Graph of the Intermittent Time Period

The Phillies responded to an embarrassing four-game sweep at the hands of the Houston Astros at home by flying out to San Diego and sweeping the Padres in three games. The offense didn’t exactly wake up, but the starting pitching was immaculate, and in the finale yesterday, the Phillies did manage to score more than three runs for just the second time since August 19.

Just how good has the pitching been?

Date Opp Result IP H R ER BB SO HR HBP SP GS Starter
8/13 @ NYM L,0-1 8 5 1 1 2 8 0 0 74 Hamels
8/14 @ NYM W,4-0 9 6 0 0 0 10 0 1 81 Halladay
8/15 @ NYM W,3-1 9 6 1 1 2 7 1 0 61 Kendrick
8/17 SFG W,9-3 9 7 3 3 0 10 2 0 60 Oswalt
8/18 SFG W,8-2 9 8 2 2 0 8 2 0 56 Blanton
8/19 SFG L,2-5 9 8 5 5 1 9 1 0 37 Hamels
8/20 WSN W,1-0 9 10 0 0 3 7 0 0 63 Halladay
8/21 WSN L,1-8 9 12 8 8 5 6 1 0 29 Kendrick
8/22 WSN W,6-0 9 5 0 0 1 11 0 0 74 Oswalt
8/23 HOU L,2-3 9 10 3 3 0 11 0 0 70 Blanton
8/24 HOU L,2-4 16 13 4 4 4 15 1 2 66 Hamels
8/25 HOU L,2-3 9 6 3 3 0 8 2 0 59 Halladay
8/26 HOU L,1-5 9 13 5 5 0 3 1 0 39 Kendrick
8/27 @ SDP W,3-2 12 7 2 2 1 10 1 1 74 Oswalt
8/28 @ SDP W,3-1 9 6 1 1 1 6 0 0 58 Blanton
8/29 @ SDP W, 5-0 9 5 0 0 0 6 0 0 80 Hamels
TOTAL 153 127 38 38 20 135 12 4
AVERAGE per 9 IP 7 2 2 1 8 1 0 61

A quality start corresponds to a game score of 50. Phillies starters have put up a game score of 50 or better in 13 out of the last 16 games. Roy Halladay (avg. 68), Roy Oswalt (69), and Joe Blanton (61) are 3-for-3 in quality starts in their last three starts, while Cole Hamels (64) is 3-1 and Kyle Kendrick is 1-2 (43).

Since August 13, the Phillies’ overall pitching (including the bullpen) has averaged a 2.24 ERA with 7.9 strikeouts and 1.2 walks per nine innings, a ratio of nearly seven-to-one. Halladay leads all of Major League Baseball (min. 100 IP) with a 2.87 SIERA and should now be the favorite to win the National League Cy Young award. Hamels went into yesterday’s start fifth in the NL with a 3.22 SIERA and Oswalt wasn’t far behind in seventh at 3.34.

As ill as fans felt after the Astros series, the Phillies are still winners of nine out of their last 15 games, good for a winning percentage of .600. As mentioned on Friday, the Phillies still have plenty of opportunities to gain significant ground in both playoff races. With a 1-2-3 punch of Halladay-Oswalt-Hamels that rivals that of any other in Major League Baseball, the Phillies will be ready to play some good baseball in the month of September. Should they be fortunate enough to reach the post-season for a fourth consecutive year, be it via the Wild Card or by winning the NL East, they will be the team the rest of the field least wants to meet.

Leave a Reply

*

20 comments

  1. Aaron H

    August 30, 2010 07:52 AM

    I knew that the 1-2-3 of Halladay, Oswalt, and Hamels would be brilliant, but can we get a little love for Joe Blanton? Ever since giving 5 to Cincinnati on the 7/13, he’s been a fantastic #4 starter on most nights. He’s only had three starts with three or more walks while tossing 7+ K’s most nights.

  2. Morton

    August 30, 2010 08:41 AM

    Pitching is great, but they still can’t score runs on a consistent basis!

  3. Dan

    August 30, 2010 10:16 AM

    Blanton is good for a #4, but I still don’t like Kendrick in our rotation at all.

    Then again, he is essentially being compared to an elite 1-2-3 punch, so maybe it’s just that we’re becoming spoiled in the pitching department.

    I’m curious what we decide to do for our #5 starter after this year, though. Do we sign Moyer to another 1 year deal and assume that he won’t get hurt? Or do we turn to the market which has some decent arms entering free agency? Or do we stick with Kendrick (I still don’t like that option)?

  4. nik

    August 30, 2010 10:25 AM

    For our #5 starter for 2011, I would look for an outside arm. It is obvious that our offensive core may be on the decline and we need to compensate with stronger pitching. Having a more reliable #5 would improve our chances quite a bit. Kendrick would still be an excellent long-reliever/spot starter/injury replacement though.

