12 January 2013

Mr. Frustration -- Jake Arrieta in Norfolk

One of the more frustrating Orioles' pitchers over the past three seasons has been Jake Arrieta. Promoted to the Orioles' starting rotation in 2010 after a superficially impressive first half at Norfolk (1.85 ERA in 73 innings, 11 starts), he showed promise in 2010. In 2011 and 2012, he failed to deliver on that promise; his 2012 was especially disappointing. He was the Orioles' opening-day starter, but was demoted to Norfolk in July after going  3-9, 6.20 2012.

When I work a game in Norfolk, I dread it when Arrieta is the starting pitcher; his games seem to take forever because he throws many pitches. At least that's my impression; the rest of this article will look at Arrieta's Norfolk starts I've worked to see if my impression matches reality. I have worked six of Arrieta's AAA starts from 2009; three from 2010; and two from 2012. I will look at those games, hoping to answer three questions — Am I justified in thinking Arrieta works a lot of deep counts; Is there a connection between Arrieta's deep counts and his effectiveness; Has Arrieta changed over the past four seasons?

I consider any at-bat to which a pitcher makes six or more pitches to be a deep count. There can't be a full count unless a pitcher makes at least six pitches, and in 2012 the Orioles averaged just about 3.94 pitches per batter. So, a pitcher making six or more pitches to a batter is making at least 50% more pitches than the average.

Now, let's look at Arrieta's starts for which I have scoresheets:
DateIPHRERBBSOBFPDeep
Counts
PCT
6/26/0967113226623.1
7/12/095.25663625625.0
7/29/09610551628621.4
8/03/0958442424729.2
8/29/094.29431225624.0
9/04/094.14316223939.1
4/09/1053004220525.0
4/19/1066213526519.2
5/20/10740021028725.0
8/02/125.25224526830.8
9/03/1253111920420.0

Am I justified in thinking Arrieta works a lot of deep counts?
Of course, I don't really know what "a lot" is because I don't have a standard of comparison. However, approximately one out of four batters Arrieta faces is able to work a deep count, and that seems like a lot.

Is there a connection between Arrieta's deep counts and his effectiveness?
Again, we're dealing with a small number of instances, and it's hard to draw hard-and-fast conclusions. There are three starts in which he had significantly more deep counts than normal, and in only one of them could he be considered effective (the start of last August 2).There were two starts in which he had significantly fewer deep counts than normal, and in one he was effective. The evidence is inconclusive, which at least means that there's no obvious connection.

Has Arrieta changed over the past four seasons?
Not based on the games I've seen. Last season, he had one good, almost dominant start and one mediocre start. In his good start, he didn't have many deep counts; in his bad start, he did. There's no evidence to believe that he has learned to become more efficient over time.

Conclusion
There was a lot of hope when Jake Arrieta was moving through the Orioles' system. Since he's reached the major leagues, he's teased us with occasional outstanding games but has generally disappointed. In 2012, he was the Orioles' opening-day starter but was sent down to Norfolk in August. He's listed among the Orioles' rotation options for 2013. Based on what I've seen in Norfolk, both now and in the past, there's no reason to expect anything different from Jake Arrieta in 2013.

7 comments:

Anonymous said...

Give Jake some time - he will come around - I just hope its with us.

I don't want another Curt Schilling scenario

Anonymous said...

This is a good start, but it really wouldn't be that hard to grab some MLB data and figure out how to compare your Arrieta data (e.g. deep counts) to MLB league averages. Instead of just saying it "seems like a lot."

Jon Shepherd said...

"Grab some MLB data" - Where do you suspect that data would be located?

I mean...I can get count data, but with Joe's pitch F/X going away...not sure where to get information like that anymore.

Unknown said...

Curt Schiling is not a particularly relevant comparison for Jake Arrieta. Arrieta pitched 2012 at 26; at age 26, Schilling had been a rotation starter for two seasons and had pitched a shutout in the World Series. Second, the Orioles traded Schilling after he had pitched well out of the bullpen. Third, in his entire career Schilling had one season (of at least 20 innings pitched) in which he pitched worse than Arrieta's best season.

Unknown said...

Average pitches per batter is readily available, but I don't know of any source that keeps track of the number of one-pitch, two-pitch, three-pitch, etc. counts. I could go through each game and count them myself, but I'm not sufficiently induced to do that work.

Stuart Wallace said...

Joe,

The closest thing would be Baseball Reference's pitching splits, where you can parse out state by pitch count - here's 2012 for Arrieta:

http://www.baseball-reference.com/players/split.cgi?id=arrieja01&year=2012&t=p#count

It's not perfect, but it gives you a general idea of his outings by count, and we also see that most of 2012 for Jake was spent in hitters counts.

Jon Shepherd said...

Yeah, I was looking at those yesterday. I'm just not comfortable with the number of foul offs at two strikes to be sure what that looks like. But, yes, it is the best data that I know of to easily get to.