May 142010

First off, I want to thank everyone who commented, both positively and negatively on yesterday’s article. I really appreciate your insights.

Moving on, I know that such a system is not perfect, since it has many limitations: A 5 inning, 5 walk, 4 hit, somehow-no-runs, 100 pitches performance looks the same as a perfect game. Conversely, a one inning disaster, think Johan vs. the Phillies, can sway performance wildly. Basically, all segments of 100 pitches are not created equally.

Still, I want to take a look at it over a whole season, because I’m curious as to whom it will overrate, underrate, and get right. I’ll compare the Runs per 100 Pitches number to the pitcher’s ERA/FIP/tRA and see if it comes close to any of those, though I doubt it will come close to the latter two. After all, I’m not accounting for hit type or weighing anything differently. Let’s see what happens when I apply this pseudo-stat to the 2009 Yankee starters. I have a feeling it’s definitely going to overrate the pitchers.

CC Sabathia
R/100P: 2.67
ERA: 3.37
FIP: 3.39
tRA: 3.70

A.J. Burnett
R/100P: 2.86
ERA: 4.04
FIP: 4.33
tRA: 4.88

Andy Pettitte:
R/100P: 3.07
ERA: 4.15
FIP: 4.16
tRA: 4.77

Joba Chamberlain (not including relief outing):
R/100P: 3.45
ERA: 4.75
FIP: 4.82
tRA: 5.30

Alright, what’ve we got? Well, we’ve got something really, really stupid. My first attempt at stat making? Epic Fail.

I thought this would be something worth trying, considering the importance of the 100 pitch mark in today’s game and lack of nine inning performances by starters, but it was not to be. Live and learn, right? I think I’ll leave the stat-making up to the big guys from now on.

5 Responses to “More on Runs per 100 Pitches”

  1. All stats appear to be ridiculous at first. I love baseball, but until recently didn’t follow all of the recent advances in analytics. After reading a number of descriptions that entail the UZR and BABIP, I furthered my understanding of the numbers side of the game. I don’t feel that these are necessary to everyday fans, as most just want to see baseball played, and like to win, but don’t have to every time and will still come out to games. The current state of baseball and the lowered numbers of attendance have to do with the economy, not stats, and not baseball itself. Remember the 80’s, no one came around Yankee Stadium, and most people talked about how it wasn’t safe. Once money was spent on safety and the team, people came around in hoards. I think you just have to search for how this statistic would be properly represented. Possibly couple it with pitches thrown per inning and strike to ball ratio. You constantly hear that managerial decisions are based on # of pitches thrown, and then they consider innings pitched. As well, you cannot compare the actual ERA numbers to those of the R/100PT because they are actually different statistics altogether. That same 5IP, 5 hit, 4 walk performance effects the pitchers ERA positively too don’t forget, even though it isn’t a quality outing. Another route could be Total Bases/100 Pitches Thrown. Keep questioning, those gurus didn’t just come out of their mothers that way…  (Quote)

    [Reply To This Comment]

  2. I think you have something going here actually. Now, let me put out a disclaimer and say that I’m not a stat guru; I’ve never had any statistics training so I can’t validate anything. I’m just hoping someone can pick it up and do it for me.

    Maybe you’re evaluating it the wrong way – instead of trying to make it comparable to ERA/FIP maybe this should be a quantifier of durability? Look at Roy Halladay, his ERA last year was 2.79 and his R/100P was 2.18. He made 32 starts with an average of 7.4 innings pitched/start. Maybe this can be used not like an ERA but a qualitative means of evaluating how deep a pitcher goes into games and how effective he is with his pitches? Or maybe you need to find a way to evaluate the difference between ERA and R/100P and what that means exactly. Roy’s 2009 difference would be 0.61, CC’s 0.70, AJ’s 1.18, Andy’s 1.08, Joba’s 1.30.

    I just took a quick look at Brandon Morrow’s numbers this year, he’s got an ERA of 6.69 and a R/100P of 4.03 this year. His difference is 2.66. He is averaging 5 IP/game this year. Daisuke in 2008 (his last healthy year) yielded an ERA of 2.90 and a R/100P of 1.85, a difference of 1.05. So maybe he’s not yielding a lot of runs, but he’s also not working deep into games? A quick calculation shows that he averaged 5.75 IP/game in 2008.

    I think this is promising. Just figure out a way to apply it.  (Quote)

    [Reply To This Comment]

    bg90027 Reply:

    When you look at differential rather than the stat itself though you lose a lot of whether the pitcher is giving up many runs or not as long as he’s pitching deep into games. Jon Garland for example might look pretty good based on a differential stat even if his ERA or ER/100 are mediocre at best. It’s less ambitous than what you or Matt are trying to come up with but I think IP/100 would be a better stat. I also think Quality Starts is a stat that attempts to measure the same thing and gets underappreciated. I know giving up 3 ER in 6IP is a 4.50 ERA and that’s not great but especially with their offense and bullpen, the Yankees are going to win the vast major of their quality starts. Consider AJ Burnett last year who seemed to either pitch great or terrible. His overall ERA was respectable but his record was disappointing because of his inconsistency. The Yankees were 21-12 in games he started. Only 3 of the 12 losses were quality starts. Maybe there could be a similar stat called “exceptional starts” which might be something like 7 or more IP with 2 or less ER.  (Quote)

    [Reply To This Comment]

  3. Like Steve P says don’t be discouraged. While I expressed doubt at how useful the stat was I don’t mean to discourage innovation. Your premise remains the same, that ERA seems to be a flawed metric, so you are on to something and if you don’t fail you never will succeed. So keep at it!  (Quote)

    [Reply To This Comment]

  4. I don’t think its an epic fail. I don’t even think its a fail. It does seem to correspond well with other pitching stats, which implies validity. Perhaps it needs another dimension for added context, such as innings per 100, baserunners per 100, or simply pitchers per batter.  (Quote)

    [Reply To This Comment]

Leave a Reply



You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

© 2011 TYU Suffusion theme by Sayontan Sinha