Week Two: The Bullpen Strikes Out

OK, that's a bad title.  But I couldn't get it out of my head.

[NOTE: insert gratuitous pic of Kylie Minogue's backside here.  We need *something* to distract us from this mess...]

Following Andrew Cashner's not-stellar debut, the story, either pushed by the Rangers or copied by the beat writers from each other, was that despite the outcome, the Rangers were pretty happy with Cashner.  And they were right to be.  While it's still a Small Sample Size (of one, sort of) a pitching performance can say a lot more than one at bat, or even a game of at bats.  I'm not gonna say that Cashner's performance had a real gem hidden inside; it didn't.  But it should have been worse than it was.  Cashner himself made a lot of good pitches, although his future success is going to largely depend on getting his strikeouts up.

Cole Hamels.  Geez, I don't want to have another "I told you so" year.  Having three pitchers that can't make it through six stresses the bullpen.  That's beyond the problem of having a clear #1 pitcher, a couple of MOTR guys, and a couple of back-enders.  But that could be worse.  A lot of teams don't even have that.

The bullpen is what has everyone worried, but there are several factors at play.  One is the apparent organizational philosophy of multi-inning relief efforts.  Except when you let some of your more capable pitchers only face one batter.  I...I just don't get it.  The multi-inning thing is real; they've done it over and over the last two years.  I didn't imagine it.  But if that's a philosophy thing, and not just a knee-jerk decision by *somebody*...then why do you bring your sharper guys in for only one batter?  Last year I surmised that Banister was going to use the first six weeks of the year to test his bullpen and find out what people could do.  I think he sort-of did that?  Maybe?  Except he kept doing the multi-inning thing a lot...but without any rhyme or reason that I could discern.  I really hope we find out what's going on, someday.

I didn't create this image.

I didn't create this image.

Once again, I got some people upset when I mentioned Banister's history of over-using arms.  The thing is, even when there is disagreement about the statement, I suspect the difference is actually over whether Banister is *right* to overuse arms...not that he *does* it.  Or to put it another way, that Banister does what is necessary to win ball games.  Banister had a bad bullpen to start the year last year, and he overused the arms that worked.  BUT, I would still make the point that he *continued* to overuse several relievers even when new bullpen pieces were provided.  In fact, several times over the past couple of years fresh arms were shipped to the dugout after some heavy usage...and Banister refused to play them.  A couple of these guys have since been traded away, both in fairly surprising moves.  My own supposition is that Banister wasn't going to use them, so Daniels traded them.  (I'm *guessing* about this, not making an accusation or anything like that.)

But anyway...let's throw the stuff we can argue about to the side, and look at how the actual relievers are performing.

Sam Dyson:  We're not talking about Sam Dyson.  He'll either be DL'd, at which point he will likely be done as Ranger reliever, or he gets one more shot or two BEFORE being DL'd, at which point he will likely be done as a Ranger reliever.  When Banister doesn't trust you, you will not be played.  Healthy Dyson will be traded or released.  NOTE: a few hours after publishing this piece, Dyson was sent to the Disabled List with a very plausible hand contusion. The severity of the bruise is unknown, as is the projected length of the DL stay. I suspect recovery from the bruise will require some rehab games, at least.

Mike Hauschild:  This one might surprise you.  The Rule 5 pick has to stay on the roster all season or be shipped back to Houston.  Here's what he's done, including the bad stuff:  ERA 9.64 FIP 7.65 xFIP 3.45.  That xFIP isn't bad, and indicates maybe he's had some bad luck.  Still, you would like a reliever closer to 3.2 and down, rather than 3.5 and up.  Batting average on balls in play:  .385.  That *is* some bad luck.  9.64 K/9:  That's pretty darn good.  1.93 BB/9:  That's decent, too.  A Home Run allowed every other inning:  that's not good.  But, I hope you can see why Hauschild will likely stick around...unless Banny stops using him.  Incidentally, Hauschild was a star pitcher on the Fresno Grizzlies team that beat Round Rock for the PCL championship in 2015.

