Archive - pitchf/x RSS Feed

Edward Mujica: A case of gopheritis

June 14, 2010 in Myron Logan, pitchf/x, Sabermetrics, San Diego Padres with 5 Comments

Edward Mujica, right-handed pitcher for the San Diego Padres, is a pretty good middle reliever. Check out his numbers since coming to the Padres:

Year Innings ERA K/9 BB/9 FIP xFIP tRA
2009 93.7 3.94 7.3 1.8 4.03 3.93 3.61
2010 31 3.19 9.3 1.2 4.79 3.10 3.94

The strikeouts and walks look good. There is one thing I purposely left out, however, and that is home runs allowed. Mujica has given up 22 home runs in 124.7 innings as a San Diego reliever. That is a 1.6 HR/9 rate. He will have trouble staying in the majors if he continues to allow home runs at this pace. Of course, perhaps he has just been unlucky or in a slump.

We now have a variety of tools at our disposal to investigate Mujica’s home run troubles. First, let’s look at his ball-in-play rates (note that 2006-2008 with Cleveland is a pretty small sample, about 70 innings total):

Year Line Drive Ground Ball Fly Ball IFFB HR/FB
2006 18.5% 26.2% 55.4% 19.4% 2.8%
2007 16% 26% 58% 17.2% 10.3%
2008 23.4% 30.5% 46.1% 10.2% 8.5%
2009 17.2% 39% 43.8% 7.9% 11%
2010 12.3% 40.7% 46.9% 0% 21.1%
Career 17.9% 35% 47.1% 9.7% 10.7%

IFFB: Infield Fly Balls
HR/FB: Home runs per fly ball

According to FanGraphs, the approximate 2010 averages are:

LD: 18.5%
GB: 44.6%
FB: 37.2%
IFFB: 9.5%
HR/FB: 9.3%

What we have in Mujica, then, is a pretty extreme fly ball pitcher. He has a career fly ball rate of about 47%, while the league average is 37%. He is pretty average on line drive%, IFFB%, and a bit over the league average on HR/FB%.  One interesting thing to note is that his infield fly ball percentage is at zero so far this season. So, apparently, every fly ball he has allowed has gone into the outfield. That is probably not helping his HR/FB rate.

Anyway, the sabermetric literature suggests that home runs are essentially a function of fly ball percentage. In other words, HR/FB should regress heavily toward the league average. Let’s take a look at Mujica’s actual home run rate and his home run rate after adjusting his HR/FB to the league average 9.3%.

Year HR/9 ADJ. HR/9
2006 .49 1.6
2007 2.48 1.9
2008 1.16 1.3
2009 1.35 1.1
2010 2.32 1
Career 1.43 1.24

Since 2009, if Mujica had allowed a league average HR/FB rate of 9.3%, his overall home run rate would be about average. While Mujica is a fly ball pitcher, his strikeout and walk rates are good enough that his propensity to allow fly balls should not be an issue, *if* he puts up league average HR/FB rates. Over these past two seasons, however, he has actually given up 22 homers on 135 fly balls, good for a 16.3 HR/FB%.16.3% is a different story.

Hit Tracker gives us another interesting tool to analyze Mujica’s home run struggles. Before looking at the data, the theory would go that, since Mujica has probably been “unlucky” so far in his Padres career (especially 2010), a lot of his home runs allowed should be unlucky home runs, homers that just barely got out.

Here are some definitions from Hit Tracker:

Std Distance (Standard Distance) – The estimated distance in feet the home run would have traveled if it flew uninterrupted all the way down to field level, and if the home run had been hit with no wind, in 70 degree air at sea level. Standard distance factors out the influence of wind, temperature and altitude, and is thus the best way of comparing home runs hit under a variety of different conditions.
“Just Enough” home run – Means the ball cleared the fence by less than 10 vertical feet, OR that it landed less than one fence height past the fence. These are the ones that barely made it over the fence.
“No Doubt” home run – Means the ball cleared the fence by at least 20 vertical feet AND landed at least 50 feet past the fence. These are the really deep blasts.
“Plenty” home run – Everything else, except for the 2 above Homerun types
Lucky Homer – A home run that would not have cleared the fence if it has been struck on a 70-degree, calm day.

