How the Rays offense improved BABIP in 2013

Last off-season I had serious concerns that the Rays had a problem with luck.

For two seasons, the Rays had out performed their in-play batting average by posting offensive metrics above league average, even though the offense had the second lowest BABIP in all of baseball in 2011, and third lowest in 2012. These were levels we hadn't seen since Andrew Friedman took over in 2006.

Season	Team	BABIP	wOBA	wRC+	WAR
2006	Devil Rays	.285	.316	89	10.2
2007	Devil Rays	.318	.335	103	17.3
2008	Rays	.302	.335	105	30.4
2009	Rays	.303	.342	109	32.8
2010	Rays	.293	.326	106	30.5
2011	Rays	.281	.319	103	28.8
2012	Rays	.284	.311	100	21.2

My biggest concern was that this trend of above average performance despite poor BABIP was unsustainable and I believed that the Rays could improve their offense greatly (and remove their dependence on superior pitching) by adjusting for luck.

While making adjustments to mitigate the risk of luck is somewhat like trying to catch lightning in a bottle, some research last season indicated that such a thing was possible.

At last year's Sloan Analytics Conference, Dan Rosenheck presented pitching research that showed how aspects of BABIP can be directly impacted by specific batted ball tendencies and the plate disciplines of hitters, which he discussed further on Fangraphs.

Rosenheck's observation was that the future BABIP production of pitchers can be correlated to their past performance in contact rates inside the strikezone and their infield flyball rates. Transitively, this logic is applicable for the luck of hitters as well.

Based on this research, I postulated that if the Rays could improve their contact rate on swings inside the strike zone (Z-Contact%, which the Rays were league-worst in 2012) and infield flyball rates (IFFB%, second worst), then the Rays' in-play batting average (BABIP) would improve overall. In fact, my observation was that the Rays had already made the moves necessary.

Instead of signing blame to defenses faced, ballparks played in, or talent of the batters, I suggested last March that the Rays would improve on each of those metrics, based on the changes the team made around the diamond.

Here's how that played out, with a comparison of 2012's role players to 2013's:

First Base	PA	Z-Contact%	IFFB%
Carlos Pena	600	75.5%	16.5%
James Loney	598	92.0%	6.0%

Last off-season I had this to say about Loney's potential to increase the Rays' performance in batting:

When the Rays signed Loney, we knew the first baseman boasted a strong glove, one third the strikeouts of Pena, an inability to hit southpaws, and not a huge fall off for on base percentage. In 2012 the league average wOBA for first baseman was .329, Loney's same mark in 2011 and giving him bounce-back potential for 2013.

Loney not only delivered what the Rays needed by bouncing back from his career lows in 2012, but also proved he wasn't a platoon bat by hitting well against same-handed pitching. He exceeded expectations by delivering a .339 wOBA. He even improved his phenomenal contact outside the strike zone, offering at 32.4% of pitches beyond the zone and making contact on 85%. His O-Contact% was the fifth best in baseball, and only Victor Martinez offered at more pitches with a higher contact rate than Loney.

This offensive complement to his defensive game (which was nominated for a gold glove, with the rest of the infield) gave the Rays enough confidence to hand him the largest free agent contract ever under Andrew Friedman.

UTIL	PA	Z-Contact%	IFFB%
Jeff Keppinger	418	95.6%	11.2%
Kelly Johnson	407	85.8%	12.1%

In 2012, Keppinger filled in at second base while Ben Zobrist primarily manned right field. In 2013, Zobrist returned to the infield and Keppinger's at bats were effectively replaced by Kelly Johnson. My expectation was that this would be a wash in transition, with Johnson posting lower contact but lower infield flies. However, Johnson was unable to keep his IFFB% down and his luck suffered because of it.

Kepp's .332 BABIP and .352 wOBA were replaced by Johnson's .276 BABIP and .314 wOBA -- a fall from rather above average to somewhat below. Johnson played primarily in left field, which is not his traditional position.

The bat performed well enough in the first half of the season, in spite of these peripherals, but his playing time diminished as Wil Myers transitioned into the line up. Matt Joyce moved to left field, and then injuries and the acquisition of David DeJesus effectively ended his playing time all together. This is the upgrade worth focusing on.

After 291 plate appearances in the first half, Johnson would only find 116 more across the rest of the year. By comparison, in the second half, DeJesus picked up 117 plate appearances and displayed a 93.5 Z-Contact% and 5.7 IFFB%.

DeJesus was featured prominently in the postseason run as the Rays' lead-off hitter, and looks to have earned his spot in the rotation for at least two more seasons. This offseason, the Rays let Johnson walk (to the Yankees, no less) and then signed DeJesus to a two year extension with an option for a third.

Short Stop	PA	Z-Contact%	IFFB%
Elliot Johnson	331	84.9%	14.5%
Sean Rodriguez	301	81.8%	17.6%
Yunel Escobar	578	90.2%	6.8%

The other significant replacement on the roster was Yunel Escobar, our other substantial improvement in luck. After a slow start in April, Yunel was able to put together a .297 BABIP performance from May through the end of the season.

Escobar gave consistency to what was a revolving door for the Rays in 2012, and there is an argument to be made that Escobar has become the best overall shortstop in the American League. Part of that is his successful approach at the plate, one that mitigates the risk of luck as much as possible.

Third Base	PA	Z-Contact%	IFFB%
Evan Longoria	312	85.8%	4.6%
Ryan Roberts	209	86.7%	12.3%
Evan Longoria	693	83.4%	4.4%

Finally, the other significant improvement was having Longoria healthy for a full year.

Due to a hamstring injury which would require surgery, Longoria was limited to 312 plate appearances in 2012. Removing his .313 BABIP was not helpful in improving the team's overall luck. At the time of Longoria's injury the Rays were last in batting average in all of baseball, and those peripherals did not suggest hope for it to improve.

Longoria was indeed able to play a full season with 693 PA's in 2013, posting similar numbers to his career averages.

Conclusion

After consecutive years of ranking among the worst in categories related to luck, the Rays offense was able to significantly improve in-play batting average to a mark of .295, comparable to the MLB average of .297.

Season	Team	BABIP	wOBA	wRC+	WAR
2011	Rays	.281	.319	103	28.8
2012	Rays	.284	.311	100	21.2
2013	Rays	.295	.324	108	30.3

In kind, the offense improved from the tenth best offense by WAR in 2012 to the second best overall in 2013.

The improvement was just in time, as the pitching that ranked third overall in 2012 had injuries throughout the rotation in 2013, falling to fourteenth in WAR. If there was a time for the offense to step up and carry the team's production, it was last season, and the team delivered with another post season run.

Perhaps much of that was a result of the adjustments made to improve in-play batting average.