Paw bumps!

Paw bumps!



Jeremy Jeffress:  His K's are way down, but so are his walks and BABIP.  FIP of 2.41 and xFIP of 3.45.  Really good ground ball percentage.  I think he'll be ok.  Maybe he'll ramp up the K's as the season progresses.

Alex Claudio's FIP and xFIP and complete lack of strikeouts indicate someone who isn't pitching as well as he appears to be.  A BABIP of .154 supports that conclusion.  Regression should still leave him as a really good long-man/middle reliever.

Dario Alvarez is the fill-in lefty specialist, and is probably best used in this role until he can be sent down to get some coaching at AAA.  He's likely to strike people out, but when he doesn't it's a walk or a home run.  A three true outcome pitcher.

The velocity is nice on Matt Bush, and I hope his brief DL stay actually helps, because he's not pitching as well as he did last year.  And because of the shoulder problems, I'm wary of predicting his upcoming performance, and a bit nervous of putting him in the closer role.

Jose Leclerc is actually pitching the way you usually want a closer to pitch.  I understand the arguments against making him a closer, but I think he's already shown he can just get up and pitch and stay within himself.  By far the best reliever in the bullpen.

You guys are gonna get sick of hearing it, but Tony Barnette is pitching a lot better than he looks.  Going into tomorrow his FIP will be about 3.5, with an xFIP around 3.  K/9 almost 1 per inning, no walks.  BABIP was .267; now .333 after Easter Sunday.

Diekman isn't going to be back soon.  He just had a new colon built this past weekend; or part of it.  I think he's still got six weeks recuperation,  then strength training, then some kind of spring training, then minor league games...I'll be amazed if he pitches for the big club this year.  I know that's not what the Rangers say, but that's what my math says.

Kela is in AAA for an alleged "bad behavior" incident in the clubhouse.  I know Kela is known for that, but clubhouse misbehavior bothers me.  It could mean either Kela is beyond regular control or the management group wasn't on top of things.  And if Kela and Banister have issues...then bringing Kela up from AAA isn't going to make much of a difference, because Banny won't use him.  NOTE:  A few hours after publishing, Kela was recalled when Dyson was DL'd, and was used to close out a game the same day.  Make of that what you will.

I know it seems like I'm closing out with doom and gloom here, but I really do think most of the guys up there are able to be included in a bullpen that can go three innings of one run baseball.  Surely we all want three innings of no-run ball, but that's not very realistic.  I am, despite all the reports about the starting rotation's ERA, more concerned about the starting rotation going forward than I am the bullpen.  Mainly because I know Daniels can keep finding sacrificial arms for the bullpen, regardless of the bullpen management problems.  Finding five guys who can keep five runs from crossing the plate is a different matter.  Texas is supposed to have three starting pitchers who can do that reliably, but Perez still has some history to overcome; and Hamels is.......whatever Hamels is now.  Both Griffin and Cashner have shown signs that they can do the job, but neither is a lock, either.

As far as the offense goes....I haven't been paying attention, honestly.  The Rangers seem to either score lots of runs or get shut out.  I don't know what that means.

Whither the Rangers in 2017? Insights on Hope and Despair Part One

Is it Darth Vader?  Or is it all in your mind?  Yes.

I had three articles in the planning stages, and I decided to generally combine them into one.  This one.  But then I didn't.  This is the one where I talk about Derek Holland, and the implications of his departure on the starting rotation.

Derek Holland wasn't very good in 2016.  Actually, there wasn't a lot about Derek Holland's 2015 to offer any reason to expect better in 2016.  And Derek Holland's 2014 wasn't anything to write home about, either.  And come to think of it, 2013...no wait.  He was good in 2013.  In fact, 2013 was the best Major League season Holland ever turned in.  After a solid 2011 in which Derek Holland showed not only a pretty good year for a Middle-of-the-Rotation lefty, he showed the promise of being *even better*, followed by a full season in 2012 that looked a lot more like a decent season for a back-end, innings-eating guy.  Then the really good 2013 that was an improvement of 2011 all the way around.