Here are Mujica’s eight home runs allowed this season:

Hitter Location Type Std. Distance
Davis NYM No Doubt 446
Davis SD Just Enough 408
Blake LA Just Enough 384
Lee HOU Plenty 406
Pence HOU Just Enough 426
Sandoval SD Plenty 436
Reynolds SD Plenty 381
Upton ARZ No Doubt 454

Mujica’s average Standard Distance is 418 feet (NL average: 395). The home run to Blake in Dodger Stadium is the only “lucky” home run Mujica has given up:

mujica HR

Here are the landing spots on all of Mujica’s home runs allowed this season:

2010_Mujica_Edward_pscatter

He has given up some absolute bombs, some normal home runs, and a couple of borderline ones. It is tough to make any real conclusions based on this data because of the sample size. For a little additional context, last year Mujica’s Standard Distance on his 14 home runs allowed was 387 feet. He gave up four in Petco Park and 10 on the road. Six were classified as “plenty” and eight as “just enough.”

If anything, I think we can say that this season Mujica’s homers allowed have not been of a particular unlucky nature. For the most part, they have been hit hard. Of course, that is not to say that he hasn’t been unlucky. After all, it is still a small sample and there are a lot of variables that go into a home run: bad pitch selection, bad execution, the hitter’s swing, etc. Perhaps Mujica has thrown good pitches that hitters have handled, or maybe he’s just thrown a few bad pitches that have been hammered.

We have yet another tool at our disposal: PITCHf/x. I am just going to look at 2010 data, identify Mujica’s pitch types, usage, and then specifically look into his home run pitches.

According to PITCHf/x, here is what Mujica throws:

Pitch Type 4Seamer 2Seamer Change Curve Slider Cutter
Velocity 92 89.9 86.9 81.4 82.8 89.7
Usage 54.8% 4.5% 27.6% 5.1% 6.9% 1.1%

Here’s the pitch movement graph (x-axis: horizontal movement, y-axis: vertical movement, view from catcher’s perspective):

mujica pitch movement

Mujica throws a split-finger fastball quite often (20% of the time for his career, according to BIS) that PITCHf/x apparently does not pick up on. It appears to me that the changeups classified by PITCHf/x are actually splitters. Here’s a good look at Mujica’s pitches from a bird’s eye view:

mujica flight path

If you look closely enough, you can see the four and two seam fastball, along with the splitter (classified as a change), move in on a right-handed hitter. The cutter, slider, and curve all break in the opposite direction.

Anyway, the last thing I wanted to look at is the location and pitch type on Mujica’s home runs allowed this season:

Mujica, homers allowed

Four homers on four-seamers, three on splitters (I reclassified the changes as splitters), and one on a slider. All located in prime home run territory, up and out over the middle of the plate.

If we’ve learned anything from this, it’s that we can analyze a player’s performance in a lot of different ways and, in the end, still be left with more questions than answers. Mujica is a flyball pitcher who is going to give up his share of home runs. However, if small sample size and regression mean anything to us, he won’t continue to give them up at this rate. At the same time, nobody wants to be testing any mathematical theories in the middle of a pennant race.

Strasburg is human

June 14, 2010 in Myron Logan, pitchf/x, Stephen Strasburg with Comments Off

Stephen Strasburg made his second start of the year in Cleveland yesterday, giving up just one run (a Travis Hafner homer) in 5.3 innings. While he was still dominant, striking out eight and allowing just two hits, he did have some serious control issues as he walked five batters. As the game went on, it appeared Strasburg was having issues with his landing spot on the mound.

Anyway, here’s his pitch movement graph, with horizontal movement on the x-axis, vertical movement on the y-axis, and from the catcher’s perspective:

stras2 

If you remember, in Strasburg’s first start the gameday classifications were not picking up his two-seamer, classifying all fastballs as the four-seam variety. I am guessing that after Strasburg’s start, someone went in and “trained” the algorithm to pick up his two-seamer, since it is clear that he throws one. As you can see, his two-seamer generally moves more down and in on righties than his four-seamer.

