The Change: The Pop-Up, And When To Start Worrying

It’s tricky to write about batted ball mix changes, for the most part. If you’re talking push and pull, adding the ability to go the other way can increase your batting average, sure. But it can also decrease your power output. Ground balls and fly balls act the same way — there’s really an ideal mix for each hitter, and we’re trying to figure out just as much as they are which is the best way forward.

There’s one batted ball type that just plain sucks, though. The pop-up. The infield fly.

98.5% of the time, that’s an out. It’s a bad idea, plain and simple.

Admittedly, there’s a little give to this idea. In that, pop-ups are part of your batted ball mix, and if you’re changing your swing plane, pop-up percentage will change along with it. So ground ball rate and fly ball rate do stabilize before pop-ups, probably for this reason: a change in your plane could result in a few more ground balls before it results in fewer pop-ups.

And there are fewer pop-ups in general, so you just want a bigger sample to say too much about them, anyway. We aren’t quite there, either. Derek Carty found that 194 balls in play is where you begin to get more signal than noise out of the PU% rate, and that’s about two weeks away. But we’re getting close, and it makes sense to identify the at-risk players before then, anyway.

The easy take away is that pop-ups are bad, mmkay. And we’re talking about pop-ups divided by balls in play, not by fly balls, because that’s a superior stat. It has better year-to-year correlation, and it really just makes more sense. Why would we divide pop-ups by fly balls anyway?

First, let’s look at the worst offenders right now. You’ll see they have bad BABIPS, for good reason. Remember that league average PU% is 3.5%. So, Todd Frazier, wow.

Name	BABIP	GB/FB	LD%	GB%	FB%	PU%	HR/FB	Hard%
Todd Frazier	0.225	0.73	22.8%	32.5%	44.7%	9.7%	23.5%	38.6%
Colby Rasmus	0.322	0.50	22.7%	25.8%	51.5%	9.1%	20.6%	40.9%
Edwin Encarnacion	0.238	1.22	17.3%	45.5%	37.3%	8.2%	22.0%	25.5%
Mike Moustakas	0.333	1.13	20.0%	42.4%	37.6%	8.0%	8.5%	24.0%
Mookie Betts	0.226	0.77	16.3%	36.4%	47.3%	7.8%	8.2%	30.0%
Hanley Ramirez	0.245	1.58	21.3%	48.1%	30.6%	7.4%	30.3%	34.3%
Starlin Castro	0.319	2.55	15.6%	60.7%	23.8%	7.4%	10.3%	25.4%
Kolten Wong	0.336	1.34	24.8%	43.1%	32.1%	7.4%	14.3%	31.3%
Jose Bautista	0.234	0.56	15.7%	30.1%	54.2%	7.2%	13.3%	34.9%
Jose Altuve	0.331	1.26	19.3%	45.0%	35.7%	7.1%	10.0%	24.7%
Carlos Santana	0.240	1.19	18.2%	44.4%	37.4%	7.1%	10.8%	28.0%
Brian Dozier	0.286	0.57	31.6%	24.8%	43.6%	6.8%	11.8%	30.5%
Xander Bogaerts	0.301	1.15	17.3%	44.2%	38.5%	6.7%	2.5%	24.5%
Asdrubal Cabrera	0.279	0.76	15.4%	36.5%	48.1%	6.7%	2.0%	23.8%
Lonnie Chisenhall	0.238	1.02	20.6%	40.2%	39.3%	6.6%	7.1%	24.6%
Alexei Ramirez	0.269	1.61	21.1%	48.6%	30.3%	6.4%	6.1%	29.1%
Juan Lagares	0.330	1.51	16.2%	50.5%	33.3%	6.3%	2.7%	19.5%
Trevor Plouffe	0.295	1.11	19.2%	42.4%	38.4%	6.1%	13.2%	24.0%
Marcus Semien	0.375	0.98	21.2%	39.0%	39.8%	5.9%	12.8%	31.1%
Luis Valbuena	0.194	0.56	20.6%	28.4%	51.0%	5.9%	19.2%	33.0%
Average	0.281	1.11	19.9%	40.4%	39.7%	7.2%	12.5%	28.9%

One thing you might notice is that it’s really impossible to typecast these players. We’re not talking about a bunch of fly ball hitting sluggers, and we’re not talking about the secret problem with being a slap-hitter. This group hits the ball about as hard as the league average (28.6%), and has hit slightly more home runs per fly ball (10.8% HR/FB is the league average). We’ve got Juan Lagares and Starlin Castro here, and we’ve got Colby Rasmus and Jose Bautista here. All types.

