The Enigmatically Productive Wei-Yin Chen

Coming in to the 2013 season, one of the perceived flaws of the Baltimore Orioles was the quality of their starting pitching. Many saw the rotation as one full of guys who realistically were best suited for the back end of a rotation, with no real ace or top flight starter who would carry the load. Chris Tillman has appeared to step up and take on this role of rotation ace and has had an admirable and productive year, whether your statistics of choosing to measure this is wins or something more sabermetric in origin.

Beyond Tillman, the rotation still remains a touch murky, with Jason Hammel still not performing to what he is capable of and the likes of Miguel Gonzalez and newcomer Scott Feldman pitching admirably, albeit inconsistently. Overall, the O's bring Tillman and a handful of guys hovering around the .500 mark in terms of their win-loss record. Included in this bunch is lefthander Wei-Yin Chen, whose 2013, while interrupted by an oblique injury, has been befuddling. Currently sporting a 7-7 record and tied for the team lead in pitching Wins Above Replacement (with Gonzalez) at 1.6, Chen's season for the most part has been one lost in the shuffle.

As of late, Chen's outings have been lackluster -- his August 27 outing versus the Boston Red Sox generated a -0.434 win probability added (WPA), while his most recent start, a no decision against the New York Yankees on September 1, again saw him provide negative WPA (-0.128). Here is a breakdown of Chen's last two starts, versus his first 16:

Starts	K/9	BB/9	HR/9	BABIP	GB%	FIP	xFIP
First 16	5.71	2.37	0.9	0.271	35	3.88	4.34
Last 2	11.76	9.31	2.46	0.455	18	7.49	5.79

...and the same stats, this time broken down by decision:

Decisions	K/9	BB/9	HR/9	BABIP	GB%	FIP	xFIP
W	5.48	2.86	0.37	0.238	35.00%	3.38	4.57
L	7.49	3.6	2.03	0.317	25.56%	5.63	4.88
ND	6.03	2.81	0.61	0.340	41.23%	3.53	3.72

In a world that is dominated by the sentiment 'what have you done for me lately?', Chen has done little to pick up the slack seen in non-Tillman Baltimore starts. Yet, with the help of these two tables, we do see some encouraging signs. In general, when Chen keeps the ball down and minimizes walks, he is successful. Not a huge breakthrough in terms of how to approach a lineup, but one that Chen still seems to have departed from as of late. Yet, we also see some enigmatic things with respect to Chen, beyond his pitch to contact approach, in spite of having one of the better fastballs velocity-wise amongst starters, per Fangraphs' Pitch Type statistic. 

Aside from his velocity and approach, the biggest enigma thus far is Chen's strikeout rate in losses -- it's higher than in wins and no decisions (ND). One would think that a pitcher generating outs and minimizing runners on his own merits would be a good thing, and it is -- when you keep the ball in the park, which he hasn't done in losses (2.03 HR/9). In general, when Chen gets wild and is unable to outpitch his peripheral stats (when FIP < xFIP), he gets in trouble.

Let's dig a little deeper into Chen's last two starts, with the help of Brooks Baseball. Let's take a look at Chen's last two outings as well as one that closely mimics his 2013 averages in strikeouts and walks per nine innings and is also his best outing (per WPA) since his return from the disabled list -- Chen's outing on July 10th versus the Texas Rangers. I use this outing in order to compare/contrast his less than stellar outings as of late, to see what he might be doing differently that could explain some of his subpar production the last couple of weeks.

First, let's look at Chen's pitch selection and breakdown:

7/10:

8/27:

and 9/1:

Here, the following graphs look at Chen's aforementioned outings, looking at pitches (and their speed) across inning:

7/10:

8/27:

9/1:

To summarize the Brooks Baseball and PITCHf/x charts, Chen is throwing his offspeed pitches less frequently as of late, compared to his successful outing on 7/10 and is also throwing said offspeed pitches with less difference in velocity between his fastball. If you look at the bottom three charts, the top 'row' of dots are fastballs and sinkers, the second 'row' are his sliders and changeups, and the bottom 'row' are curveballs. In general, his last two outings show a less distinct gap between pitch types, as well as fewer dots in the bottom two 'rows'. While he is throwing his fastball at a slightly higher velocity on average as of late, he is not getting enough of a change of speed on his offspeed pitches to fool hitters. Add to this his curveball not being a particularly sharp pitch, judging not only by the spread on the PITCHf/x charts, but also their linear weights, and you get the ineffective Chen of the last two starts.

Overall, Chen's season has been productive, sometimes in spite of his best efforts. With this Saturday's start against the White Sox, using the past our guide to predict future outings, Chen should not only look to create a greater difference in velocity between his fastball and breaking/offspeed offerings, but also look to rely more upon his slider and changeup more so than his curveball to keep hitters honest against his fastball and to keep batters guessing. In doing so, he has the potential to not only salvage an enigmatic season, but also give the Orioles a solid 1-2, righty-lefty combination that could keep the team in the thick of the playoff hunt.