It’s been kind of a weird season to be a Tigers fan for a few reasons. It hasn’t been weird that the bullpen has struggled but it was pretty weird that the biggest question mark of the bunch was the steadiest arm for the first few months of the season. Joba Chamberlain was risk, a fine one, but a risk nonetheless. And for a while, it looked like Dombrowski had cooked up another pot of magic. And then, things went south.

Let’s dig into the first and second half splits for Joba. The All-Star Break is arbitrary. But it’s all arbitrary. I could split it from August 1 or July 25. It doesn’t really matter.

Joba’s strikeout rate is down, the walks are up, the home runs are up, and the BABIP isn’t much higher. Those are bad signs. Granted, it’s a very small sample and we can’t make too much of it, but all of the indicators are trending in the wrong direction.

Let’s check the results:

Nothing surprising here, either, but plenty to make you concerned. This isn’t simply a little batted ball luck or a poorly timed home run. Joba’s peripherals are bad and so are his results. What’s going on here?

There are essentially three possible options. Bad luck, regression, or some sort of change in his process and talent. The bad luck doesn’t hold water because his K/BB/HR numbers are all moving negatively and his BABIP isn’t. That’s not what bad luck looks like.

So we’re really asking ourselves if Joba got demonstrably worse or if this is simple a matter of regression and random variation? Was he just over-performing during the first half, leaving this as a market correction? Or did he forget how to pitch?

Let’s start with a peak at his preseason projections (Steamer). Granted, projections for relievers (especially ones with injury histories) are to be taken with a grain of salt. But let’s look anyway:

This is pretty interesting. Steamer (which thought more of him than over projections), actually expected him to be worse than he’s been to date. He’s getting more strikeouts, limiting home runs better, and running lower ERA and FIP even after the post break swoon. You could look at this table and assume that Joba’s struggles are just things evening out. You might be right, but let’s look a little bit more closely to be sure.

I didn’t find much in his pitch usage or velocity, but let’s look a little bit at location.

He seems to be having more trouble filling up the strike zone since the break, and that’s not a surprise given the walk rate.

Batters are going after his strikes more often and they’re connecting. He’s also throwing fewer strikes and fewer early strikes in the second half. It’s hard to say if this lack of command is just a bit of noise or if it’s deteriorating over time, but it’s certainly not helping his case.

My bet is that Joba pitched above his talent level during the first half and is pitching a little below his talent level right now. I’m not worked up about the fact that he’s getting worse because I think it’s probably safe to say he’s a high 3’s ERA and FIP guy. Relievers like that are going to go through good times and bad. We have enough information to know he’s not that lights out relief ace he was in the early goings, but I also think we have enough information to suggest that he’s not totally useless.

This isn’t an encouraging finding, but it’s probably also not doomsday. Joba is an okay reliever. That’s who the Tigers signed. They just happened to get a couple of great months out of him before he came back to Earth.

Pitch calling is one of the ultimate black boxes in baseball. Not only is it difficult to discern which pitches are called and in which locations with perfect accuracy, it’s difficult to know who called the pitch, period. The catcher put down the sign, but the pitcher has final say, the coaches could have signaled the choice to the catcher, and the entire group could have outlined a plan of attack before taking the field. Additionally, pitch calling is so dependent on so many contextual factors that it’s hard to tell what the “right” pitch to call would have been. In other words, game calling is very hard to evaluate.

This has always been an interest of mine but after recent comments from Ausmus suggesting that James McCann can’t catch during meaningful innings because he doesn’t know the pitching staff or opposing hitters seems to indicate that Ausmus thinks McCann would call the wrong pitches.

The reason I say this is because 1) Ausmus is specifically worried about McCann on defense and 2) I haven’t heard any concerns about McCann’s actual ability to receive, block, or throw. It’s about working with the pitchers and that indicates that he’s worried about how McCann would call a game given his lack of familiarity.

I think this is silly. McCann is more than capable of doing his homework just like anyone else and he actually does know about half the guys on the staff from time in Toledo. Also, you could just call pitches from the bench if you are so worried about him. Let’s leave that aside and look at something related though. How do Avila and Holaday call a game differently?

We’re just going to use 2014 data here, first from Baseball Savant. There are occasionally PITCHf/x classification issues, but they should be uniform and random errors. Here is the basic breakdown, with the knowledge that they have not caught the same pitchers with the same frequency (FF = Four Seam, FT = Two Seam, CU = Curveball, SL = Slider, CH= Changeup).

Right away, the big difference is that Avila calls more sliders and Holaday calls more curves. That’s probably about individual pitchers, but let’s take a look. I’m going to use another data set I have to break it down, but the classifications are basically going to line up. Don’t stress about the exact comparison to the data above, just compare across pitchers.

Here are the percentage of a catcher’s pitchers that came from each main starter:

You can see Holaday catchers Verlander and Scherzer more often and Avila is a Porcello, Smyly, Price guy with very similar numbers for Sanchez. None of this is surprising, just important to know. Let’s drop out Price because there’s no way to compare something to zero. Pitch type by pitch type. (Note: I don’t think this data is perfect, but it is identically collected so it shouldn’t be wrong between the two. Anything hard is a fastball, splitters and changeups are called changeups).

What do we notice? Holaday likes to go to the changeup more than the slider with Verlander. Holaday likes Scherzer’s curveball more than his slider-changeup combination. The same is true with Holaday and Sanchez. Holaday likes Porcello’s slider more than his curve and change. He was also willing to use Smyly’s changeup a bit more too.

I should note that some of this could be driven by the particular opponents and handedness. Also pitchers sometimes don’t have command or feel for a pitch and the catcher can’t call it as much on a certain day. Basically, we have no idea what causes these differences, but the differences exist as such. Over the course of the season, Holaday and Avila seem to be calling slightly different games for their pitchers.

Going deeper than this gets even harder. What were the results on these pitches? In what order did they come? Where did they land in the zone? Oh, and framing and blocking skills. It’s so hard to capture catcher defense.

One thing I will point out, which I find to be a relevant stat is that once each catcher gets into a two strike count this year, the strikeout rate for both is around 38%. There are a lot of factors that go into this, but for me it’s telling. When their pitcher in position to deliver a strikeout, both seem to call the right pitch with equal frequency.

We can dig deeper than this in the future, but it’s very labor intensive. The differences appear small enough that it wouldn’t be hard for McCann to get up to speed about how he is supposed to call pitches for each pitcher. It would only take a little bit of time to go through the history of each pitcher and talk to them about how comfortable they are with certain pitches in certain counts. It’s not like Avila and Holaday are calling the exact same game, how much worse could McCann be?