Ricky Nolasco and the disappearing curveball

Shortly after I released my first part of looking at Ricky Nolasco, Matthew Carruth at Fangraphs wrote a brief piece about Nolasco’s recently signed extension. The article went on about Nolasco’s Super 2 status and the complications teams have when deciding to buy up remaining arbitration years even with a talented pitcher remaining on board. The article was quite positive and figured this should be a good deal for the Marlins, considering that Nolasco should remain healthy.

What was interesting, however, was the high volume of comments made about Nolasco being far too “unlucky” for too long and that maybe it’s time we re-evaluted a pitcher’s FIP-ERA differential when considering his value. This is an argument primarily spurned on by those who fundamentally disagree with how one measures a player’s WAR value. Since I have no desire to firmly take up arms for one side over the other, I will note that Nolasco’s curious streak of ERA underperformance has occurred in just 342.2 innings.

I know this may seem like a lifetime, but in the field of statistical baseball analysis, that still qualifies as a small sample… and to further compound how small this portion is, take into account an article published before the 2010 season which argued that if we omitted three bad starts from Nolasco’s 2009 season: his stat line for that year would look remarkably different.

Before we go further, let’s see where Nolasco ranks among other active starters from 2008 to present:

{exp:list_maker}Ranks 25th in terms of FIP (3.66)
Ranks 25th in WAR (10.6)
Ranks 12th in K/9 (8.63)
Ranks sixth in BB/9 (1.93)
Ranks sixth in K/BB (4.47){/exp:list_maker}

Those rankings are based on Fangraphs leaders database and were based upon a starting pitcher appearing in a minimum of 500 innings from 2008 to 2010. Also, it must be said that not all things have been rosy for Mr. Nolasco.

{exp:list_maker}Tied 15th in HR/9 (1.23)
Tied 12th highest BABIP (.315)
Tied eighthth in total home runs allowed (75){/exp:list_maker}

Essentially, it’s been a combination of lousy overall BABIP along with a healthy case of balls flying over the fence that’s been troubling our subject. In my last part, I charted the percentage of pitches Nolasco has thrown since 2008. Here is a chart provided by Joe Lefkowitz detailing the movement on his pitches in 2008:

According to that, Nolasco threw a healthy mix of fastballs, curves and sliders with the occasional change. In comparison to his 2009 and 2010 season:

Looking at these side by side and in comparison to the 2008 data, we see that ‘09 and ‘10 were relatively similar in terms of less usage of his fastball along with a drastic decrease in his curveball.

Commenting on my last piece, one reader, along with our own Josh Smolow, wisely brought up the problems with PITCHf/x camera calibrations through MLB Advanced Media. Looking at the 2010 data we see the introduction of a cut fastball (FC) along with the appearance of a split-finger fastball (FS). It’s entirely possible that these pitches have always been in Nolasco’s arsenal but with corrections made in 2010, PITCHf/x may be recognizing them now.

It’s a point open for debate and one that may never be fully settled, but it seems certain that Nolasco is using his curveball less and less.

In 2008, Nolasco threw his curve 791 times. In 2009 that number decreased to 480 and, finally, 389 times in 2010.

Ricky Nolasco’s curveball data 2008-2010

The data above suggest that the curveball is a valuable pitch and one that Nolasco throws to both right-handed and left-handed hitters. According to its Pitch Type Value chart, Fangraphs has rated his curveball well and classified it as his best pitch in 2008.

Obviously, one has to be selective when throwing the curve. In situations with runners on or in some hitter’s count (especially in situations of 2-0 and 3-0) it probably wouldn’t be wise to pull out the curve. Looking over the above graph I was reminded of an article written by Josh Kalk which examined the problems when professionals hitter recognize a released curveball.

The view provided above charts the trajectory of Nolasco’s 2010 pitches from a side vantage point. Obviously, the high arching curveball stands out and if you haven’t clicked the above link to Kalk’s piece, I suggest you do so. It has a very nice graphic showing the “batter’s view” of a high approaching curve in comparison to a standard fastball.

Looking over his opponents’ swing percentage and seeing how low it is compared to his fastball and slider (both average over 50 percent), this is probably not the pitch Nolasco will turn to if he needs to fool a batter or induce a groundball swing in a 3-1 count.

In 2009 Nolasco fell into a lot of so-called hitter’s counts (2-0, 3-0, 3-1 and 3-2 are considered the standard) as you can see in this chart:

Before we sit amazed at the reduced 2010 numbers, it must be said that Nolasco made only 26 starts before sustaining a knee injury in late August. However, if you jog the numbers a bit, you will see that he probably would have been closer to 2008 numbers in terms of overall hitter count situations.

It seems that when any extended talk centers on Nolasco’s troubles, we are reminded of his problems handling hitter’s counts and pitching with runners on:

Of course one also has to look at the role of BABIP in these situations:

Taking BABIP into consideration, it’s also handy to look at the UZR scores of both the infield and outfield of the Marlins:

It’s no secret that the Marlins defense ranks near the bottom half over the past three seasons, but does a less than optimal infield defense really matter for a pitcher with a career 38.9 percent groundball rate? It could in certain situations, but until we gather more data, things will seem incomplete and it still may take more than 342 innings to see if we are dealing with luck or an obvious lack of skill.