Betancourt, xBABIP and Batted Balls

I think most people who frequent Brew Crew Ball are familiar with the stat BABIP (batting average on balls in play), which is usually taken as a measure of how "lucky" or "unlucky" a hitter has been depending on the difference between his BABIP and the baseline of .300.  However, some batters are able to consistently maintain high or low BABIPs, leading one to believe that the simple formula for BABIP doesn't always tell the whole story.  As a result, two writers at the Hardball Times came up with a metric called xBABIP (expected BABIP) a few years ago that uses a number of advanced factors to give a better estimate of what we should expect a hitter's BABIP to be (i.e. not necessarily .300) [link]. 

These stats can be useful in examining Yuniesky Betancourt as his extremely low BABIPs over the last two or three seasons (.256 and .267 in the previous two) - seasons which helped lead to a lot of hyperbole about Betancourt being the "worst player in baseball" -  suggest something weird may have been going on.  Did Betancourt suddenly start doing something different that made his offensive production plummet in 2009 (and continue to be depressed by a low BABIP in 2010)?  I decided to take a quick look at xBABIP and Betancourt's batted ball data to see what was going on:

xBABIP uses a number of advanced stats, but the Hardball Times has also made a simple tool available for making a quick estimate of xBABIP [link].  For the sake of consistency, I'm using this simple xBABIP calculator throughout the post, even though this is a fairly rough estimate of the more advanced xBABIP metric.  First, it might be useful to see how Betancourt's xBABIP has compared to his BABIP across his career:

Betancourt's xBABIP has actually been remarkably consistent (the largest outlier in 2005 is only over 228 PA) and for the first few years of Betancourt's career his BABIP mirrored his xBABIP incredibly closely.  However, his BABIP began to fall relative to his xBABIP in 2008 and in 2009/2010 it fell off a cliff relative to his xBABIP. 

Using xBABIP, we can come up with a number for xHits and from there xAVG/xOBP/xSLG (conservatively assuming that each additional hit would have been a single).  For the record, here are Betancourt's career lines (excluding the short 2005 season) with a rough estimate of wOBA [(OBP*1.75+SLG)/3] and with the previous three seasons adjusted for xBABIP:

Those lines not only make Betancourt look remarkably consistent, but also a potentially very solid offensive shortstop.  To put things into context, the .328 expected wOBA last season would have ranked fifth among qualified shortstops (right behind Jose Reyes' .329).  Betancourt's recently-terrible defense is much easier to swallow if he's hitting like that, and the xwOBA from his seemingly "worst player in the majors"-defining 2009 season is 36 points higher than his actual wOBA, which over 508 PA works out to a difference of somewhere around 1.5 wins above replacement! (again a rough estimate)

How can we explain the huge drop in BABIP versus xBABIP?  The advanced factors that go into xBABIP suggest that Betancourt has been doing everything basically the same, and his batted ball percentages seem to support that:

If anything, he's hitting slightly more fly balls and slightly fewer ground balls, but the difference from 2007 on is minimal.  There doesn't seem to be any major systematic changes in the number or type of pitches he's swinging at or making contact with either (although this Fangraphs article suggest that a higher percentage of his swings have been at pitches out of the zone, I don't think that by itself can come close to accounting for the drop in BABIP, and the trend is inconsistent). 

Is Betancourt just making weak contact and hitting a lot of weak flies, or has he gotten fat and slow and no longer reaching base on grounders?  Breaking his BABIP for each year into the different types of batted balls (ground balls, fly balls, line drives) gives some useful insights:

Although showing a steady decline, his BABIP on ground balls stayed relatively close to the league average in 2008 and 2009, but plummeted relative to the league average in 2010.  His BABIP on fly balls fell in 2009, probably due to the spike in IFFB%, but was back above average last season. It seems feasible that Betancourt is getting older and slower, and thus his gbBABIP has fallen, but even if he has below average speed now I don't think he's to such an extreme as to drop his gbBABIP versus the league average that much (xBABIP does take speed into account, albeit in a very limited manner in the simple xBABIP calculator that I used).

In 2007-2009 his line drive BABIP was inexplicably low, but it recovered last season.  Generally line drives are the ideal balls for hitters to be putting into play, so the ridiculously low ldBABIP definitely had a large negative effect on Betancourt's numbers (especially since he's a guy who puts a lot of balls in play and who has a roughly average LD%).  I'm not sure what explanations there could be for why so few of his line drives fell for hits in those years, but I would be interested to hear if anyone has any theories.  If you're hitting line drives, you're usually hitting the ball pretty well.  The extremely low ldBABIPs in those years also suggest that the xBABIP adjusted batting lines above are very conservative, as a lot of the balls Betancourt was getting "unlucky" on were line drives, which can often result in extra bases instead of just singles.

If Betancourt has suddenly become a terrible hitter who cannot make solid contact (one plausible explanation for the fall in BABIP) he's being strange about it.  Based on the random fluctuations in the above component BABIPs, that explanation seem unlikely to me (remember that his GB%/FB%/LD% have remained relatively constant).  Betancourt hasn't become worse at one particular thing (i.e. he hasn't started consistently hitting weak grounders or weak fly balls), instead it looks a lot like he's just been getting very, very unlucky with balls in play.  He's average in gbBABIP one season and terrible the next, terrible in fbBABIP or ldBABIP one year and above average the next.  It seems fairly random, as opposed to a systematic decline in his skill set. 

If Betancourt had put up the xBABIP-adjusted lines posted above, including the very-good-for-a-shortstop .328 xwOBA from last season, I think a lot of us might be much less worried about having Betancourt on the team next season.  But what about his terrible defense? 

This is purely speculation, but based on a lot of the comments I've read about Betancourt around the web, it seems like he's a player who has the tools but can have some serious motivational problems on the field.  Perhaps not coincidentally, his defense fell off a cliff alongside his BABIP.  Could the sharp increase in balls not falling in for hits and line drives being snared have contributed to Betancourt's frustration and his "lazy" approach at shortstop?  I don't know, but it's interesting to think about.

I think there's plenty of reasons to expect Betancourt's BABIP to rebound next season, and I don't see any compelling reason to think that he's suddenly become the type of player who can sustain such a low BABIP indefinitely (nor has the xBABIP metric).  The boost in offense alone substantially raises Betancourt's value, but it's fun to be optimistic and think that maybe the resulting increase in Betancourt's "luck" at the plate (as well as playing on a winning team) might help motivate Betancourt to play better defense too.  I'm not saying he's a great player, but at this point it seems we're basically left with finding things to be optimistic about in Betancourt and praying for the Deity's durability.