The First 10,000: A Review of the Rays Bullpen Usage

The cat is starting to get out of the bag as more and more folks realize the Rays just might be cooking with gasoline. This past day saw good reads from Jeff Sullivan giving a very nice overview of what has gone wrong and how that hasn’t really mattered. He astutely noticed that it’s no coincidence that the Rays starters have a nice, shiny ERA, but also lead the league in fewest innings. That means the bullpen has needed to pick up the slack. USA Today’s For The Win blog picked up the torch with a nice little unearthed gem:

And Cash has managed to limit his starters and get plenty of effective innings out of his bullpen without abusing his relievers. Through Tuesday night’s play, Major League pitchers have worked on at least three straight nights a total of 183 times in 2015. The Rays are only responsible for six of those instances — ever so slightly less than their share — and only lefty specialist Xavier Cedeno has done it twice — once for three straight games, once for four.

While the fact itself is incredible I don’t think it goes far enough to show what an incredible job the Rays have done stealing guys some rest as if it were Hell Week. These two fine organizations have started down the rabbit hole and while their torch has dimmed I’d love to light my own and walk around for a while. Maybe see how deep this chamber goes and hope to pan a few gems of our own. Come with me if you seek adventure.

The tricky thing with research like this is that it’s inevitably going to get cluttered. It’s very difficult to show the work of 25 pitchers (all numbers throughout are for 70 games) who saw varying levels of work with several walking out of the kitchen through that swinging door and then back in as roles so far have been mostly amorphous. To that end I’ve prepared a few different ways to catalogue how the Rays pitching has been deployed thus far. Emphasis will be on the pen, but it is unavoidable to mention the staff so they’ll get their time too.

First off, let’s look at just relievers used for all teams through the Rays 70th game. As mentioned by For The Win the Rays lead the league in relief innings by nature of their starters leading the league in fewest innings pitched. By extension, the implication is that the Rays have burnt out their pen and it’s only a matter of time before these guys start to wilt as if they were a not-s0-freshly picked petunia. Watching the Rays ride the shuttle all year and constantly bringing in fresh arms that can hold their own in one-to-four innings stints it’s hard not to have a high level of admiration for how the team is spacing all these innings out. The most important horses are working a lot, but they’re not frothing at the mouth and wild at the eyes.

My hope is that this isn’t all that confusing, and to prove it I’m goinRays Peng to give you this image that looks at the relief pitches for all Rays pitchers. I’ve listed the raw number of pitches for each guy, but I’ve also ran a standardized z-score for each guy across the other 420 guys that have attempted to spell relief this year and the nearly 103,000 pitches that they’ve cumulatively pitched. For the initiate, a standardized score of zero would imply that that guy has pitched the average amount of innings this year. Anything in the positive tells us that that guy has pitched more than the average and the more positive the further he is from the norm. Same in the other direction for the negative values. In this example Kevin Jepsen leads the Rays relievers in pitches thrown and is around 1.7 standard deviations from the norm. He checks in well below Justin De Fratus who leads all MLB pitchers in relief pitches thrown with 654, which equates to 2.4 standard deviations.

Of these 21 guys you can see that six have thrown an above-average amount of pitches with three of them more than one deviation from the norm. I should note here that Geltz’s pitches as a “starter” are not included so you can add another 50 there to put him just north of Boxxxy if you’re a stickler. We see a lot of guys helping out and a few guys shouldering the load more than the others. To that end I think it would be beneficial to just focus on the three guys that have been worked the hardest to see where the Rays rank:

Top 3

Again, if you want to add in Geltzy’s innings as a starter then you want to bump us up to eighth to basically be tied with the Reds instead of actually tied with the Indians for ninth. No surprise that the Phillies lead the league in over-working their guys (De Fratus, Luis Garcia and Jeanmar Gomez), but we also see some other teams in there that would call themselves contenders. The Rangers are killing three of their guys (Anthony Bass, Keone Kela and Shawn Tolleson) with the latter two representing really nice relief options. The Yankees (Esmil Rogers, Dellin Betances and Chasen Shreve) aren’t all that far behind and neither are the Cardinals (Kevin Siegrist, Trevor Rosenthal and Matt Belisle). These are teams that fancy themselves a contender and the best team in baseball, respectively. It seems odd that folks would decry that the Rays pen is ready to fall apart, yet I hear no mention of these teams over-working their own very good pens.