And then began the injuries.  In 2014, Dutch pitched 37 innings for the Rangers.  His results were *fabulous*, although his xFIP (eXpected Fielding-Independent Pitching, a statistic that produces an ERA-like number based on normalizing a pitcher's fly ball and home run results, as well as applying a leavening factor to strikeouts and walks.  This will, with effective pitchers, usually show an ERA-like number better than their actual ERA.  Less-effective pitchers will frequently sport a higher xFIP than ERA.  These are generalizations, not absolutes) indicated a performance similar to 2013 and 2011 rather than the giant improvement his 1.46 ERA would suggest.

In 2015, Holland was injured again...this time his return to the Bigs was disastrous.  Over 58.2 innings, he produced numbers very similar to 2009; his first appearance with Texas.  The good news was that his xFIP *this* time indicated that he should return to at least being that serviceable, back-of-the-rotation pitcher that he had looked like at his *worst*.  Because of that, it was widely assumed that the first of his two contract options would be picked up by the Rangers at the end of 2016.  The prices for back-end pitchers had started to skyrocket, and Holland's $11 million price tag was a bargain.

And THEN 2016 happened.  In some regards, Dutch *technically* improved his performance in some areas.  Very, very slightly.  But the important number was:  his strikeout percentage was WAY down.  Derek wasn't throwing nearly as hard as he used to.  Despite similar Batting Average on Balls in Play to his career numbers; despite a similar walk rate; a similar Ground Balls Induced percentage; and a similar Home Runs allowed percentage, his ERA and FIP (Fielding-Independent Pitching, like xFIP except it doesn't normalize home runs allowed) were still around 5.00.  Worse, his xFIP was 5.14, almost a full point higher than 2015.

The stinker is, there is very little evidence of a "smoking gun" statistic.  His numbers were all in line with recent years (keeping in mind those were mostly bad years).  Where you see it all fall together is in comparing 2016 with 2011 - 2013.  Everything was just a little bit worse, and almost everything WAS worse.  Just a little bit, mostly.

When you combine pitching just a little bit worse with a fastball that just wasn't quite good enough to get people out anymore, you wind up with a guy who was actually LUCKY he didn't do worse in 2016.

And *that* is what I was referencing when I kept mentioning during the year that Holland wasn't actually pitching very well, even though he got good results in several games.  However, even then I still expected Texas to pick up his contract option, considering how expensive Middle Rotation and Back End pitchers have become.  It genuinely surprised me that they didn't.

There are possible reasons, of course.  Texas may know, or at least suspect, that Holland has serious or long-term damage in his shoulder or some other important body part.  It could simply be that the right people see the negative performance trend and believe (probably rightly) that pitching at that presumed level could be filled by cheap, internal options.

In conjunction with that is the possibility that Texas really is committed to improving the rotation.  As much as you can blame the Toronto series that ended 2016 on very, very, bad officiating; you also can not escape the fact that the starting rotation wasn't as good as it was expected to be.  In particular, Cole Hamels wasn't very good in the second half of 2016.

The bottom line is:  while everyone keeps saying "you don't worry about Cole Hamels", I think we all need to consider how often we heard that line repeated in a reassuring tone by many, many bloggers and analysts last year.  Hamels' 2016 was his worst year since 2006, his first season.  The big culprit was a huge increase in walks; all year batters sat back and made him make his pitch, and he didn't with the same consistency as previous years.  Considering Martin Perez *also* threw up career worsts in strikeouts and walks; this isn't a questionable rotation after Cole, Yu and maybe Martin...it's a questionable rotation after Yu.  And Yu has not been extended past this next year.  And may not be, considering he can expect a minimum contract AAV north of $20 million and will almost certainly at least be talking about $25 million for five or more years.

