Should Neftali Feliz be the Texas Rangers’ Closer?

facebooktwitterreddit

When I started this series of articles two weeks ago, I assumed that Neftali Feliz would become my number one choice. After some research and, more importantly, more news reports on his health and status, I am not so sure.

May 13, 2012; Arlington, TX, USA; Texas Rangers starting pitcher Neftali Feliz (30) throws a pitch during the third inning of the game against the Los Angeles Angels at Rangers Ballpark. Mandatory Credit: Tim Heitman-USA TODAY Sports

It is extremely difficult to discern what Feliz might be in 2014 because not only he missed essentially all of 2013 (4.2 innings), but the brief time he spent pitching in fully healthy in 2012 was as a starter, making trends such as diminished velocity irrelevant.

First of all, let’s gauge how good Feliz has actually been. His career WAR (over 210 innings) would have tied with Yu Darvish’s last season and beaten Derek Holland’s by .2. His career ERA would have been second in the AL last season and 5th overall (between Anibal Sanchez and Zach Greinke). His career xFIP would have tied Kyle Losch–for 59th. Of course, these comparisons involve starting pitchers, who typically post worse inning-by-inning numbers but have more value overall.

So let’s use relief pitcher as well, and start this incredibly simple and unscientific comparison at 2009, Feliz’ rookie season. Feliz’ WAR is 30th, behind Andrew Bailey and ahead of Darren O’Day. His ERA is 18th (between Josh Collmenter (?) and Scott Downs). His xFIP is 77th, behind Josh Collmenter again and ahead of Troy Patton. It would be inaccurate to say that Feliz is a middling relief pitcher, but totally accurate to say he has, so far, beat his peripherals.

So what does that mean, exactly? Some pitchers, like Clayton Kershaw consistently beat their peripherals. (Others, conversely, like Ricky Nolasco, never do). It is too early to discern whether or not Feliz is one of these frankly rare breeds. I venture that he is not. After all, the poster boy for this is the best pitcher in baseball, with two 70 pitches and an 80 (20-80 scale). Feliz has/had an 80 fastball and 60 (plus) slider, but a 50 (average) third pitch. (I believe I said Scheppers may be one of these pitchers as well. In hindsight, I wish I could take that back. I was too rash.)

How much is fair to expect him to regress is really hard to ascertain. ZiPS puts him at a 3.65 ERA and 4.10 FIP. Fans project him at 3.04 and 3.61, respectively. There seem to be two totally different views of Feliz and what he is likely to accomplish. I agree more with ZIPS, but am overall split between the two. His FIP  has gotten progressively worse each season, from 3.21 to 3.50 to 4.28 to 4.98. (This excludes 2013). Some pitchers beat regression, but Feliz probably won’t, at least not entirely. That doesn’t actually address my issues with ZiPS, just my issues with Feliz.

Sep 1, 2013; Arlington, TX, USA; Texas Rangers relief pitcher Neftali Feliz (30)throws during the fifth inning against the Minnesota Twins at Rangers Ballpark in Arlington. Mandatory Credit: Kevin Jairaj-USA TODAY Sports

If Feliz had put up his ZiPS-projected numbers last year, he would have ranked 91st in ERA and 117th in FIP (among relievers) That explains my issues right there. My methodology isn’t fair in that it compares ZiPS ERA to  “real” reliever ERA, and not ZiPS ERA to ZiPS ERA. You can find the entire ZiPS database for 2014 here. As you can see, it doesn’t separate relievers from starters, and there is no way to parse the info to find them (such as setting a innings range of 35-65). So I use 2014 projections with 2013 results.

So while Feliz may not be 91st in ERA if ranked by ZiPS ERA, he would still be rather high, or at least higher than I would project. Someone with his stuff and some semblance of control shouldn’t have an ERA only slightly better than Wesley Wright did in 2013 and an FIP tied with Kevin Gregg in 2013. I would drop both numbers some, but how much I don’t pretend to know. Just know that I think he will beat ZiPS by a legitimate amount, but will still do worse than fan projections.

*****

Everything I have written so far assumes that Feliz is healthy. It is important to remember that he is, as of now, literally throwing 87-91, which is to say more Burke Badenhop (115th in reliever fastball velocity) than Trevor Rosenthall or Jake McGee–or Tanner Scheppers.

And that explains to me why Scheppers, should the Rangers even consider him in the race, seems like the best option. He has better stuff than Soria without Soria’s LOB% dependence. He is currently healthy, with an equally electric fastball to Neftali Feliz in 2010-2011 (his peak). (Seriously, look it up on the profiles I linked to. Scheppers had equal fastball velocity last year to 2010-2011 Feliz with more movement.) Feliz, should he come back to full strength, should still be a good pitcher, but Scheppers at this point is better. He also doesn’t need to rely on coming back to full strength.