To take this literally one step further and to make sure that this wasn’t just luck that I decided to go three deep let’s take a look at the fourth most worked reliever for each team:

Top 4

Again, we see that the Rays check in ninth, which could just as easily be eighth, but no where near numero uno. Here we bring in Brandon Gomes and while we still see the Rangers (Tanner Scheppers), Yankees (Andrew Miller), and Cardinals (Carlos Villanueva) ranking highly I notice another team creeping up that’s rather interesting. The Pirates have used the fewest relievers in baseball this year with only nine gentleman throwing a pitch in relief for them. As you can imagine they’re working those nine dudes pretty hard. Tony Watson leads the pack, but he’s joined by Arquimedes Caminero, Mark Melancon, Jared Hughes and Rob Scahill as guys with a zScore at 1.0 or above. Among contending teams only the Yankees can boast a fifth guy at that level leading me to believe they’re another team where the bullpen could probably stand to aerate.

Now that we’ve seen that the Rays bullpen workload is probably a bit overstated, let’s delve into a couple of ways of visualizing what they’ve done so far. Again, this is a delicate balance between providing information that’s intuitive, but also not so cluttered that it’s impossible to read:


From left to right we have every pitch thrown by the Rays this year and working north I’ve attributed them to each pitcher. You can tell the starters as there is a more consistent density to their pitch groupings with even spacing in between for the most part. The relievers get a little more tricky, but I think this does a good job of showing when and where the Rays sought their production. An interesting note from this is the reminder of just how very long this season is. It’s easy to forget some of the truly bad performances that we received from Frieri and Balfour, amongst others. These guys are long gone, but their numbers are going to still skew the team totals somewhat.

I have color-coded each 1,000 pitches just to give an idea of spacing. For instance, over the first 1,000 we see that Jepsen, Geltz, Boxberger, Andriese, Frieri, Yates, Balfour, and Beliveau saw work and all of those guys stuck around for the next thousand pitches, but the third sees some shake up as Beliveau, Yates and Riefernhauser hit the DL and Balfour hit the glue factory. I think you can do your own perusing from here, but please hit me up on Twitter (@sandykazmir) if you have any questions. Now I want to move on to another way to look at this:


I would love to have this all on one graph, but it would be impossible to have any information conveyed so I’m going to split the 70 games into five 14-game segments. In this first one you can see how many pitches each guy threw on each day. There’s a lot here and I wanted to mention that these are not in order of appearance, but this does include all pitchers on a given day. The starters are fairly obvious, but I think this works best for relievers to isolate a guy and then check out his daily usage. For instance, Geltz pitched three of the first five games, but he did get a two-day rest period in the middle and then a three-day layoff before his next showing. Let’s move on to the second segment:


Here we see Smyly make an appearance or three with Archer, Karns, Colome and Odorizzi also making starts as Erasmo moved to the pen. There’s so much here that I hope you can dig in, but I’m going to move on to the third segment:


I like what we did on 5/14 where Erasmo and Andriese combined to throw a similar number of pitches and give everyone else a day off. You can see Bellatti’s effect as he was a guy that could throw quite a few pitches in an appearance. It’d be nice to get him back. I could go through this all day so let’s go to the 4th segment:


The game on 5/26 had a real chance to muck up the week, but Archer and Boxxxy were able to give everyone else the day off the following game and then the whole team got a day off. Stuff like that has helped immensely and is a driving force that allows the team to do what they’re doing. We see a liberal dose of Cedeno in here and Boxberger through a ton of pitches from 5/26 to 5/31 even with the two-day break in the middle. Erasmo and Karns aren’t throwing a ton of pitches, but their contributions were noteworthy. Let’s get to the last segment:


After using 14 or 15 pitchers across the previous segments we step it up to 17 here, though two of those were position players when Nick Franklin and Jake Elmore were called to duty. I’m surprised that the position player game wasn’t even the most pitches in a game , but obviously we needed to bite the bullet as the next few days still required a number of guys. The most recent games see a lot of guys getting used so this is starting to feel like a train that is picking up steam. Will it run off the rails? I have no idea. This is a bit unprecedented so I don’t think anyone can tell you if the Rays can continue to have it work, but getting Moore and Odorizzi back in the next couple of weeks should hopefully make it a moot point, somewhat.

I’d love feedback on this or any of my work so don’t be afraid to shoot me a tweet or leave something in the comments. Take care of yourselves, you’re the only one that will.


About Jason Hanselman

Rays fan.
This entry was posted in Uncategorized. Bookmark the permalink.

2 Responses to The First 10,000: A Review of the Rays Bullpen Usage

  1. Pingback: Rays’ Historically Heavy Bullpen Usage Key to Relevance | FanGraphs Baseball

  2. Pingback: Further Exploring the Bullpen Usage of the Rays in 2015 |

Leave a Reply

Please log in using one of these methods to post your comment: Logo

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s