Cleveland Sports Insiders

Ernie Camacho's favorite blog…

Asdrubal Cabrera and Inconsequential Bouncebacks

9 Comments

Presumably a Popular 'Pick 3' Combination After the Wild Card Game

Presumably a Popular ‘Pick 3′ Combination After the Wild Card Game

After a 2013 Wild Card game performance in which Asdrubal Cabrera went 0-4 with a strikeout and the single most deflating play of the season, a double play in the bottom of the 4th with the bases loaded, the phrase ‘bounce back’ was not an appropriate word for Cleveland audiences. However, as the Browns’ and Cavaliers’ seasons have progressed and impressed upon fans’ minds the nature of true cosmic despair in sports, where ‘hope’ is a meaningless word even in jest, the author feels he can now make the argument for an Asdrubal Cabrera bounce-back without fear of mob justice.

The term ‘bounce-back,’ in this case, does not refer to a 40-point increase in Batting Average or a 20-home run leap from 2013; nor, moreover, does it refer to the concept of a ‘contract year,’ since it’s difficult to believe that baseball players, who are already competing against the toughest competition in the world, can merely will themselves to perform better than they already have. Baseball’s hard – Major League Baseball, orders of magnitude harder.

A probable bounce-back, rather, would be both less dramatic and less mystical than either of these two possibilities. Instead, bad luck relative to the mean caused Asdrubal Cabrera’s numbers to undersell his actual performance; namely, Cabrera made frequent, solid contact at the plate, but his batting average did not reflect his solid contact. Assuming his contact ability remains going into 2014, this bounce-back alone figures to improve Asdrubal Cabrera’s value to the club by about 1 WAR – which still, admittedly, leaves him a slightly below-league-average shortstop.

These bounce-backs do not take away from two serious, unlikely-to-regress flaws going into 2014, the most notable of which is his fielding. Ultimate Zone Rating, a defensive metric based on three factors (a defender’s range, double-play conversion rates, and errors), estimates that Asdrubal Cabrera’s defense cost the Indians 12.8 runs on defense relative to the league average defensive shortstop; in other words, Cabrera’s defensive performance – removed from game context – cost the Indians about one win in 2013. On one hand, there’s no reason to believe that, at age 29, Cabrera’s defense is poised for apocalyptic decline, but on the other hand, said defense is equally unlikely to be anything better than below-average, given that Asdrubal’s range and defense have been below average for the entirety of his major league shortstop career. Entertaining though Asdrubal’s defensive flair might be, such flair exists only because his range is below league average, meaning that even an average defensive shortstop would make Asdrubal’s exhilarating plays, as well as many more plays, in a high-percentage, routine manner – definitively proving that baseball stat-heads like the author are indeed destroying the excitement of the game. Behind-the-back fireball throws to first don’t win championships; close-cropped, clean-shaven throws do. Suit up, John Wayne.

In light of this, Cabrera’s offense is almost the entire reason he is valuable to a major-league team; going into 2014, however, legitimate concerns do exist about his plate discipline. In 2013, Cabrera’s strikeout rate, clocking in at 20.4% of his Plate Appearances, was both the highest of his career and the first time since 2008 that he struck out at a worse-than-league-average rate, with career-highs in both his Swinging Strike Rate (9.5%) as well as his Out-of-Zone Swing Percentage (34.4%). Plate discipline, moreover, is not a bounce-back trait, so as of this point, there’s no reason to believe that Cabrera’s 2014 K rate will be much different from 2013.

Yet while it is true that Cabrera was little better than a replacement-level player this past year, the fact remains that from 2009 to 2012, Asdrubal was the fifth-best offensive shortstop in the game by most holistic metrics. Whatever value his defense lacked, his offense more than made up for. And while Cabrera’s strikeout rate is highly unlikely to rebound, an uptick in strikeouts is not, on its own, enough to cause Cabrera to decline from a top-tier offensive shortstop to John McDonald at the dish. The reason for his decline is almost entirely his Batting Average.

  AVG OBP SLG OPS BB% ISO BABIP
2011 0.273 0.332 0.460 0.792 6.6% 0.187 0.302
2012 0.270 0.338 0.423 0.762 8.4% 0.153 0.303
2013 0.242 0.299 0.402 0.700 6.2% 0.159 0.283

Asdrubal Cabrera: A Batting Life (Stats: FanGraphs)

While Cabrera’s On-Base Percentage and Slugging Percentage both declined in 2013, his walk rate and ISO (Isolated Power: Slugging Percentage minus Batting Average, a way to measure extra base hits and hence power) each varied only slightly from 2012 – in short, it was his poor batting average that caused his decline in OBP, SLG, and (unsurprisingly) OPS.

