshed flyballs and line drives.

I did ratios on my own from the flyball and ground ball data, taken from the sets of data that do list fly balls separately from line drives.
The ratios that I used did reflect's "pure" flyball data.
10:20 AM May 31st
I'm sorry, Maris — I should have been more clear. That sentence almost confused me, and I know what I meant. :—)

To be clear, the issue didn't damage your findings. Our results were pretty much identical. I was just trying to explain my preference for FG in [i]way too few words[/i]. So now I'll do it in way too many. . . .

On the Splits page, the Hit Trajectory table obviously separates FB, LD and GB (plus bunts) into distinct categories.

But on the Advanced Batting page, the "Batting Ratio" tables include line drives in every reference to fly balls.

This is the entire glossary entry for:
"GB/FB -- Ground Ball to Fly Ball Ratio
[/i]"Includes line drives as fly balls.
"Batted ball type and location is not complete prior to 1988."[/i]

Same thing with IF/FB (which is ridiculous) and HR/FB.

I might have noticed this first because Votto had surprisingly high IF/FB percentages at BBRef. Every Reds fan knows Votto almost never hits infield flies. FG has his career percentage as 1.2%, including 0 in 2010 and 2016, and five other seasons with just 1. At BBRef, it's 3% (without a decimal, which I don't like), because Votto [i]does[/i] hit line drives that infielders catch.

Another reason I prefer Fangraphs for FB/LD/GB/bunt data is that it gives raw numbers in addition to percentages. But BBRef has two other problems — one perhaps trivial and one definitely not.

On the Splits page, this is how BBRef explains it's table for "Vs. Fly Ball/Ground Ball Pitchers":
[i]Fly Ball pitchers are in the top third of the league in ratio of fly ball outs to ground ball outs. Ground Ball are in the bottom third of the league in the ratio of fly ball outs to ground ball outs. . . .[/i]
Why would they say it that way, flies-to-grounders? The issue is a pitcher's propensity to induce ground balls. Extreme fly ball pitchers are just pitchers who don't induce many grounders. No pitcher works on inducing more flies. Even BBRef's own Ratios stat GO/AO (ground outs / air outs [including line drives]), is expressed as grounders-to-flies. The explanation also doesn't specify whether "fly ball outs" include line-outs. It ought to and probably does, but why not say so? . . . Maybe I'm being unfair, but this kind of inconsistency and imprecise language is a big red flag to me.

And this is a bigger problem. The sum of FBs+LDs+GBs+bunts should equal ABs+SFs+SHs-Ks, right. On the BBRef Splits page, in the Hit Trajectory table, it often doesn't.

To refresh my memory, I checked out nine Pujols seasons, early, middle and last year. Because this seemed to be one of the few active threads without Pujols' name.

In two of those 9 seasons, Pujols' FBs+LDs+GBs+bunts did not equal ABs+SFs+SHs-Ks (he has bunted 6 times, with 1 SH, so I included all 4 of those seasons). They were close, but there's a place to indicate incomplete data, and it wasn't indicated. Besides, the reason I looked this up is that I've noticed the problem before. I haven't noticed it at FG.

For older data, BBRef goes back twice as far as FG, so there's no choice for those years. It's not as if BBRef's numbers are horribly inaccurate. They're not quite precise in some cases and confusingly inconsistent in others. But large enough samples should smooth out the imprecision, and most important, understanding the inconsistencies should help avoid those pitfalls.
2:12 AM May 31st
(......but, terrific breakdown and analysis.
Just wondering what you meant about the merging-line-drives thing.)
11:23 PM May 30th
KL: What do you mean about baseball-ref "merging line drives into fly balls"?
Sounds like you mean either that they don't give separate data for line drives (which would be false; they do), or that there has been some shift over the course of time in how they handle the data, about which I don't know, but which is a kind of thing I did allow for but said I doubted there hae been changes in how they've tabulated and expressed the data on hit trajectory since they started doing it 30 years ago.

What did you mean?
(Gonna chase you down on Reader Posts about this too.) :-)
11:18 PM May 30th
Baseball Reference is a little misleading, because it merges line drives into fly balls. I prefer Fangraphs' flyball and groundball data for that reason, so I wandered over there. Going back to 2001:

---In 2006, the fly ball percentage spiked from 34.9% the previous year, up to 36.7%. (The decrease in line drives was much greater than the decrease in grounders.) The FB% stayed in that vicinity through 2011, peaking at 37.8% in 2009 (with the LD% down from its 22.5 peak in 2003 to 18.2 in 2010).

So: as drug testing began, fly balls rose by 3% and line drives went down by a bit more than 4%, circa 2003-2011.