For the most part, these guys have already seen their BABIPs influenced by their pop-up problems. A .281 BABIP is not a good one. And you can’t just say that the guys on this list with good BABIPs have been hitting the ball hard when they line it up: Todd Frazier has the second-best Hard%, and Juan Lagares has the worst Hard% on this list of 20.

I will say that enjoying a .330+ BABIP while hitting the ball softly and also popping up the ball has *got* to scream regression, especially when it doesn’t come with great speed. So, sorry, Mike Moustakas, I think your batting averages are headed south. You might even say the same for Starlin Castro and Xander Bogaerts, despite their modest BABIPs: there’s little reason to believe they deserve average BABIPs with their batted ball mix.

Even a Juan Lagares type — who might be benefitting from soft contact with an infield hit percentage that’s about 50% better than league average — shouldn’t be hitting pop-ups. Look at Dee Gordon. Hits the ball softly but has a 1.5% PU%. That’s how you speedster.

The other side of the coin.

Name	BABIP	GB/FB	LD%	GB%	FB%	PU%	HR/FB	Hard%
Chris Davis	0.313	1.00	22.2%	38.9%	38.9%	0.0%	28.6%	41.7%
George Springer	0.222	1.61	23.1%	47.4%	29.5%	0.0%	26.1%	34.6%
Howie Kendrick	0.351	2.90	28.7%	53.0%	18.3%	0.0%	19.0%	27.8%
Ryan Howard	0.284	0.73	27.3%	30.7%	42.0%	0.0%	18.9%	37.5%
Jay Bruce	0.213	0.85	14.9%	39.1%	46.0%	0.0%	17.5%	40.2%
Brandon Belt	0.434	1.21	32.9%	36.7%	30.4%	0.0%	12.5%	45.6%
Michael Morse	0.288	2.72	18.3%	59.8%	22.0%	0.0%	11.1%	34.2%
Lorenzo Cain	0.370	1.66	23.4%	47.7%	28.8%	0.0%	9.4%	31.5%
Khris Davis	0.321	1.03	19.8%	40.7%	39.5%	0.0%	9.4%	34.6%
DJ LeMahieu	0.382	3.06	27.8%	54.4%	17.8%	0.0%	6.3%	26.4%
Nori Aoki	0.316	3.91	18.8%	64.7%	16.5%	0.0%	4.5%	21.6%
Jace Peterson	0.337	1.71	27.8%	45.6%	26.7%	0.0%	4.2%	27.5%
Matt Kemp	0.331	1.61	20.8%	48.8%	30.4%	0.0%	2.6%	34.4%
Angel Pagan	0.377	2.10	23.8%	51.6%	24.6%	0.0%	0.0%	21.3%
Ben Revere	0.285	2.31	32.8%	46.9%	20.3%	0.0%	0.0%	15.4%
Jon Jay	0.278	3.67	22.2%	61.1%	16.7%	0.0%	0.0%	23.3%
Ryan Zimmerman	0.258	1.52	17.2%	50.0%	32.8%	0.8%	9.5%	28.1%
Joe Mauer	0.347	1.83	28.0%	46.6%	25.4%	0.8%	0.0%	24.6%
Eric Hosmer	0.369	1.83	26.7%	47.4%	25.9%	0.9%	23.3%	29.4%
Nick Markakis	0.365	2.23	27.0%	50.4%	22.6%	0.9%	0.0%	31.3%
Average	0.322	1.97	24.2%	48.1%	27.8%	0.2%	10.1%	30.6%

Joey Votto, the patron saint of this leaderboard, has hit a pop-up this year. So he’s tied for 25th and doesn’t make our short version here. Probably should’ve just expanded it to lend the list more legitimacy, but the average BABIPs for this group also helps in that regard.