Yet Batting Average is not itself a root cause of offensive success but a result of the manifestation of his batting skills: namely, of strikeout rate, of home run rate, and of Batting Average on Balls in Play (BABIP). The first of these increased and has been accounted for, and the second has held steady; the last of these, BABIP, explains how frequently struck balls fall in for hits, and it is this stat that is the linchpin of the bounce-back argument for Asdrubal Cabrera.

Put simply, BABIP is mostly a product of four factors: how fast the batter can get to first, how many balls in play were hit well, how many balls in play were hit poorly, and where the defenders were. These four factors are represented by the following four quantities: Speed, Line Drive Percent, Infield Fly Ball Percent, and ‘Luck.’ There are a number of other BABIP-influencing factors, but these four are by far the most important. Given that Asdrubal is about as fast as a league average player using the FanGraphs Speed metric, we are left with three main variables: Line Drive Percent, Infield Fly Ball Percent, and ‘Luck.’

  AVG BABIP LD% IFFB%
2011 0.273 0.302 17.5% 5.9%
2012 0.270 0.303 23.4% 10.6%
2013 0.242 0.283 23.0% 5.6%
LEAGUE AVERAGE 0.253 0.297 21.2% 9.7%

Cabrera’s Batted Ball Profile vs. 2013 League Average (Stats: FanGraphs)

In 2013, Cabrera’s BABIP was well below league average. However, his solid contact – measured by Line Drive Percent – was well above league average; moreover, because Line Drives fall in for hits at an incredible .693 clip per this article on Talking Chop, LD% is the prime mover in a high BABIP. In contrast, his rate of Infield Fly Balls, which result from poor contact and almost never fall in for hits, was also better than league average. In fact, using the league-average BABIP rates in the Talking Chop article coupled with this Expected BABIP (xBABIP) equation from FanGraphs, Cabrera’s xBABIP was .314; this means that, given the rates at which Asdrubal did those things which generally lead to a high BABIP (i.e.: hit line drives, avoid hitting IFFBs), Asdrubal’s BABIP would have been .314 with neutral luck. In short, if one subtracts Cabrera’s xBABIP from his BABIP, we find that Asdrubal lost a full 31 points of BABIP from bad luck.

How would that have affected his offensive output in 2013? Given that he struck out or homered in 31% of his at bats, a 31-point BABIP increase would have increased his Batting Average by .020, up to .262. This 20-point increase in Batting Average, in turn, would have effected a 20-point increase in slugging (assuming equal ISO) and a 14-point increase in OBP (assuming unchanged walks and HBPs). The result of this thought experiment is the offensive season he might have had without bad BABIP luck; hence, we end up with our Haphazardly-‘Regressed’ Asdrubal Cabrera Triple-Slash Line Of Science that looks like .262/.313/.422, with an OPS of .735 – an OPS about 20 points higher than league average.

The point of this is not that Asdrubal was above-average offensively in 2013 – as much as one regresses his luck-based numbers, that doesn’t change the fact that in the real world, his offensive numbers were below-average, making Asdrubal little better than a replacement-level player. However, when one projects forward into 2014, there’s no reason to believe he’s going to continue to be unlucky. Hence, counter-intuitive though it may be, the .735 pseudo-regressed OPS above is very  likely more predictive of his 2014 than is his actual, real-world .700 OPS, since said pseudo-regressed OPS makes no unfounded assumptions about future luck. Moreover, a .735 OPS is quite good for a shortstop. Will Asdrubal’s actual offensive performance be exactly .735? Doubtful. However, he’s equally likely to improve on this estimate as he is to decline from it – like most regressed predictions, it’s not a guarantee, just a mid-point.

What that in turn means for 2014 is that, with that sort of offense and a repeat of this year’s defense, Asdrubal Cabrera will be, overall, a below-average shortstop in 2014. On many teams, a ~1.5 WAR shortstop making $10M for one year would be an entirely reasonable allocation of resources; given the precarious position of the Indians’ finances and the possible pursuit of a different 2013 Indians player, the Indians will be forced to gamble – sell low and risk Cabrera returning to 2011 form, or call and risk expedited decline. The correct answer will only present itself in retrospect, yet if this segment of Cabrera’s career is at all like the rest of his genuinely bizarre career arc, even retrospect might not help us.

Stats courtesy of FanGraphs and Talking Chop

About these ads

Author: John Grimm

John can be contacted on Twitter at @JHGrimm, or via e-mail at john.h.grimm@hotmail.com.

9 thoughts on “Asdrubal Cabrera and Inconsequential Bouncebacks

  1. Pingback: Asdrubal Cabrera: Above-Average Players and Contract Extensions | Cleveland Sports Insiders

  2. Pingback: Nick Swisher’s Zemblanitous Season | Cleveland Sports Insiders

  3. I don’t see how supplying a bare list of players with high career O-Swing % from 2002-13 proves your point in any shape or form. Apart from anything else (and there are some pretty mediocre players on that list, as well as the All-Stars), there’s potentially a huge selection bias in your sample. For all we know, many of the players with high O-Swing % were failures as a result and never even made enough PA to meet the qualifications you specify. To put it another way, the so-called ‘unlucky’ high O-Swing % guys who do experience low BABIP often don’t last 1000 PA.