  5. Aaron H

    August 30, 2010 11:05 AM

    I agree with nik on this one-they should try to send Kendrick to the pen if they have other options, particuarly considering his platoon splits. Righties have a career .696 OPS against him (Brian Schneider), but lefties are hitting .924 OPS off him (Joey Votto).

    As for getting a #5-why can’t they give Drew Carpenter, Vance Worley, or (gasp!) Brandon Duckworth a shot next season? They’re all cheap options, and how many teams have a “lockdown” #5 anyway?

  6. Ryan

    August 30, 2010 11:41 AM

    @ Aaron H – I think Worley is going to be the guy that pushes. There’s a chance he gets to start one game of the double header coming up next week, and he’s really done all he’s needed to do at the minor league level.

    With some of these fringe type starters, eventually you have to give them a shot. At his age, after posting sub 3.50 at AA and AAA this year, he’s going to need a look.

  7. MikeC

    August 30, 2010 11:59 AM

    Bill,

    I really appreciate these little data dumps that you give us, so I hope you take this minor observation as constructive criticism: I think a bar graph makes more sense than a line graph for this type of analysis.

    Line graphs imply trends; ie, the line between the points you are graphing has meaning. This line graph makes the starting pitching look rather erratic, which I know is not the point you are trying to make.

    If you want to compare discrete data points to each other or to an established standard (quality start = 50), then a bar graph would tell a better story here.

    Mike

  8. hunter

    August 30, 2010 12:00 PM

    I was really hoping for another photoshop of Pat Gillick.

    It can be hugz tiem now?

  9. Steve K

    August 30, 2010 12:39 PM

    I cannot for the life of me figure out why Blanton is so good and Kendrick sucks when Kendrick has pitched more innings, has a lower ERA, a lower WHIP, fewer hits allowed, and a winning record. Someone enlighten me.

    Steve

  10. hunter

    August 30, 2010 01:08 PM

    “I cannot for the life of me figure out why Blanton is so good and Kendrick sucks when Kendrick has pitched more innings, has a lower ERA, a lower WHIP, fewer hits allowed, and a winning record. Someone enlighten me.”

    I guess I would agree with you if the time period I was evaluating was just this year and I didn’t factor in to the equation that Blanton was coming off an injury.

  11. Bill Baer

    August 30, 2010 03:02 PM

    @MikeC

    Line graphs work well when displaying data over a continuous period of time, like August 13 to 29 above. I also disagree that it makes the starting pitching look erratic, as you can still easily read that there are only three data points below the bolded average (50) line.

    In the end, there really isn’t much of a difference between reading the information presented here in bar or line graph form.

  12. Bill Baer

    August 30, 2010 03:06 PM

    @Steve K

    Why Kyle Kendrick has better numbers than Joe Blanton:

    - Much lower BABIP: .286 to .334
    - Slightly lower strand rate: 70% to 66.7%

    Going forward, expect Blanton to perform at a level at least about a half run of ERA better. (Blanton has a 4.12 SIERA to Kendrick’s 4.88.)

    (That doesn’t mean that Blanton WILL achieve results that much better, but that if both pitchers continue to pitch in a similar fashion, Blanton should emerge with better results.)

  13. Bill Baer

    August 30, 2010 03:08 PM

    In agreement with Ryan above — I have no qualms about filling the #5 spot with an internal option, such as Vance Worley. Why go out and spend several million on a veteran #5 when you can pay a kid the Major League minimum to do the same job?

  14. dejesus54

    August 30, 2010 03:38 PM

    I was thinking Duckworth, Worley or Carpenter as well. It would never happen, but I would also think four-man rotation. Wouldn’t an extra 150 innings between Halladay/Oswalt/Hamels/Blanton be better than 150 innings of any #5 starter in the majors? (I’m aware you would need a better bullpen to make this work best– which should be a focus in the offseason anyway.)

  15. Bill Baer

    August 30, 2010 03:49 PM

    I think eventually that smaller rotations will come back in vogue but right now, given the amount of credence given rituals and routines, I doubt any team — especially the Phillies — will deviate.

    According to this article:

    www.hardballtimes.com/main/article/how-good-is-your-4-starter/

    …the average #5 starter in the National League puts up a 6.26 ERA. That actually makes Kendrick look quite appealing, relatively speaking.

  16. Ryan

    August 31, 2010 09:09 AM

    Bill – I understand looking at the NL average, and I’m right there with you. There have been contracts in the range of 3-5 million paid for pitchers slotted into the no. 5 slot.

    You can reasonably expect league average production, even from a rookie – but at 400k. That seems like no brain way to save your team 3.5-4.5 million to spend elsewhere.