--In 2012, the FB percentage plunged from 36.0 to 34.0. That rate stayed pretty steady through 2015, when it bottomed out at 33.8%. In that same four-year period, ground balls went up by 0.9%, making 2015 the century's highest year for GB/FB ratio: 1.34. The line drive rate had started going up a year earlier, in 2011, and rose by 2.7% in five years, to 20.9%, the highest since 2003.

Recapping so far: After rising and then falling, the FB% had a net decrease of 1.1% from 2005 to 2015, LDs fell and rose to a net change of 0.0% in 10 years. And ground balls decreased, then increased, with a 10-year change of +1.1%.

--Fly balls have gone up each year since 2015, never by as much as a full percent, but with a three-year gain of 1.8 percent to 35.6% so far this year, still well under the spike of 2006-11. And the increase is a tick under the ONE-year increase of 1.9% in 2009 (to its recent peak).
--The LD%, steady since 2012, has risen by only 0.3% since 2015.
--And ground balls, down 2.1% in a steady fall since 2015, have reached their lowest level since before 2002, which apparently is FG's first year of this data.

Also: wasn't 2015 also the first year of StatCast?

Maris is right about the more important, broader picture, too. We're not in a fly ball revolution. We're in a LAUNCH ANGLE REVOLUTION, and the angle being aspired to is kind of in between a line drive and a fly ball. This revolution has been driven by two things: Statcast's confirmation of the optimum revolution (which has given a revival to the Ted Williams axiom of meeting the pitch's downward plane with an uppercut swing).

And the routine infield shifting, which makes ground-ball hits increasingly harder to come by. That's the thinking, anyway. The data says the MLB batting average on ground balls has either increased or stayed the same every year since 2003, rising from that year's low of ,215 to last year's .241 (while the number of ground balls has fallen by more than 3,00). Go figure. (There were also two ground ball home runs in 2012.)

This article is fine for a general readership, but for BJOL subscribers, it is likely to draw the observation that "pulling fly balls" is actually two different things. Fly balls tend to go toward the opposite field more than the pull field, so pulling them is another matter entirely, and is probably as old as the live ball. However, the philosophy comes and goes, and two players' successes revived interest in pulling the ball to increase HR power. They did it several years before the fixation on launch angle. Jose Bautista began hitting home runs at a much higher rate toward the end of the 2009 season, and continued until he had hit 54 in 2010, which was more than his combined total for 2004-08. The next year, 2011, Curtis Granderson began consecutive seasons of 41 and 43 home runs, after hitting just 24 in 2010 and peaking at 30 in 2009.

The pull rate is up 1.5 percent since 2015, to 40.6%, but remains at least 2% below the pre-2006 plateau, which ended one year after routine PED testing, if I'm remembering the date correctly. Pulling the ball has not captured as much interest as optimum launch angle. But the article is right that combining the two is a prescription for home runs.

10:10 PM May 30th
........I looked into it, and unless the way that the ground ball and fly ball data on is done has been changed at various points and therefore is misleading, which I'd guess is doubtful.....

The opening phrase of the article is mistaken.
There isn't a fly ball revolution.

What there is, in terms of fly balls vs. ground balls, is a partial reversal of a downward trend in fly balls that occurred from 2012 to 2015, and which still isn't up to the prior levels.

I looked at the ratio of fly balls to ground balls every year from this year down to 2010, and then down to 1989 with some years skipped (for convenience).
1989 is the first year for which has the data for a full year. (It has partial data for 1988, and none for prior.)

Flyball-to-groundball ratio in 1989 was .77.
In 1990 it was .78.
It stayed in the .70's through 1998, range .70-.79 (the .79 was in '92).

Then it went into the .80's, mostly in the low .80's, and stayed there (basically) through 2011 (I didn't look at every year in there. I did look at all the years from '89 to '99, and '10 to '18.)

Except for the years '94-'97, when it was only .70-.72 every year, it was always at least .75 during the whole period from 1989 through the year 2012, when it was .78.

2013: .69
2014: .64
2015: .58
2016: .63
2017: .67
2018 so far: .71

I'll save my data for a while in case anyone is interested in more details.
12:12 AM May 26th
Side question:
Assuming there really are more fly balls now (is that for sure??), would that mean that fielding for outfielders has become relatively more important than before?

Seems it would have to.
At least relatively, compared to fielding for infielders; it wouldn't necessarily mean an absolute increase in importance, because of the general decrease of balls in play.
10:59 PM May 25th
©2024 Be Jolly, Inc. All Rights Reserved.|Powered by Sports Info Solutions|Terms & Conditions|Privacy Policy