Mediating projections, Part III: Lucas Duda

Welcome back to Mediating Projections, wherein I bloviate about baseball forecasting systems and rely on the most subjective of means to determine their legitimacy. For some additional background, check out the first two installments. Today, we'll scrutinize the case of Lucas Duda, the mighty (and maladroit) Metropolitan who broke out in 2014. Like Julio Teheran, the first subject of this series, Duda receives more praise from ZiPS than he does from Steamer:

Projection	PA	BB%	K%	ISO	BABIP	wRC+	Off	Def	WAR	WAR/600
Steamer	624	11.6%	24.4%	.186	.278	115	9.2	-12.5	1.7	1.6
ZiPS	562	11.4%	21.2%	.219	.291	137	22.9	-10.3	3.2	3.4

According to ZiPS, Duda's 2015 will mirror his 2014; according to Steamer, it'll resemble the years that preceded it. Which system should we believe? Let's find out! (Because the two don't differ much with regards to walks, I won't look into those.)

Strikeouts

Coming up in the minors, Duda had a reputation as a free-swinger. Across 2007 and 2008, his first two professional years, he went down on strikes in 20.9% of his plate appearances...in single-A ball. Nor was that a fluke — he owns a 0.32 career strikeout rate z-score in the minors (courtesy of Chris St. John's JAVIER). Needless to say, that didn't inspire much hope for his success at higher levels. Sure enough, his first four major-league campaigns saw him fan in 23.4% of his plate appearances for a 121 K%+. As a 28-year-old on the wrong side of the aging curve, he would probably get even worse in the years to come.

That didn't happen, though. For 2014 as a whole, he ended up striking out in 22.7% of his 596 trips to the dish, a rate only 11% worse than the major-league average. This obviously helped him hit for more power — a strikeout generally doesn't go for extra bases — and significantly improved his hitting overall. Will it remain this low, as Steamer suggests, or bounce back to its previous level, as ZiPS predicts?

The thing about strikeouts is that they can jump around without much change in peripherals. Thus, before reaching any judgment regarding Duda's fluctuation, we should look into those. We'll start with strike rate (as a percentage of all pitches), which didn't change much for him:

Years	Str%	Str%+
2010-2013	60.0%	94
2014	60.1%	93

Since a batter can strike out only on, well, strikes, having fewer of them will help him eschew the K. Of course, not all strikes are equal — a batter can go down solely on those of the looking and swinging variety. And as with his strike rate, Duda didn't see much of a difference there:

Years	L/Str%	L/Str%+	S/Str%	S/Str%+
2010-2013	33.2%	119	17.0%	109
2014	32.2%	117	18.5%	113

But we can go even deeper than this. PITCHf/x data can give us O-Swing% and Zone% (which can give us an expected strike rate*), Z-Swing% (which can give us an idea of how many looking strikes a hitter should have), and Swing% and Contact% (which can give us SwStr%**).

*xStrike% = 1 - (1 - O-Swing)*(1 - Zone%)

**SwStr% = Swing%*(1 - Contact%)

So how do these metrics appraise Duda? For the most part, they reach the same conclusion as their counterparts:

Years	xStrike%	Strike%+	Z-Swing%	Z-Swing%+	SwStr%	SwStr%+
2010-2013	59.1%	92	58.1%	93	9.5%	104
2014	59.5%	93	58.6%	93	10.2%	107

The absence of any support for this strikeout decrease makes it seem rather flukish, which Steamer believes (as do I).

Isolated power

Then we come to power — the defining element of Duda's 2014 breakout. Finally tapping into the prodigious potential that he displayed in 2010, he slugged his way to a .228 ISO, despite playing in a park that heavily depresses the hitting folks. ZiPS thinks he won't miss a beat in 2015, while Steamer feels that he'll fall back to his 2010-2013 level. In which projection should we place our faith?

For one thing, there's the aforementioned stadium, which makes his task much harder. Plus, power surges tend to dissipate to some extent in the subsequent year, as research by BtBS's Jon Roegele has shown. Nevertheless, this one looks to me like it'll stick around, because Duda's made legitimate changes to his game.

Mike Petriello pointed out in August that Duda began hitting the ball to all fields in 2014. This balanced approach has served him well — by going with the pitch more, he no longer felt compelled to drive it where he wanted. That showed itself in his exceptional fly ball distance (also noted by Petriello), as well as his higher fly ball rate: 49.0% of his balls in play went airborne, compared to 44.5% before.

This shouldn't come as a surprise, since Duda's always possessed power. And it didn't shock Sandy Alderson, the Mets' GM, who has stated that clout came into play when determining Duda's fate:

The tiebreaker in trading Ike Davis over Lucas Duda was that the exit velocity of baseballs off Duda's bat was harder.

— Adam Rubin (@AdamRubinESPN) March 15, 2015

At the end of the year, that decision still looks like a sage one — and one that will continue to bear rewards for the foreseeable future. For that, I comfortably award ZiPS the victory here.

Batting average on balls in play

But what about when Duda doesn't knock the ball out of the park? In those situations, he actually declined a bit in 2014: His BABIP fell from .296 in the first four years of his career to .283 in the fifth. It could bounce back (according to ZiPS), or it could sink even lower (according to Steamer). Which outcome has a greater chance of coming to pass?

It certainly appears to be the latter. For one, Duda hit more fly balls in 2014, which helps his power but hurts him here, because fly balls don't go for hits very often (unless they leave the park). If he reverts to hitting as many grounders as he did before making the leap, his power will decrease, and he'd almost certainly like to avoid that. The park in which he plays makes that worse, as it'll likely give him no advantages when he puts the ball in play.

Most importantly, there's the matter of the batted balls which caused his lower BABIP. Don't blame fly balls or line drives — they turned into hits just as often in 2014 as they did from 2010 to 2013. Rather, ground balls didn't serve him like they had before, with a .250 BABIP prior to last year giving way to a .214 one during it. But why did their lack of quantity also bring a lack of quality?

The answer lies in every sabermetrician's favorite topic: the shift. While Duda altered his methods with regards to fly balls, he stayed largely the same on grounders:

Duda became even more of a pull hitter than he ever had been, and the opposition caught on. Jeff Zimmerman's research showed that Duda hit against the shift 245 times in 2014, with a drastically lower BABIP in those instances (.253) than in the normal ones (.321). Unless he counteradjusts, Duda will probably live up — or down, I suppose — to Steamer's expectations.

Conclusion

In the end, we're a little more pessimistic about Duda, but not too much. This highly scientific exercise proves that he'll become more of a true power hitter in the coming campaign — low average, oodles of strikeouts and walks, and home runs galore. After four years of general awfulness from both parties, he and the Mets would probably accept that.

. . .

All data courtesy of FanGraphs, Baseball-Reference, Brooks Baseball, 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.