Eric Campbell: Was this Souper Utility guy just Souper Lucky Last Year?

Photo from mets360.com

            For a lot of Mets fans, Eric Campbell came out of nowhere in 2014 to be an interesting right-handed bat off the bench who could play the corner positions acceptably. For those that follow the minor leagues closely, he was a long-time depth guy (drafted by the Mets in the 8^th round of 2008) who had been beating up some on AA and AAA pitching since 2012, but who never hit for enough power to profile as a starter at the corner positions he played less than spectacular defense at. Last year, Campbell was one of the last cuts in spring training, and then a slow start from Josh Satin opened a door for a red-hot Soup to get the call. Soup came up and had some big hits in his first week with the club, especially in the pinch-hitting role, and he quickly became a go-to guy for Terry Collins. But Soup didn’t hit in the majors like he had in the minors, so I wanted to take a closer look at his season.

            To start, let’s compare his overall major league stats with what he’s been doing in the PCL the past two years.

Table 1a – Eric Campbell’s stats with Las Vegas in 2013-14 and the Mets in 2014

Team	PA	AVG	BABIP	OBP	ISO	K%	BB%	XBH (HR)	wRC+
2013 LV	425	.314	.357	.435	.161	14.1	15.5	36 (8)	148
2014 LV	163	.355	.398	.442	.170	12.3	12.3	18 (3)	157
Mets	211	.263	.348	.322	.095	26.1	8.1	12 (3)	97

Table 1b – Batted ball and extra base hit rates (GB/BIP, FB/BIP, LD/BIP, XBH/AB)

Team	GB%	FB%	LD%	XBH%
2013 LV	55	28	17	10.6
2014 LV	47	23	30	12.8
Mets	55	27.5	17.5	6.3

            I should point out that the 2014 sample sizes are small, and of course we’re forced to compare his stats starting in the minors to his stats mostly coming off the bench with the Mets (he started 36/85 games played, or 36/128 games the Mets played after his promotion). Obviously batted ball data doesn’t tell the whole story, but his batted ball rates with the Mets are nearly identical to what he did in 2013 with Las Vegas, and his BABIP’s are fairly close too. His line drive rate with Las Vegas in 2014 was probably unsustainably high, but it was also likely a big factor in his BABIP being so high, as well as his increased XBH-rate, which were mostly doubles. As I showed in my last post, which compared Mets players 2014 BABIP with their xBABIP calculated by Jeff Zimmerman using inside edge data, Campbell significantly outperformed his xBABIP last season, and he was by far the biggest offender on the club. To recap, in addition to batted ball rates, the xBABIP formula included whether batted balls were hard, medium or weak contact (thanks to inside edge) and a player speed score, and had a better season-to-season correlation with BABIP than other metrics. Using Zimmerman’s data, Campbell’s xBABIP was .283, a difference of .065 from his actual BABIP. This suggests that if Campbell hits like he did in 2014, he will likely see a significant decrease in BABIP during the 2015 campaign. That’s a problem, because most of his offense was tied to having a high BABIP last year.

            But there are other red flags in Campbell’s 2014 beyond the high BABIP, such as his strikeout rate doubling after his promotion, while his walk and extra base hit rates took significant hits after his promotion. His walk rate was still average with the Mets, so it’s not that concerning, and even less so when you consider that it was trending up throughout the season, as shown in table 2.

Table 2 – Eric Campbell’s half season splits with the Mets in 2014 (1^st half represents Mets games 1-81, 2^nd half represents Mets games 82-162)

Month	PA	AVG	BABIP	OBP	ISO	K%	BB%	XBH%	wRC+
1st half	82	.307	.400	.341	.133	25.6	4.9	10.7	122
2nd half	129	.235	.313	.310	.070	26.4	10.1	3.5	81

            Unfortunately, even though he started walking more in the second half, he wasn’t nearly as good of a hitter. It’s interesting, despite poor K: BB rates in the first half, his overall numbers are great, and despite a high BABIP, his average ISO made it seem like he wouldn’t regress too much in the 2^nd half. But, when you consider that he had a 59% GB-rate in the first half, his high BABIP and average ISO were somewhat sketchy. His GB% dropped to 52% in the second half, but his BABIP dropped near average, and his ISO dropped to Ruben Tejada levels. Without a high BABIP, his high strikeout rate makes it impossible for him to maintain a decent batting average and on-base percentage. Combine that with below average power, and we’re looking at the production of a backup middle infielder, and those guys generally provide value with their glove. While there is value in having Campbell’s defensive versatility, he is a scratch defender at most corner positions, and below average up the middle, so it’s just the value of having a body who can play many positions, not a defensive upgrade anywhere. If Campbell had kept up his first half ISO down the stretch, his 2^nd half production would’ve been around a 90 wRC+, which is more acceptable for a bench player. The concern is that the league adjusted to Campbell by the 2^nd half, and Campbell didn’t make any adjustments to the league’s adjustments, finishing the season with merely a 9 wRC+ over his 33 September PA’s.

Table 3 – Eric Campbell’s L/R splits with the Mets in 2014

Split	PA	AVG	BABIP	OBP	ISO	K%	BB%	wRC+
L	105	.250	.328	.324	.109	26.7	9.5	98
R	106	.276	.366	.321	.082	25.5	6.6	96

            Finally, let’s look into those L/R splits, which appear pretty even by wRC+ and PA. But you can see that much of Soup’s success against righties was BABIP driven, as he provided a below average K%, BB% and ISO. And just like with his first half stats, his high BABIP in those splits seem at odds with the 61% GB-rate he had on balls in play. His BABIP against lefties was more reasonable, and he still provided nearly average offensive production despite a high K%. Campbell generally provided more offense against lefties throughout his minor league career, so it’s not surprising he would do the same at the highest level, but you hope for better than average production against one side from your bench players, and Campbell did not do that in 2014.
            Overall, Campbell is an interesting potential bench piece for the Mets for the next several years – they’ll have team control of him for the next 5 years, and he has all 3-option years left per TPGMets. To answer the question posed in the title, no, I don’t think Soup was just lucky in 2014, though it seems clear that BABIP played a large role in his overall production. Considering this, and Zimmerman’s xBABIP results, I think there is real concern for Campbell having a disappointing 2015 at the major league level. Because of this, I don’t think his spot on the bench should be a foregone conclusion. Of course, the season ended before Campbell had a chance to readjust to major league pitching, so we’re not working with a full dataset yet, and he had a full offseason to study how the league adjusted to him. While he can backup multiple positions, he’s not the first backup at any spot (other than maybe 3B, though the Mets have other options who can play there), and that’s not very useful if he’s not hitting off the bench. With Cesar Puello at risk of being lost if he doesn’t make the Mets out of Spring Training, and having shown improvements throughout 2014, I think it might be best for the franchise if Campbell starts 2015 back in the minors.

A new understanding of the Sun's magnetic field. Visitors often ask us about this! #astronomy http://t.co/GrX30mw6SN pic.twitter.com/MusAno1Vim
— Public Observatory (@SIObservatory) February 26, 2015