Mediating projections, Part IV: A.J. Pollock

Welcome back to Mediating Projections, the series involving baseball forecasting disagreements and my attempted arbitration thereof. Be sure to check out Parts I, II, and III for more. Like Lucas Duda last week, A.J. Pollock — Arizona's toolsy outfielder, and our current subject — gets praise from ZiPS and derision from Steamer:

Projection	PA	BB%	K%	ISO	BABIP	wRC+	Off	Def	WAR	WAR/600
Steamer	518	6.2%	15.9%	.133	.296	93	-4.1	8.6	2.2	2.5
ZiPS	399	6.8%	17.3%	.179	.328	119	10.1	7.5	3.2	4.8

(As with Duda, Pollock's projected walk rates sync up, so we'll forget about those.) Up to whose expectations will Pollock live? We shall see!

Strikeouts

In the era of the K, putting the ball in play as often as possible matters. Pollock surely knows this, and for the most part, he's succeeded at it. In the four minor-league seasons that preceded his 2012 callup, he fanned in 12.9% of his 1,428 trips to the dish, finishing 0.78 standard deviations below the mean (according to Chris St. John's JAVIER). Results like those meant he'd probably make contact in the show, and indeed he did, putting up strikeout rates of 17.0% and 16.0% in 2013 and 2014, respectively.

Those two don't clash significantly, or so it would seem. And yet, they coincide with vastly different offensive production — in the former year, he hit at a below-average level, whereas the latter season saw him ascend to stardom. Thus, the fact that Steamer and ZiPS project strikeouts to occur at rates resembling Pollock's 2013 and 2014, respectively, may matter. So which should we believe?

There's good reason to think Pollock will continue to avoid going down on strikes. He didn't miss on his swings as often in 2014 (84.5% Contact%, compared to 81.9% in 2013), which reduced his SwStr% from 10.5% to 8.9%. Moreover, his Zone% and Z-Swing% didn't fluctuate much, so he had a similar amount of looking strikes. While he did swing more at pitches outside the strike zone, that shouldn't hurt him too much here. Assuming he keeps up these peripherals, I feel that ZiPS has the upper hand here.

Isolated power

Of course, slightly fewer strikeouts alone didn't elevate Pollock to the ranks of the elite. He chiefly predicated his rise on two things: BABIP (which we'll discuss in a moment), and ISO. In the latter regard, Pollock didn't impress as a youngster, with a .114 mark and a -0.14 z-score across his minor-league career. At first, that feebleness carried over to the majors, where he posted roughly average ISOs in his first two seasons. Approaching the age when power begins to dissipate, he didn't have much hope for the future.

And then he slugged his way to a .196 ISO in 287 believable* 2014 plate appearances. This obviously had a hand in his massive step forward, and maintaining it — as ZiPS predicts — would help him stay there. Will he be able to do that, or will Steamer's subpar projection come to pass?

*Russell Carleton's research has shown that ISO stabilizes at about 160 at-bats for hitters.

Based on the evidence, I wouldn't expect Pollock to hit too many more home runs. In 2014, he hit his fly balls an average of 277.8 feet, a lower distance than he had in 2013 (284.0 feet). Along with an analogous fly ball rate across the two campaigns, this probably means he won't hit as many home runs going forward. But long balls didn't define his power rise — his dinger rate in 2014 still sat below the major-league average, and replacing it with his 2013 rate would yield a .173 ISO, in line with ZiPS's projection.

So from where did that power come? Mainly, doubles and triples — and those, Pollock will probably continue to hit. For one thing, his park will help him considerably with the latter: Chase Field has never ranked lower than third in the majors in triple-enhancing ability, and it's benefited right-handed batters to a greater extent than their counterparts. It's also finished with significantly above-average marks for doubles, although it generally grades out evenly by handedness.

