Comparing Individual Mets 2014 BABIP and xBABIP | Astromets Mind

Tuesday, February 24, 2015

Comparing Individual Mets 2014 BABIP and xBABIP

            At this point in the season (offseason?), there aren’t many new topics to discuss until the games finally start. The biggest news story to break in Mets spring training camp so far is that Lucas Duda has been dealing with a slight left intercostals/oblique strain from overworking this offseason. While that doesn’t sound serious right now, it’ll be very important that Duda gets passed that issue before the season starts. While I was looking for other interesting Mets news on twitter today – and tweeting out some fun David Wright #FaceOfMLB votes – I ran across an interesting tweet from ESPN’s Mark Simon. The tweet highlighted the following information:

Table 1 – Results by batted ball type from 2014
Batted Ball Type
Ground Balls
Fly Balls
Line Drives
*tOPS+ is the OPS for the split relative to a player’s total OPS

            While this information isn’t particularly new, it got me thinking about how the current Mets fared in terms of BABIP in 2014. BABIP isn’t particularly a new concept, and if you like advanced stats even just a little, you probably have a rough idea of either how most Mets fared in terms of 2014 BABIP, or at least which players outperformed/underperformed league average BABIP expectations. Even though league average BABIP has been pretty steady over the past 10 years, individual players BABIP can vary quite a lot from season to season, and so analysts have come up with various formulas for an expected BABIP formula (xBABIP). After some fairly simple xBABIP equations initially, these xBABIP calculations have evolved to take into account more factors than are sometimes even available to the casual fan. For example, Jeff Zimmerman at Rotographs introduced one that uses Inside Edge data and player speeds. I collected the relevant xBABIP information for the 2015 Mets and compiled them into Table 2.

Table 2 – Comparing Mets 2014 BABIP and xBABIP
Lucas Duda
Dilson Herrera
*xBABIP values from this Fangraphs piece

            Those players with a higher xBABIP than BABIP in 2014 were ‘unlucky’ (highlighted in blue), while those with a lower xBABIP were ‘lucky’ (highlighted in orange). I would've included at least one Mets pitcher (Jacob deGrom) had they made the list, but Zimmerman did not include them.
            Granderson and Duda ran into the most infield shifts among players on the Mets in 2014, coming up to the plate against a shift nearly every time up, so it’s not surprising to Zimmerman that their BABIP is less than their xBABIP – he concludes something similar about heavily shifted players like Chris Davis in the piece linked after Table 2. As John Dewan mentions in the post linked above (and here), the BABIP on groundballs and short liners is about 30 points lower when there is an infield shift as opposed to no infield shift. So while these two have the most room for improvement by dBABIP, their ~30-point drop from xBABIP to BABIP is actually within range.
            Aside from the Bill James handbook, which I do not currently have, I haven’t found where to get shift info for each player. Because of this, although I know Travis d’Arnaud pulled a high number of grounders to the left side of the infield, I can’t be sure if he ran into a higher number of shifts for a righty. If not, he is the Mets player most likely to see an increase in BABIP in 2015, which is obviously great, since that was one of the biggest reasons he wasn’t more valuable offensively in 2014. His career minor league BABIP is .323, so his .259 BABIP in the majors last year was surprisingly low.

Figure 1 - Heat map courtesy of

            Flores, Recker, Mayberry and Herrera are the other 2015 Mets who would likely see a boost in BABIP this season per this xBABIP, which is a good sign for the Mets SS situation. Like d’Arnaud, Flores had a much better career minor league BABIP (.318) than he showed in the majors during the 2014 season, so it’s not crazy to expect an improvement there. Recker’s 2014 xBABIP is similar to his 2013 BABIP, and his decreased BABIP and BB% in 2014 hurt his offensive contribution, as he saw a drop in wRC+ from 90 to 75. Mayberry had a career low BABIP in 2014, dropping 51 points below his .280 career BABIP, but still managed to produce a 107 wRC+, so maybe the Mets can get even more from him if used properly in 2015.
            Wright, Lagares, Campbell, Cuddyer and den Dekker all appear to have outperformed their 2014 xBABIP by at least 30 points. We know David was not himself in 2014, so it’s not clear how helpful that is. Cuddyer was obviously the big free agent acquisition that many Mets fans, myself included, weren’t a big fan of. His BABIP’s while playing half of his games in Colorado were career bests by a good margin, and it’s hard not to attribute a large portion of that to playing in such a big, offensively charged stadium. He had a .401 BABIP with 46% of his plate appearances coming at home in 2013 (he did have a .365 BABIP on the road that year, so 2013 gets more of a pass), and a .421 BABIP with 43% of his plate appearances coming at home in 2014 (compared to a .303 BABIP on the road). Still, a .320 BABIP from Cuddyer would be fine, so let’s just hope he ends the season with more plate appearances than Eric Campbell this year.
Speaking of Campbell, Mr. Versatility appears to have also been Mr. Very-Lucky. His best months by wRC+ (125 in June and 145 in July) were also months when he hit to a .500 BABIP without providing much power and while striking out a lot. After his initial BABIP success bought him more playing time in Terry’s eyes, Campbell became a dead weight on the offense, hitting to only a .217 BABIP and 47 wRC+ over his last 97 PA. Matt den Dekker reworked his swing prior to 2014, so it took him a little longer than normal to get going, but he got red hot with the 51s in July, and brought that hot streak with him into the majors (minus the power). The potential drop in BABIP in 2015 should be offset by more power, as he was able to provide a .200+ ISO over his last 3 months with Las Vegas, and finally cracked the .100 ISO mark over 68 PA in September.
Finally, considering Wright’s bum shoulder played a role we can’t quantify, the most important player from this group is Juan Lagares, who provided a significant boost in offense in 2014 thanks in large part to a boost in his BABIP. I’m not 100% sure which speed score Zimmerman used for his study, but Lagares may have been underrated by the score, as he showed a new level of stolen base ability later in the season that would raise his score if sustained over a full season. Lagares had a .334 career minor league BABIP, so his .341 BABIP in 2014 never seemed to high, but a drop to a .297 BABIP would have dropped his overall value close to the 3 fWAR provided in 2013. Still, considering he only had 450 PA’s in 2014, that’s perfectly fine for a Mets centerfielder.
            Murphy, Tejada and Nieuwenhuis are the final three on the list, and while they all outperformed their xBABIP’s in 2014, none did so by more than 16 points. Perhaps the most interesting thing from the entire chart is that Kirk didn’t outperform his xBABIP by a significant amount, as a .361 BABIP is unusually high – it ranked 19th of 410 players with at least 130 PA in 2014 – even compared to his .338 career minor league BABIP.

            In total, the oranges outnumber the blues 8-7, but two of the blues aren’t expected to see improvements in 2015 unless Rob Manfred quickly introduces anti-shift rules (I hope he never does). Also, of the two groups, the orange group contains more significant players for the Mets success in 2015. The good news is that players are constantly over and underperforming what they’re supposed to do, and 2015 will be no different, so just sit back and enjoy what should be the best Mets team in years.

  • 0Blogger Comment
  • Facebook Comment
  • Disqus Comment

Leave your comment

Post a Comment

comments powered by Disqus
submit to reddit