I mean, look at how short the list of go-get-em-before-the-BABIP-rises players is: George Springer, Jay Bruce, full stop. All right, throw old man Ryan Zimmerman in there. But this list isn’t strong enough to recommend Michael Morse and his crazy ground-ball rate right now, nor is it enough to make Ryan Howard ownable in more leagues than he is already.

Maybe it’s more useful on the other end. Maybe the regression won’t be as crazy for guys like Brandon Belt, Lorenzo Cain, DJ LeMahieu, Eric Hosmer, and Angel Pagan. Regression will come, but what if it’s to .350 instead of .300?

Let’s look at the guys that improved their pop up rates the most. You might want to believe their pop up rates less than guys that have been doing it for years. So call this the list you’ll want to watch closely for the next couple of weeks, to see if their old ways creep back into their swings.

Name	BABIP	GB/FB	LD%	GB%	FB%	PU%	HR/FB	Hard%	DIFF PU
Freddy Galvis	0.393	1.42	23.8%	44.8%	31.4%	1.9%	3.0%	13.8%	8.5%
Brett Lawrie	0.347	1.25	21.4%	43.7%	35.0%	2.0%	5.6%	29.5%	5.2%
Logan Morrison	0.229	1.11	14.4%	45.0%	40.5%	1.8%	13.3%	35.1%	5.1%
Derek Norris	0.343	0.88	14.8%	39.8%	45.4%	0.9%	6.1%	30.6%	4.4%
Jake Marisnick	0.341	1.58	23.0%	47.1%	29.9%	2.3%	11.5%	24.7%	4.2%
Josh Reddick	0.330	1.00	24.5%	37.7%	37.7%	1.9%	15.0%	32.1%	4.1%
Will Middlebrooks	0.221	1.03	23.3%	38.9%	37.8%	2.2%	11.8%	35.6%	4.0%
Brandon Belt	0.434	1.21	32.9%	36.7%	30.4%	0.0%	12.5%	45.6%	4.0%
Brian McCann	0.253	0.80	17.3%	36.7%	45.9%	1.0%	8.9%	28.3%	4.0%
Chris Carter	0.222	0.73	17.4%	34.8%	47.8%	4.3%	18.2%	33.3%	3.9%
Charlie Blackmon	0.344	1.43	24.0%	44.8%	31.3%	1.0%	16.7%	32.3%	3.8%
Brad Miller	0.296	2.56	25.6%	53.5%	20.9%	1.2%	27.8%	37.5%	3.6%
Nolan Arenado	0.272	0.92	16.4%	40.0%	43.6%	2.7%	14.6%	32.7%	3.6%
Yoenis Cespedes	0.333	1.15	14.7%	45.7%	39.7%	2.6%	10.9%	29.3%	3.5%
Jhonny Peralta	0.330	1.36	24.1%	43.8%	32.1%	0.9%	16.7%	38.4%	3.3%
George Springer	0.222	1.61	23.1%	47.4%	29.5%	0.0%	26.1%	34.6%	3.3%
Jason Heyward	0.291	2.95	17.9%	61.3%	20.8%	0.9%	13.6%	27.4%	3.1%
Omar Infante	0.290	1.02	21.7%	39.6%	38.7%	2.8%	0.0%	19.6%	3.0%
Nick Markakis	0.365	2.23	27.0%	50.4%	22.6%	0.9%	0.0%	31.3%	2.9%
Salvador Perez	0.289	1.27	19.0%	45.2%	35.7%	4.0%	11.1%	21.4%	2.9%

Jake Marisnick and Brandon Belt leap off this list. Marisnick has changed his pull/oppo splits, though, so maybe this is all part of his revamped approach. Belt? Maybe he just has gotten a bit lucky with the infield fly balls recently. Maybe if he starts swinging for the fences a bit more, the pop-ups will come back. Derek Norris and Brett Lawrie deserve a bit of a view askance — neither profiles as a BABIP monster otherwise, even if Lawrie goes a bit more oppo than league average and has some speed.