THAT is why people are making "absurd" calls to trade Yu now.  Looking *only* at the rotation, it's not an absurd notion.

But the Rangers, like every other baseball team, is comprised of not only pitching but also fielding and hitting.  And the other parts look pretty good.

More Trend Analysis: Pitcher xFIP

More trend lines!  xFIP, or eXpected Fielding Independent Pitching, is intended to not only take team factors out of a pitcher's performance (FIP), but also "normalize" home run outcomes by applying a constant to Fly Balls Allowed based on averages.  The stat has a pretty good history of telling you the true talent level of a pitcher once you take most of the variables out.  It also, therefore, does a good job of telling you when a pitcher is legitimately pitching worse or better.

xFIP works like ERA; high is bad.  The trend line follows suit; high side on the right is trending up which is bad, low side on the right is trending down is good.  A flat line indicates stability and is good, as long as the number it represents is good (like, under 4 is groovy).

Cole is, until we have a better understanding of how Darvish is going to play out, the most important member of the roster.  Hamels has been steady, and is legitimately pitching slightly worse this year...but not by much.  He's been pretty solidly a 3.5 xFIP guy, and as you can see, Cole is sitting around 3.7/3.8 this year.  The trend is good though, so there's every reason to believe he'll finish the year in fine form.

Speaking of Darvish, there obviously isn't a trend line, but his one game is one of the best starts by a Rangers pitcher this year.  Optimism!

How 'bout Martin Perez?  The trend is right, but that's as much because of a bad start.  It wouldn't surprise me to see Martin stabilize right around 4.00; anything under that would thrill me.  And take note that would put him on par with Hamels.

Solid as a rock.  In fact, he was trending down to 4.00 like Perez before his last bad start.  This is probably what Colby looks like all year, and once again puts him in that "#4 starter" territory.

Oh, dear.  The good news is, without the one game, the trend isn't quite as bad.  But it's still bad.  Pitchers have managed to outpitch their xFIP for entire seasons.  It's not unusual.  But it's not the norm, either.  If/when Holland blows up, it shouldn't be unexpected.

As with Delino Deshields, we must ask ourselves:  "Was Shawn Tolleson actually performing that poorly?"  As you can see, the bad appearances were getting worse and worse, and he only had a few good appearances.  More importantly to Jeff Banister, I suspect, is that Tolleson wasn't predictable.

That's what you like seeing in a closer.  At least the recent results.  Since being made the closer, Dyson has been exceptional.  Remember that big uptick at the end is the average, which I include to "normalize" the line a bit.

Matt Bush's trendline is good, but I'm not too thrilled that it's centered on 4.00.  And of course, as has been a frequent topic of discussion, there are other factors to consider with Bush's quality of pitching. 

Jake has been very off-and-on this year which makes his 2.5 xFIP stable trendline somewhat misleading.  He's still the second-best reliever behind Dyson.

Here's Tony Barnette with two really bad appearances.

Here's Tony Barnette without the two disaster appearances.  More stability would be nice, but trending from 4.00 to 3.00 is a decent indicator that he can keep filling in reliably.

This is both starting and relieving.  It's about as good as it gets in a spot-starter/long-reliever.

I almost forgot about Alex Claudio, and it looks like that would have been a mistake.  Not only is better at his worst than some of the other relievers, he's steadily improved.  Again, remember that last big uptick is the average.  I would bank on Claudio getting a lot more time, soon.

There is still a good core group of relievers here.  Looking at the performance histories and trends, I'm seeing consistency as a bigger problem than actual ability.  Consistency can be a function of the pitcher of course, but it can also be a function of the catcher; a position that has been highly unstable for Texas this year.  It can also be a problem with game planning, and Banister is working with two new coaches this year as well.

Taking all of that in, I'm still a lot more worried about injuries and arm fatigue than I am who's on the roster.  As the roster itself stabilizes, I think the relief corps will, as well.