Additionally, Pollock is fairly fleet of foot. He's stolen 26 bases in the past two years, and catchers have only cut him down on six occasions. That reinforces a trend from his minor league tenure, during which he swiped 74 bags at a 79.6% clip. With that speed to aid him, he'll be able to stretch singles into doubles, and doubles into triples.

But the park factors or Pollock's quickness didn't suddenly appear in 2014. What, then, could have precipitated the gains in power? Perhaps a changed technique, as noted in an Arizona Republic piece from earlier this month:

Pollock did make some mechanical tweaks — he hates talking about that kind of thing, but videos of his swings from 2013 and 2014 seem to suggest he lowered his hands and added a bigger leg kick.

That insightful interview might have hit on something. Pollock did undergo something of an evolution, as his spray charts show:

Originally depending mainly on pulled balls for extra bases, Pollock began to take a more balanced approach, and it certainly appears to have paid off.

Of course, this all comes with a significant caveat: Doubles and triples are crazy flukish. As Matt Klaasen's work has illustrated, they have a worse year-to-year correlation than virtually every other hitting metric. This means that, while skill and ballparks can factor into them, luck plays a sizable role as well. With that in mind, I'd say the likeliest outcome is something in the middle of the two projections, but I'd nevertheless prefer ZiPS over Steamer.

Batting average on balls in play

Then we come to that most capricious of statistics, and that which played a key role in Pollock's exceptional 2014: BABIP. In the two campaigns before that wonderful year, 30.4% of the balls he put in play fell in for hits; suddenly, that figure jumped up four percentage points, to .344. Based on the data, I can't say I foresee it staying there.

Line drive rate fluctuates a lot — more so than double rate, triple rate, or even BABIP (via Klaasen's aforementioned analysis) — so we shouldn't jump to conclusions about a year-to-year drop. With that said, Pollock did hit a lot fewer of them in 2014, as his rate fell from a respectable 18.5% to a putrid 14.2%; had he qualified for the batting title, that would have ranked third-lowest in baseball. He did replace those with ground balls, which helps some, but it won't compensate entirely for the absence of liners.

So with that altered batted-ball profile, how did Pollock increase his BABIP? All three possible ball outcomes saw an uptick in that regard, but one stands out:

Year(s)	GB_BABIP	FB_BABIP	LD_BABIP
2012-2013	0.259	0.114	0.722
2014	0.298	0.224	0.774

More hits on ground balls and line drives helped him out, but he wouldn't have improved at all without their airborne counterpart — if he'd posted a .114 fly ball BABIP in 2014, with everything else equal, that would drop his overall BABIP to .304. Will those effective fly balls show themselves again?

Almost certainly not. Among qualified hitters in 2014, the highest fly ball BABIP belonged to J.D. Martinez, at .198. Maintaining something as lofty as Pollock's rate is nigh on impossible, and those types of things don't tend to occur that often. And that doesn't even account for the other batted balls; if Pollock gives back some of the gains he made on line drives and ground balls, he'll plummet even further.

To a certain extent, I can't see Pollock — who BABIPed .336 in the minors, and who still possesses that dangerous speed — falling as far as Steamer prognosticates. But I'll still pick Steamer, simply because there's too much here that can go wrong, and not enough that went right last year.

Conclusion

The fourth (and probably final) edition of this influential and venerated series leaves us without as many firm conclusions as the three that came before it. I feel confident that Pollock will meet ZiPS's strikeout expectations, but the BABIP and ISO could go either way, as they often tend to do. In the end, though, I know I'll keep my eye on this high-flying, face-smashing outfielder, because his future — like the game as a whole — could go anywhere. For a Diamondbacks team that will likely dwell in the cellar for years to come, that'll have to do.

. . .

All data courtesy of FanGraphs, Baseball-Reference, and Baseball Heat Maps.

Ryan Romano is an editor for Beyond the Box Score. He also writes about the Orioles on Camden Depot and on Camden Chat that one time. Follow him on Twitter at @triple_r_ if you enjoy angry tweets about Maryland sports.