The guys that might regress positively when it comes to PU%? The guys whose PU% is much worse this year than last? Here are those guys.

Name	BABIP	GB/FB	LD%	GB%	FB%	PU%	HR/FB	Hard%	DIFF PU
Colby Rasmus	0.322	0.50	22.7%	25.8%	51.5%	9.1%	20.6%	40.9%	-5.8%
Steven Souza	0.329	1.46	20.3%	47.3%	32.4%	5.4%	25.0%	35.5%	-5.4%
Todd Frazier	0.225	0.73	22.8%	32.5%	44.7%	9.7%	23.5%	38.6%	-4.6%
Adam Eaton	0.282	2.90	16.3%	62.2%	21.4%	5.1%	0.0%	24.3%	-4.3%
Mike Moustakas	0.333	1.13	20.0%	42.4%	37.6%	8.0%	8.5%	24.0%	-3.9%
Adeiny Hechavarria	0.372	1.66	18.4%	50.9%	30.7%	5.2%	5.7%	27.0%	-3.9%
Asdrubal Cabrera	0.279	0.76	15.4%	36.5%	48.1%	6.7%	2.0%	23.8%	-3.8%
Hanley Ramirez	0.245	1.58	21.3%	48.1%	30.6%	7.4%	30.3%	34.3%	-3.7%
Lonnie Chisenhall	0.238	1.02	20.6%	40.2%	39.3%	6.6%	7.1%	24.6%	-3.7%
Starlin Castro	0.319	2.55	15.6%	60.7%	23.8%	7.4%	10.3%	25.4%	-3.6%
Yadier Molina	0.336	1.35	25.2%	43.0%	31.8%	4.7%	0.0%	25.0%	-3.5%
Xander Bogaerts	0.301	1.15	17.3%	44.2%	38.5%	6.7%	2.5%	24.5%	-3.5%
Trevor Plouffe	0.295	1.11	19.2%	42.4%	38.4%	6.1%	13.2%	24.0%	-3.4%
Kevin Pillar	0.280	1.15	16.3%	44.7%	39.0%	5.7%	2.1%	26.2%	-3.4%
Adrian Beltre	0.254	1.03	16.1%	42.7%	41.3%	4.2%	8.5%	28.7%	-3.4%
Buster Posey	0.282	1.17	18.7%	43.9%	37.4%	4.1%	13.0%	35.0%	-3.3%
Luis Valbuena	0.194	0.56	20.6%	28.4%	51.0%	5.9%	19.2%	33.0%	-3.1%
Mookie Betts	0.226	0.77	16.3%	36.4%	47.3%	7.8%	8.2%	30.0%	-3.0%
Adrian Gonzalez	0.375	0.86	28.6%	33.0%	38.4%	4.5%	20.9%	40.7%	-3.0%
Russell Martin	0.325	1.31	22.1%	44.2%	33.7%	5.8%	24.1%	32.2%	-2.8%

The low-sample guys here are interesting. Are we sure we know what Mookie Betts will do when it comes to popping the ball up? Let’s not decide just yet. Let’s say the same for Steven Souza and Xander Bogaerts — a week without a popup might significantly alter their distance from league average in this stat, and their samples before this year weren’t robust anyway.

But the large sample guys? Guys like Todd Frazier and Colby Rasmus? I tend to believe there is something going here. Those two in particular have batted ball profiles that are way out of line with their career numbers. Then again, Frazier’s hard-hit percentage is way high (38%), and Rasmus’ BABIP is already tiny.

Maybe this final list actually should make us most worried about Asdrubal Cabrera? He’s not hitting the ball hard, and he’s hitting more pop-ups. Give him a week or two, maybe, but the signs aren’t positive.