Passing out POTY ballots

Who’s your Texas Rangers Pitcher Of The Year for 2007? There’s still a little way to go, but the contenders narrow themselves down by the day. In the rotation, most of them have sucked for most of the year, with occasional bursts of good performance, but not enough to lift them into POTY contention. McCarthy might have been, what with his 3.41 ERA since the beginning of May, but he’s only pitched 74 innings in 14 games in that time, and in fact today he once again went on the DL, this time with a cracked shoulder blade (which leaves one wondering how good he would have been if healthy in that time – or better yet, how do we crack some of those other guys shoulder blades?).

In the bullpen there are, as usual, several good candidates. Gagne would have been one, but he was only here half the year, so he loses credit for being traded. Aki will have missed half the year with injury, so he’s out too. Mahay did some good now and then, but he’s gone, even Willie Eyre stepped up and performed well at times when he was needed badly. But really, as you probably already guessed, there are only two real candidates: Joaquin Benoit and CJ Wilson. As all around them have come and gone, those two have been steady as rocks, coming up with the goods day in and day out, and so with a month and a half left to go, it’s time to compare them as they stand now, and set them up for a grandstand finish in the POTY race.

Benoit started the year in the seventh inning slot, setting up for Aki and Gagne to end out games. Of course, early on, the offense and the starting pitching didn’t get the team into position where the back three were needed that much. But Jack has pitched well, used a lot with just one day’s rest, and used with the game tied or with a one run lead. His strikeout-to-walk ratio is about 3.5 to 1, which is excellent, especially as he’s striking out about 10 batters per 9 innings. His ERA is 2.71 as I write, which is a little deceptive because he gave up 6 runs in 1 inning in an April appearance, and 4 runs in 2 innings in a June appearance. Take out those two appearances and for the rest of the year his ERA is an astounding 1.35.

CJ’s role was as a lefty early on (not that his role could be anything else!), coming in early or the middle of the game, wherever needed to get some innings in. As time went by though, his role got bigger and bigger, and with Aki and Gagne gone, he finally moved into the closer role at the beginning of August. I had said for a couple of months that I thought he would be the closer one day, but I did not even expect it to be this year, let alone as soon as it was. CJ has mostly pitched on one or two days rest, and has varied the times when he came in, although of course now he’s the closer he’ll get a lot more leads in the ninth to keep hold of. CJ’s ERA is now 2.29, and although he’s struck out about one batter per inning, his walks are a little high at almost one every other inning. He has improved as the year has gone on in walks though. He’s given up two or more runs in five appearances (including tonight, although he did still get the save), take those out and his ERA is 0.71. That’s not quite as fair as taking out those two for Jack though, since they’re a lot more regular. Still, excellent work, and in fact CJ did just go through a period of 12 or 13 innings without allowing a hit.

Let’s take a look at what they’ve been throwing this year, starting with Joaquin Benoit:

Jack Speed HV

He has three pitches, this chart shows the horizontal and vertical breaks on them, related to speed. For each pitch, the darker color is horizontal, lighter is vertical. You can clearly see the three pitches, fastball in green at the top, slider in the middle in blue and the red changeup at bottom. He doesn’t have a curveball. His fastball runs 90-95 mph, the slider 85-92 and the change about 80-86. Of the 596 pitches I have for him in my Gameday database, he threw fastballs just under half the time, sliders just over a quarter and changeups just under a quarter.

Jack Horiz Vert

Looking closer at just the horizontal and vertical breaks, you can see his fastball and change break to the left, his slider doesn’t break much horizontally at all. The fastball is a little high, coming in at about 10 inches, but the other two pitches average about 5 inches, which helps keep the ball down.

Jack Breaks

Looking at the breaks, the angle of his fastball is pretty high, averaging about 40 degrees, while the other two pitches have pretty good length on them, mostly in the 6-8 inch range but regularly getting up to 10 inches.

Benoit is a righty, of course, so let’s look at the lefty Wilson:

CJ Speed HV

CJ shows four pitches, although without the color coding it initially looked like just three. The fastball at the top is really two different pitches, as we’ll see in a moment, both thrown at 90-95 mph. I’ve been trying to figure out all night what the other two pitches are, and I finally decided that the green one is the slider, at 80-85, and the red one is the mythical gyroball. I decided that because the green pitch is in a similar position to sliders I’ve tracked with other players (compare to Benoit’s above), and because of the publicity about the gyroball and how variable it is. In addition, CJ has said he’s thrown the gyro about once every appearance, and with 50 appearances under his belt, the red pitch appears 54 times in my database and the green 74.