    In any case, with Asdrubal we are actually talking about a player who suddenly experienced a career-high O-Swing % in 2013, not someone who has always had a high O-Swing % and has found a way to work around that and still be successful. I believe that Cabrera’s decline in plate discipline last season contributed to his decline in BABIP, and that it wasn’t just all down to bad luck, as you suggest.

  4. http://www.fangraphs.com/leaders.aspx?pos=all&stats=bat&lg=all&qual=1000&type=c,4,6,11,-1,34,35,40,41,-1,50,61,58,46,102&season=2013&month=0&season1=2002&ind=0&team=0&rost=0&age=0&filter=&players=0&sort=14,d

    Included in the above is an ordered list of the highest O-Swing percentages in the league. I haven’t done any particularly rigorous stat work with it, but it doesn’t seem as though these most liberal swingers as a group deviate wildly from league average, at least as far as BABIP is concerned.

  5. I don’t see how Sullivan article (which relates to pitchers) has any relevance to the theory that poor plate discipline (in terms of swinging too much outside the zone) leads to a reduction in BABIP. It seems to me that you are taking the Sullivan article out of context (see also Voros McCracken’s comment to that article).

    IFFB% may be the best proxy you have for poor hitting, but taking it at face value without mentioning its limitations in this specific context is potentially worse than not using any proxy at all.

  6. To be clear, the IFFB correlation was selected just because of how limited public hitting data is. If I had access to, say, HITf/x or anything else that tracked strength of contract, I certainly would have incorporated, on one hand, weak GB% as a bad-hitting correlate and, on the other, hard-hit GB% as a good-hitting correlate. As it were, the data I personally have access to as a private citizen limits the sophistication of the conclusions one can draw, there’s no doubt about that. I picked IFFB% as a proxy largely because that’s just the best stat out there; that said, BritDawg, if you know of a public resource that tracks strength of GB contact, I’d be greatly indebted if you were to share it.

    However, I’d encourage you to read the following regarding OOZ% and IZ%: http://www.fangraphs.com/blogs/where-balls-in-play-are-allowed-and-what-it-doesnt-mean/

    So poor plate discipline, intuitively, seems like it should be a cause of bad BABIP, but that’s not really a conclusion that’s borne out by the data. In fact, at least according to the Esteemed Jay Sullivizzle Esquire, there’s almost no correlation at all. It feels like there SHOULD be a link, I agree, but one gets into very dangerous territory when one makes adventurous assumptions based on intuition rather than data.

  7. Interesting article ,thanks John.

    You said ‘with career-highs in both his Swinging Strike Rate (9.5%) as well as his Out-of-Zone Swing Percentage (34.4%). Plate discipline, moreover, is not a bounce-back trait, so as of this point, there’s no reason to believe that Cabrera’s 2014 K rate will be much different from 2013.’

    and later on ‘BABIP is mostly a product of four factors: how fast the batter can get to first, how many balls in play were hit well, how many balls in play were hit poorly, and where the defenders were. These four factors are represented by the following four quantities: Speed, Line Drive Percent, Infield Fly Ball Percent, and ‘Luck.’ ‘

    I don’t really understand why you associate hitting balls poorly exclusively with IFFB% Surely hitting balls poorly is also very much about making weak groundball contact, resulting in easy outs and lower BABIP? And weak groundball contact is partly a product of high Out-of-Zone Swing Percentage. In other words, if you swing too much outside the zone the quality of your contact is diminished and your BABIP goes down.

    I don’t see Asdrubal’s low 2013 BABIP as simply bad luck – at least in part it was the inevitable consequence of poor plate discipline. He was swinging too much outside the zone and thus making poor contact. If plate discipline is indeed ‘not a bounce-back trait’ I’m pretty skeptical that Asdrubal’s BABIP will recover as much as you suggest.

  8. Hey, Will, thanks for reading!

    Insofar as Cabrera’s career has phases, yes, Home Run hitting Cabrera is probably the better phase.

    I’m not sure one actually can classify Asdrubal’s career into phases, however. Often, you can do that with players, but despite his power seeing a sustained increase from 2011-13… nothing else has been consistent. Line Drive Rates have fluctuated wildly, contact rates likewise – these are stats that are generally supposed to stay the same unless a hitter’s changed his approach; while Cabrera did supposedly change his approach to be more power-inclusive leading up to 2011, things have changed even in the last two years despite similar results.

    In sum, Asdrubal Cabrera’s entirely mystifying.

  9. John, this was a really nice read. I like your use of numbers. I’m not familiar with “Talking Chop,” but something I’ll have to take a look at. Do you think that this home run hitting Cabrera is the preferred model?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.