Saturday, April 12, 2014

Analysis of Rick Porcello: Game 2


After a good first start last weekend versus the Orioles, Tigers right hander Rick Porcello was much less effective in his second second start last night in San Diego.  He pitched 6 1/3 innings allowing five runs on ten hits in a 6-0 loss to a masterful Andrew Cashner and the Padres.  What went wrong?

According to Brooks Baseball, Porcello threw a lot of sliders last night - 18 of them compared to just 11 curve balls.  His curve is better than his slider, so you may wonder why he would throw more sliders.  It might be because his slider was effective last week versus the Orioles - a linear weights value of -1,10 (negative numbers are good).  Last night, he was 2.90 (positive numbers are bad) on his slider.  He had favorable results on his sinker and change-up in both outings.

Porcello needs to keep his pitches down to be effective and he was unable to do that last night.  After getting 12 outs on ground balls in his first start, he he had just 7 on Saturday.

Last night, his location was off versus left-handed batters in particular.  The strike zone plots below indicate that he located fewer pitches in the middle of the strike zone versus left-handed batters on April 5 (figure 1) than he did last night (figure 2).  The result was that lefties went 6 for 14 against Porcello last night compared to 2 for 11 in his first start. 

Figure 1: Rick Porcello's Strike Zone Plot Versus LHH - April 5, 2014

http://www.brooksbaseball.net/pfxVB/cache/location.php-pitchSel=519144&game=gid_2014_04_05_balmlb_detmlb_1&batterX=&innings=yyyyyyyyy&sp_type=2&s_type=3&league=mlb&cache=1.gif
Data source: Brooks Baseball

Figure 2: Rick Porcello's Strike Zone Plot Versus LHH - April 11, 2014

http://www.brooksbaseball.net/pfxVB/cache/location.php-pitchSel=519144&game=gid_2014_04_11_detmlb_sdnmlb_1&batterX=&innings=yyyyyyyyy&sp_type=2&s_type=3&league=mlb&cache=1.gif
Data source: Brooks Baseball

So, there you have it.  Like many of Porcello's poor starts, he couldn't keep the ball down and couldn't get left-handed batters out.  Next up, he'll face the Indians on Thursday

8 comments:

  1. So looks like they changed their system to no longer show "foul bunts". And curious why they have 9 categories in the first example but with only 6 colors. This is a goofy system. I mean how do I know what a "ball in dirt" is if every ball is the same color? And wouldn't we know it was in the dirt if it hit the line on the bottom anyhow? Same with balls in play, why do we have to have 2 categories for that when they can't even distinguish which is which? They seem to want to tell us which ones create runs but why and what if the runs are caused by an error anyhow?

    This is a goofy system and I don't get it. I wouldn't even trust these people to plot the data points correctly. If they can't even communicate cleanly and clearly with the data then to me it's all compromised data that can't be relied upon.

    ReplyDelete
  2. They don't plot the data. It comes from cameras.

    ReplyDelete
    Replies
    1. Well yeah I was being facetious. But I don't know what the process is between the camera sighting a pitch and the creation of their chart that supposedly has the dots in the right place from what the camera saw. So the joke is I'm not willing to trust them with what should be the simplest part of the process which is to have a computer transfer it for you if they can't even make a logical legend to explain the dots.

      It's just disappointing. And you can barely even distinguish the red colors too. They represent on their website that they are doing the best work in their field and this stuff looks a child created the legend. On the front page they list the people who this is for, and SCIENTISTS is one of them! No scientist is going to appreciate this legend and the bizarre confusion it unleashes, this is NOT science.

      Delete
  3. I think it's an awesome site. The stuff they do is really difficult and I'm happy they are doing it.

    ReplyDelete
    Replies
    1. Yeah it looks encouraging and very promising, I would expect that in the future it can only get much better. And I'm glad that they have taken the initiative to explore these ideas so that's all fine and dandy, they just need to realize that very simple oversights can really have an effect on the impact of the story you are telling.

      And it's possible that "foul bunts" didn't make the next list if there wasn't one in that game. What would be really cool is if each dot you could hover over to get the actual results, or to filter it for various questions you might ask. If we are giving up an inordinate number of doubles as an example, well let's filter for only the double dots on these charts and take a look. Or maybe you could sort by RBIs created in clutch scenarios. Or how about the dots that are during an attempted steal. Maybe we aren't pitching with the right strategy in obvious running scenarios.

      It should be interesting to see how this evolves over time, but anything that I don't like I'm going to catch and then complain about it, cause well that's what I do, I take pictures of peoples' mistakes, I'm just a logic detecting camera and I see the good and the bad and snap the picture and post it!

      Delete
  4. Why don't you develop your own system and correct the imperfections?

    ReplyDelete
    Replies
    1. Well to breakdown pitch data just wouldn't be something that I have the infrastructure to be able to develop such a system on my own. And there would be other areas of baseball science that would appeal to me more to work on first.

      But one of the other questions I have is where does a "ball in dirt" show up on the chart? If "foul bunts" doesn't show up because there wasn't one in the second chart, then that would mean a "ball in dirt" did show up in both charts. So why are those dots so high? I'm confused by the spatial orientation of those dots, which we still can't confirm which are "balls" vs "ball in dirt".

      Delete
  5. Porcello is just not a very overpowering pitcher (5.5K/9 this year), which is pretty much on par for his career average as well.

    ReplyDelete

Sabermetrics Book

Sabermetrics Book
One of Baseball America's top ten books of 2010

Blog Archive

Subscribe

501 Baseball Books

501 Baseball Books
Recommended by Tiger Tales

Stat Counter

Site Meter