It is also a couple MPH slower:

Pitch Amount Avg. Speed Horiz-break Vertical-break
Four-seamer 38 98.5 -6.9 6.4
Two-seamer 22 96.3 -8.5 5
Change 13 91.3 -7.3 -.4
Curve 22 85.1 7.1 -6.9

While Strasburg is interesting enough to watch on the field, check out his early off-field contributions, in the form of increased attendance and ratings. Strasburg’s marquee value may be as high as anyone is baseball right now.

Strasburg dominates in debut

June 9, 2010 in Myron Logan, pitchf/x, Sabermetrics, Stephen Strasburg with 3 Comments

Stephen Strasburg made his debut Tuesday night against the Pittsburgh Pirates.

His line: 7 IP, 4 hits, 2 R, 1 HR, 0 BB, 14 K

Unreal. His lone blemish was a changeup to Delywn Young that caught a little too much of the plate, allowing Young to yank it into the right field stands. For most of the game, however, Strasburg toyed with the Pirates’ hitters, blowing 98 mile an hour fastballs by them, dropping unhittable curveballs, and occasionally resorting to his 90+ MPH change.

Watching Strasburg’s performance, I couldn’t help but think that he has the most dominating stuff I have seen since, I don’t know, Pedro Martinez. Sure, I’m probably caught up in the hype. There are plenty of great pitchers in the game today, and one start against the Pirates doesn’t make you the best in the game.

The highest drama came after the 6th, when Strasburg was at 80 pitches and the Nats, behind home runs by Adam Dunn and Josh Willingham, took a two run lead. There was much speculation from the broadcast booth that the Nationals phenom would be taken out. However, not only did Strasburg come up for the seventh, he struck out the side and left to an electric ovation.

You knew it was coming – let’s take a look at some PITCHf/x data on his debut.

Strasburg, pitch speed chart

That is simply Strasburg’s pitches, in order, accompanied by MPH. He did not lose much on the fastball as the game went on. Here’s his movement chart (from the catcher’s perspective):

Strasburg movement

Horizontal movement is on the X-axis and vertical movement is on the Y-axis, and the data points are color coded by pitch type. You can see that his pitches cluster into three pretty distinct groups; the fastball in the upper left, the change up just below it, and the curve in the lower right.

I think if we studied his pitches more closely, we’d find two fastballs and perhaps a variation of the curve (a true, knee buckling curve and more of a sharp breaking slider/curve). For now, though, the chart gives us a good indication of his pitch movement.

Using the data from Brooks Baseball, here’s his pitched usage breakdown (with average and max speed):

Pitch Type # of Pitches Average Speed Max Speed
Fastball 60 97.5 100
Curve 25 82.2 83.8
Change 9 90.2 91.6

According to PITCHf/x, Strasburg hit 100.1 and 99.9, both in the second inning. He hit 99 or faster five times and at least 98 a remarkable 28 times. It is completely accurate to say this guys *sits* in the upper-90s. At least he did on Tuesday night. I can’t wait until the next time he takes the mound.

Idle thoughts: Fastball velocity

June 2, 2010 in Myron Logan, pitchf/x, Sabermetrics, San Diego Padres with 4 Comments

The general consensus is that a four-seam fastball is thrown with less movement and more velocity than a two-seamer. Check the San Diego Padres rotation, using 2010 PITCHf/x data from FanGraphs:

Pitcher Four-seam MPH Two-seam MPH
Garland 88.8 89.4
Latos 93.5 92.8
LeBlanc 86.4 82.8
Correia 89.9 89.5
Richard 91.4 91.3

Yeah, check that rotation. Prior to looking up the data, my theory was that the four non-Clayton Richard (since we’ve already looked into the data on him) Padres starters would have four-seam fastballs that registered a few MPH faster than their two-seamers. Instead, we see not only Richard averaging the same velocity with each pitch, but also Garland (a faster two-seamer, in fact), Correia, and Latos (within one MPH). LeBlanc, the only one who has a clear difference between the two pitches, uses his two-seamer sparingly (less than 2% of the time).

What gives? Well, we have (at least) three possible explanations:

  1. The Padres employ a staff that just so happens to throw both of their fastballs (ignoring the cutter, for now) at similar speeds. Perhaps they just happen to be on the same staff together or maybe it has something to do with pitching coach Darren Balsley or the Padres organization in general.
  2. We are again encountering problems in classification, by Gameday, and are not witnessing real differences in the two pitches, rather difficulties in properly classifying them.
  3. I’m crazy, and four-seamers and two-seamers are generally thrown at the same speed.

I think we can rule out choice three, as Dave Allen says here that four-seamers are indeed thrown faster — “about 1.5 mph faster than two-seam fastballs and 3.5 mph faster than cutters.”

My guess is that it is some combination between one and two. But, frankly, I have no idea what the correct answer is and whether or not it is significant. Any thoughts?

Clayton Richard and the two-seamer

May 29, 2010 in Clayton Richard, Myron Logan, pitchf/x, Sabermetrics with 8 Comments

Our friend Zach did an analysis of Clayton Richard’s 2010 success over at Gaslamp Ball. He says:

In 2009, Richard threw his 4-seamer 51.3% of the time, and the 2-seamer only 19% of the time. In 2010, he’s dropped his usage of the 4-seamer, throwing it only 23.8% of the time, and upped the use of the 2-seamer, now throwing it 35.6% of the time.
Richard has also increased the velocity of his 2-seamer. Where in 2009 he averaged 87.6 mph, this season he’s throwing it at an average of 91.2 mph.

Indeed, if you look at the PITCHf/x data, located on FanGraphs, that’s what comes out. Check out his usage, in 2009 vs. 2010:

Year Four-seamer Two-seamer Cutter Slider Change Curve
2009 51% 19% 2% 15% 11% 2%
2010 22% 37% 13% 4% 18% 5%

It struck me a bit that he’d change his repertoire so much in one-offseason. What’s also striking is the speed of his two-seamer; it has jumped almost four miles per hour in one season, while his four-seamer – and basically the rest of his pitches – stayed at the same mph.

As I thought about this, it started to click that perhaps Richard’s pitch selection hasn’t changed that much, maybe the PITCHf/x algorithm has. I wanted to look at the raw PITCHf/x data to see if I could find anything.

The first graphs I’m going to show has horizontal movement on the x-axis and vertical movement on the y-axis, and the pitches are a different color based on their pitch type (as classified by MLBAM). Remember, the view is from the catcher’s perspective:

2010
Richard 2010 new

2009
Richard 2009

You can see the two and four-seamers clustered in the top right of the graph. In 2010 you can see the two-seamers have slightly more horizontal movement and slightly less vertical movement. What is revealing is if we remove two-seamers to get a better view.

2010
Richard 2010, no two seamers1
2009
Richard 2009, no two seamers1

Notice the range of the four-seamer, the circled purple triangles, is much wider in 2009. It looks like in 2009, the algorithm was very reluctant in calling a pitch a two-seamer. Most everything in that upper right hand corner was identified as a four-seamer. In 2010, however, it is much more equal – in fact, as you can see by the graph and the chart above, the two-seamer is now being identified more often than the four-seamer.

If we take the 2010 version of the pitch algorithm as the more accurate one, then it appears what happened in 2009 is that a lot of two-seamers were actually classified as four-seamers. So, in reality, Richard’s pitch selection has probably not changed that much – the pitch algorithm has.

And, indeed, it appears that is what has happened. In an email exchange – and here on the Hardball Times – Mike Fast explained to me that Ross Paul, the guy doing some of the behind the scenes work at MLBAM, is now using a neural networks to classify pitches. A neural net is specifically trained to identify pitches from each pitcher, and should theoretically be a more accurate way to classify a pitcher’s arsenal.

****

I also created spin graphs for Richard, and while they may not add to the point of this article, I think they are still worth sharing.

2010
Richard, 2010 spin

2009
Richard 2009, spin

Interestingly, it appears that the same thing may be going on with sliders and cutters. It appears that a lot of Richard’s pitches that were being classified as sliders last year are now being classified as cutters.

Page 2 of 9«12345»...Last »