    But.. and I just completed an article on this – lets just say for the sake of the argument that the Phillies offense is legitimately regressing. Lets say that even with a full healthy year of Chase and Howard, you can expect 50 fewer runs than they averaged from ’07-’09 because of age related regression from the older players, and with the inevitable loss of Jayson Werth.

    If the offense IS taking a step back, does that mean that the Phillies need more than the league average production from the No. 5? If they are no longer a top 2 or 3 offense, does that mean they need to continue to build a top 3 or 4 staff? If so, 6.26 from the no. 5 doesn’t get you to that level.

    Obviously I know that assuming that type of regression is premature. But it’s just food for thought.

  17. MikeC

    August 31, 2010 01:24 PM

    Bill,

    At the risk of belaboring a point that no one else is interested in…

    Even though I expected your response, I still found it somewhat disappointing. After all, don’t you spend a decent amount of time debating with people about baseball stats? And don’t you do this because many of them are ignorant and you are an expert? And yet you choose to argue about constructive criticism for your graphics when you clearly have no clue about quality graphic communication.

    Let me spell a couple of things out for you.

    First, you can debate whether or not a bar chart is the BEST graphic for this particular data set, but not whether or not it is more appropriate than a line chart.

    Second, you cannot debate my point that the graphic suggests that the pitching is erratic – that is just a fact. I’m talking about the graphic, not the data points on the graphic. The graphic paints a very different picture than the story you claim to be telling. As proof of this, simply post it to Google Finance and wait for people to start jumping out of buildings. The graphic paints a picture of instability.

    Third, the reason that you think the graphic is legible is that you know the data so well. As proof of this consider the fact that you claim “you can still easily read that there are only three data points below the bolded average (50) line”. Bill, you can barely see the “bolded average (50) line”; you just assume everyone can read it because you “know” it’s there.

    You don’t actually need the graphic because you think in spreadsheets; in fact, you could probably plot the data points on a polar chart and they would still jump out at you. This is probably true of many of your readers, too; I merely make this argument for the subset of readers who don’t think in spreadsheets, and whom you seem to want to reach when you say “and for those of you who think graphically…” But communicating graphically is a skill that has to be honed just as any other skill; it is not accessed by pressing the “make chart” button in excel.

    If you decide you are interested in improving your graphic communication, you can check out this book:
    THE VISUAL DISPLAY OF QUANTITATIVE INFORMATION, by Edward Tufte
    www.amazon.com/Visual-Display-Quantitative-Information-2nd/dp/0961392142/ref=sr_1_1?ie=UTF8&s=books&qid=1283278479&sr=8-1

  18. Bill Baer

    August 31, 2010 02:22 PM

    Mike,

    There is no need to be insulting. As I wrote in the “about” section, I encourage debate but do so civilly. If you reply with further insults, I will not hesitate to revoke your commenting privileges.

    Here are the differences you’re describing:

    (I couldn’t get Excel to ignore the off-day on August 16.)

    There’s almost no difference, especially when you consider that line graphs simply connect the top-center of each bar. The data appears just as “erratic” in both graphs.

    At any rate the interpretation of the data being “erratic” relies not on knowledge of the data display, but of the data itself. It is mentioned in the article that 50 is the average and it is bolded on the original graph (however visible it may or may not be). Having that knowledge will allow you to interpret the data correctly, regardless of the efficiency or inefficiency of the graph. The only way I could really screw with it is by changing the scale, starting at, say, 25 instead of 0.

    Bar graphs would be great if I had qualitative information on the X axis, such as “Hamels”, “Halladay”, and “Oswalt”. Line graphs are great when time is displayed on the X axis.

    Example:

    www.statcan.gc.ca/edu/power-pouvoir/ch9/line-lineaire/5214824-eng.htm

  19. Mark

    August 31, 2010 03:20 PM

    When in doubt, go to the pie chart… too funny

  20. MikeC

    August 31, 2010 05:40 PM

    Bill, I’m sorry that you thought I was being insulting and uncivil. It was unintentional. I’m unsure why you concluded that. The only obvious candidate in my post would seem to be my “clueless” comment, which may have been an overstatement of an otherwise valid observation. I don’t expect you to be an expert in my field, just as I do not claim to be an expert in mathematics, statistics and probability theory, even though I enjoy your analysis and debates. And I do believe I called you an expert…

    Anyway, I think you have proven my point even if you do not. As a form of communication graphics have a life of their own, and you cannot always control the context in which they are read. I thought the critique might interest you because you seem to be getting more and more opportunities to share your expertise in the blogosphere.

    If you have any other issues with my comments, please feel free to take them up with me personally. You have my email address.

Next ArticleRyan Howard's Malaise