Graph of the Intermittent Time Period

Happy Halladay, everyone!

Here’s something to get you through the next few hours until Roy takes the mound in Colorado tonight.

J.C. Romero had another shaky outing last night, the second out of five appearances in which he has allowed a run. Overall, he has walked four batters in two and two-thirds innings. His fastball thus far is, on average, three MPH slower than his career average of 91.5 MPH. His slider has accounted for one out of every four pitches he has thrown in 2010 compared to about one out of every six over his career.

You may recall at the end of March, I recommended that Romero be used strictly as a LOOGY in lower-leverage innings because he is not nearly as good as his ERA has indicated in his time with the Phillies. Behold:

Romero has severely out-performed his ERA estimators in his two and a half seasons with the Phillies going into 2010. He has been extremely fortunate, stranding 86 to 90 percent of base runners with a BABIP between .236 and .239 since 2007. As great as Romero has looked, those are simply unsustainable numbers, especially when coupled with his unacceptable walk rate that has been between 5.8 and 7.0 since ’07.

Simply put, Romero is a poor man’s Oliver Perez. Would you trust Ollie in high-leverage situations? Neither would I.

On the topic of relievers, check out my explanation of the new shutdown/meltdown statistics at Baseball Daily Digest.

Leave a Reply


This site uses Akismet to reduce spam. Learn how your comment data is processed.


  1. phatti

    May 11, 2010 07:08 PM


    Makes a lot of sense. Except if he’s going to be a LOOGY, he’s got to get them out. They went 2 for 3 off him yesterday (although Giambi’s ball wasn’t exactly hit hard). Hopefully this won’t be an issue as the season goes on.

  2. Bill Baer

    May 11, 2010 07:43 PM

    Romero has shown a legitimate ability to get left-handers out, so I would attribute last night’s struggles to a small sample. He also walks lefties at a lower rate than right-handers.

  3. Murgatroid

    May 11, 2010 11:03 PM

    Sorry to bother you about this shutdown/meltdown stat again, but I just don’t understand what it’s intended to do. I understand why percentages aren’t a good tool of comparison, but I don’t understand why differential is any better. For instance, a look at last year’s standings puts Matt Capps in the middle of the pack for SD/MD differential, tied with Kiko Calero. However, Calero was clearly a much better pitcher last year. Capps had a terrible season, with an ERA of 5.80, a WPA of -2.83, and a WAR of -.4. In addition, in 15 of his 57 appearances, Capps registered a negative WPA, which dipped below -.3 10 of those 15 times. Meanwhile, Calero had an ERA of 1.95, a WPA of 1.36, and a WAR of 1.4. Despite having 10 more appearances, Calero had 2 fewer appearances in which he registered a negative WPA and did not have a single game in which his WPA was worse than -.3.

    This stat claims to be blind to roles, but it seems to support closers. Even Lidge isn’t as low as he should be, considering that, by most metrics, he had a negative contribution to the team last year. Also, when Herndon picked up one of his two meltdowns this year, it was statistically impossible for him to pick up a shutdown. Isn’t one of the problems with saves that non-closers can pick up blown saves but not saves? Obviously Herndon didn’t pitch well, but to a certain extent, a player’s SD/MD differential isn’t under his control. If Joe Girardi had pitched Chamberlain only in the situations he’s used Mitre, Chamberlain would have at best 2 shutdowns and 0 meltdowns this year. The requirements to earn a shutdown or meltdown change depending on the situation, so a player’s shutdown/meltdown stats are dependent how the manager chooses to use the pitcher. Meltdown opportunities aren’t even necessarily shutdown opportunities (and vice versa).

    I think the basic intention behind SDs and MDs is to create a statistic that measures a RP’s performance in high leverage situations. If that is the intention, why not just set a leverage threshold, with a shutdown behind a positive WPA performance in a high leverage appearance? That seems like it would be more effective.

    Please tell me if I’m missing something, but I’ve spent a lot of time looking into this and I just can’t really figure it out.

  4. Bill Baer

    May 11, 2010 11:15 PM

    This stat claims to be blind to roles, but it seems to support closers.

    Right, as the game gets closer to the ninth inning, the odds of a team winning become more finite.

    Lidge, for example, will have a higher WPA if he has a perfect inning in the ninth than in the eighth.

    Where the stat shines is crediting non-closers for good and bad performances in crucial spots. If Jose Contreras comes in the seventh inning with the bases loaded and no outs with his team up 3-2 and he strikes out three consecutive batters, he will be credited appropriately. At present, there are no other ways to credit Contreras for such a performance.

    I think you actually get the idea behind the shutdown/meltdown but you may be analyzing it too much. It is not SIERA for relievers; it’s just a statistic that tells a story about whether or not a reliever helped his team’s chances of winning a game.

Next ArticlePhillies Stealing Signs? So What?