CJ Horiz Vert

Now take a look at the horizontal and vertical breaks by themselves, and you’ll see why I separated the fastball into two pitches. Funnily enough, while the previous picture was harder to tell apart without color, this one was easier to tell apart when it was black and white. At that time you could clearly see a gap between what are the blue and purple pitches. Combining this chart with the one above, which separates the green out, we see what happens with the pitches. The blue fastball is averaging a horizontal break of 4.8 inches, and vertical of 9.1. The purple fastball gets a horizontal average of 9.6 and vertical of 5.9. If you look closely at the previous picture, you’ll see the dark blue and light purple together, and dark purple and light blue together. This is what distinguishes the two pitches. Again, a lot of research uncovered an article on from Jerry Crasnick, which said he throws a live fastball and hard sinker. Because of that I conclude the purple pitch is the sinker (it has the lower vertical) and the blue is the fastball.
CJ Breaks

Break angle and length show the distinctions in the pitches as well. In this case the sinker is breaking a little more than the fastball, and about the same rate as the slider. The gyro is much more variable.

I have 584 pitches tracked for CJ, just 12 less than I have for Jack. He throws the fastball about 46% of the time, and the sinker about 32%. The slider and gyro are thrown about 12% and 9% respectively.

So far everything’s been even, they’re both throwing well with a variety of pitches, with a lot of strikes and not many walks. At this point they’re fairly even, but there’s one thing left to look at. Look at the Rangers charts on the Fangraphs site, and scroll down to the relievers.  There’s a number of stats here, but I want to focus on a couple of them.  Leverage is a rating of how crucial a situation is, for example a runner on first with none out in the first inning is not nearly as important as a runner on first with none out in the ninth inning.  In the chart, the stat pLI is the average leverage situation the player has appeared in.  You can see that Gagne and Aki are around 1.65, meaning 1.65 times the average leverage situation for all players.  Jack shows up at 1.35, and CJ at 1.03.  This means that Jack has been trusted with more important situations that CJ has, so far this year.  I would expect this to change, now that CJ is closing, but so far Jack is ahead.  The other stat to look at is WPA, or Win Probability Added.  This is the likelihood that a team will win given the game situation, and the change therein based on the player’s performance.  If you pitch badly, pushing your team towards a loss, this will go down, and if you pitch well, especially in high leverage situations, your team’s chances of winning will go up.  The better you do, the higher your WPA.  In this case, Jack has a 2.40 WPA, higher than anyone else on the team, including the batters.  CJ at 1.05 has done pretty well too, but Jack is head and shoulders above in this stat.  The final stat, BRAA, is Batting Runs Above Average, meaning how many runs more than an average player have they scored (or how many fewer have they given up, in the case of pitchers).  This somewhat tracks ERA, but gets better the more innings you pitch with a lower ERA.  CJ is leading in this category, in fact leading all Rangers including the hitters, but Jack is just a little way behind (and look at how terrible the rotation has been).

What is there to conclude from all this? They both have a range of pitches and they’re both throwing them very well this year. They’re both excelling in what they do. With the loss of other pitchers via injury and trade, these two guys have stepped up and taken charge. If the Rangers were to suddenly find themselves in contention, I would be very happy having CJ and Jack running out there for the 8th and 9th innings, and I would be confident the Rangers would get good results with them. Overall I give a slight edge to Jack right now, but there’s still a month and a half left for CJ to make his case for Pitcher of the Year.


3 Responses to “Passing out POTY ballots”

  1. Enhanced Gameday analysis catalog update « Fast Balls Says:

    […] August 16, he published “Passing Out POTY Ballots”, an article about choosing the Rangers’ best pitcher of 2007, and also classifying pitches […]

  2. Alan Nathan Says:

    Tell me the characteristics of the pitch you are calling the gyroball. You don’t label your axes so it is hard to know what you are actually plotting.

  3. Steve West Says:

    I’m bad at labeling, I need to get better at that. For CJ, in this post I have three charts. The first shows speed on the vertical axis, and the horizontal and vertical breaks (pfx_x and pfx_z) on the horizontal axis. The pitches themselves are split into a dark (horizontal) and light (vertical) version of the same color. For example, what I’m calling pitch 2 is dark red and light red for the two breaks.

    The second chart is simply pfx_x horizontally and pfx_z vertically.

    The third is break angle horizontally and break length vertically.

    In all three charts, what I’m calling the gyro is the pitch in red. It shows a wide variation in speed, from about 70-87, mostly around 85. It is the only one of his pitches that has a negative horizontal break, and has a much lower vertical break than the others. It has a slightly positive break angle and a higher break length than the others. I conclude it is the gyro based on reading about the pitches he throws, about the number of times he has thrown it (about once a game, he says), and the fact that it has a much higher spread, vertically and horizontally, than the others, which suggests both a less controllable pitch and a pitch he has recently learned to throw.

Leave a Reply

Fill in your details below or click an icon to log in: 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

%d bloggers like this: