Breaking News: Chan Ho Park Is Good

This afternoon, I talked about the reasons why Derek Jeter laying down a bunt in the seventh inning was a good idea (until there were two strikes, anyway). One of the common responses to the support of the bunt in that situation is that Jeter should have been swinging away because, to paraphrase the argument, Chan Ho Park was pitching and Chan Ho Park stinks.

I’m sorry, but this is one of those cases where I just have to scratch my head and wonder how reputations can gain such traction when they are so remarkably wrong. The idea that Park is a bad pitcher, especially out of the bullpen, is downright crazy.

Over the last two years, Park has thrown 179 innings with a FIP of 3.90, good for a value of +1.8 wins. He’s done it by racking up 152 strikeouts while also posting an above average groundball rate, which is a classic recipe for success. Just based on that performance, we’d have to conclude that he’s something like a league average pitcher.

However, those numbers don’t reflect the actual level of Park’s effectiveness as a reliever. They also include his failures in the rotation, where he was unable to sustain his velocity and got pounded as a result. When moved back to the bullpen, his stuff improved by leaps and bounds, as you can see in his velocity chart below.

1267_P_FA_20090916

Park’s FIP as a relief pitcher this year? 2.10.

Now, a good chunk of that absurdly low mark is a 0.0% HR/FB rate that isn’t his actual talent level, but even when you adjust for that, he was still a lights out reliever this year, running a 3.25 K/BB rate and holding opposing hitters to a .231/.296/.280 line.

Additionally, Park is a right-handed pitcher with a significant platoon differential established over his entire career. RHBs have hit him at a .227/.311/.355 mark over his career, compared to .271/.368/.447 for LHBs. Jeter, being a right-handed hitter, was up against a right-handed relief pitcher who performs significantly better against same handed batters.

The idea that Jeter should have been swinging away because Chan Ho park was on the mound and a big rally was likely is the opposite of the truth. In reality, he’s a very good relief pitcher with the platoon advantage, and the match-up wasn’t likely to end well for Jeter.




Print This Post



Dave is a co-founder of USSMariner.com and contributes to the Wall Street Journal.

31 Responses to “Breaking News: Chan Ho Park Is Good”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Steve says:

    first impressions are hard to shake. and for many, their impression of Park is still the guy who signed for $65M with Texas and was a complete and total bust.

    Vote -1 Vote +1

    • Joel C. says:

      How is the Texas signing a first impression? He had already logged over 1250 innings of 109 ERA+ and a 84-58 record with the Dodgers over 7 seasons. I’d say that qualifies as very good over that period of time, and he’s been mediocre in the 650 innings since 2002, but that still works out to a “good” career.

      How is it his fault that Park’s agent told him that Tom Hicks was offering $55 million? Was he supposed to say “No thanks, I’ll take $35 million over 5 years instead since that’s what I think my true market value is.”

      Considering Park’s ’00 and ’01 seasons, a pitcher with similar stats and age this year would be offered around or more than $11 million even in a depressed economy so that contract offer wasn’t as foolish as it sounds at the time.

      Vote -1 Vote +1

  2. Richie Abernathy says:

    My argument is still this: Derek Jeter will be in the Hall of Fame someday, it ain’t gonna be for his glovework.

    Vote -1 Vote +1

  3. Sam says:

    Ok, makes some sense. Park is better than what he has been given credit for. But he hasn’t done particularly well with men on this year, whereas, with no one on, he has been lights out. I don’t know the starter/reliever splits on that one, but that is certainly something to keep in mind.

    Vote -1 Vote +1

  4. Logan says:

    This is why this site is pure win.

    Vote -1 Vote +1

  5. e poc says:

    Dave:
    Can you tell us what sample size is sufficient for us to determine that a pitcher has established a new level of effectiveness? For instance, a couple of months ago you and RJ were at pains to explain to us that 40 IP samples don’t mean anything for pitchers. But 180 IP, in the case of Park, tells us that Chan Ho Park is good now, even though he had been bad for years prior to 2008. At what point does the sample size change from meaningless to meaningful? Maybe you’re right in saying that Park is good now, but your argument doesn’t convince me because 180 IP doesn’t seem like a big enough sample size to determine that he’s good when there’s a larger body of evidence that suggests that he’s awful. The argument that he’s a very effective reliever because he’s able to sustain his velocity in a way that he’s not able to do as a starter – that’s convincing until you look at the splits and realize that he was actually much better as a starter than a reliever in ’08. Plus, as you mention, he’s gotten very lucky this year on hr/fb. I don’t know that there’s any way to know definitively whether Park is good or bad right now, but I think the best guess is that he’s a pretty bad pitcher who’s had a decent last 180 innings.

    Vote -1 Vote +1

    • Dave Cameron says:

      It depends on what you’re trying to measure. Groundball rate, for instance, stabilizes very quickly, so Joel Pineiro’s increase in GB% this year is far more likely to be carried over to future years than his absurdly low walk rate. The amount of contact allowed by opposing hitters also has little variation (relative to other pitching metrics), so if you see a guy start generating a lot more swinging strikes, there’s a decent chance that there’s been an underlying change in skill.

      In Park’s case, we do see several significant changes that support his better level of pitching – the change in roles and the subsequent increase in velocity. In general, we know that pitchers perform better as relievers than they do as starters, but some pitchers adapt better than others. Park’s stuff got a lot better after he moved to the bullpen, as his uptick in velocity shows.

      When velocity increases at the same time that strikeout rate increases, you can be much more confident that the improvement is real. When both of those are accompanied by a move to the bullpen, there’s a legitimate cause for the change in stuff, and we have reason to expect it to carry forward.

      180 innings is a small sample for something like ERA, which is dependent on a lot of variables not in the pitchers control. For things like strikeout rate and velocity, though, that’s more than sufficient to establish that there’s been a real underlying change.

      Given what we know about the general trend of starter-to-reliever conversions, Park’s velocity increase as a reliever, and the corresponding improvement in his strikeout rate when his stuff got better, the conclusion that he’s “a bad pitcher who has had a good 180 innings” is definitely wrong.

      Vote -1 Vote +1

      • e poc says:

        Park’s career k/9 is 7.8. As a starter last year it was 10.8, and as a reliever 6.3. That’s in 25 and 70 innings respectively. This year his k/9 as a starter was 5.7, and as a reliever 9.4. That’s 33 and 50 Ip respectively. So what does his increase in k rate in 50 ip this year tell us? I’d have to think the answer would be nothing. In fact, over the last two years combined, his k/9 has been 7.6 as a reliever and 7.9 as a starter, and both numbers are very close to his career number of 7.8. So the evidence does not support your claim that his k rate improved when he moved to the bullpen. At best, you could claim that he got better in the bullpen this year, but that’s only a 50 IP sample, and as you’ve pointed out before, anything can happen for a pitcher in such small samples. You say that his 3.9 FIP from the last two years doesn’t tell the full story of just how effective he’s been because he’s way more effective out of the bullpen. But that’s only true in a small sample from this year. Last year he was atrocious pitching out of the bullpen, and much better as a starter.

        Basically, your contention that his stuff improved in the bullpen is unfounded (unless speed of fastball is synonymous with “stuff”), and your contention that his k rate improved is quite simply wrong. Of course we should expect him to more effective, generally speaking, out of the bullpen than he is as a starter, but there’s no evidence particular to Park to suggest that that’s the case for him. Again, it seems like the “evidence” for him being good now is his last 180 innings, and based on your response hear, I’m going to have to stick to my opinion that he’s not as terrible as we thought in 2007, but that he is basically a bad pitcher who’s had 180 decent innings.

        Vote -1 Vote +1

    • vivaelpujols says:

      There is no magical point where the sample size grows large enough to where you can say the player has “changed”. Each record of his performance simply adds on to what you know about him.

      For example, when the Dodgers first signed Park last year, he was expected to be replacement level. That was our prior estimate of his ability in this case. Then he put up a 4.4 FIP with the Dodgers in 95 innings, with some of those coming in starts so we lower that number slightly to about 4.10 to get in on a reliever scale.

      The odds of a replacement level reliever posting a 4.10 FIP in 95 innings is pretty low, so we are forced to assume that he is better than our initial estimate of him, so between average and replacement level.

      Then he goes into 09, and puts up an equivalent FIP of about 3.00. The odds of a average-replacement level pitcher doing that are very low as well, so we adjust our expectations of him once more to where he is slightly above average.

      I disagree with Dave, that since he has “changed” (better velocity, more strike outs, whatever) it means we assume that his last 180 innings represents his true talent level. We have a whole bunch of history before that says the his talent level is worse than his performance the past 2 years. However, in light of his recent performance, we are forced to conclude that he most likely underperformed his true talent level before these past two years, just like he has most likely over performed his true talent level these past two years.

      The real answer is that he is somewhere in between. He’s not as bad as he was in Texas, but he’s not as good as he was in Philly and LA. That makes him about an average reliever, maybe a little below.

      Vote -1 Vote +1

      • Dave Cameron says:

        I don’t believe, and never said, that his last 180 innings represent his true talent level.

        Vote -1 Vote +1

      • vivaelpujols says:

        Sorry, I thought that was what you were implying.

        Vote -1 Vote +1

      • Wally says:

        viva, Most of that 95 IP last year with Dodgers came as a reliever. He only started 5 games. In relief work last year his FIP was 4.86. That’s over 70 IP. I’m not saying that’s his “true talent,” but Just adjusting his 4.40 FIP over the whole season down to 4.10 is obviously the wrong thing to do here.

        This year his K/9 and BB/9 rates are pretty much exactly the same as last year. The huge difference is in HR/9, where he’s given up about half his career average of HR/9. So while I don’t think he’s a bad pitcher, I don’t think he’s anywhere close to this 3.49 FIP due to that HR/9 difference, and he’s probably closer to his 2008 numbers in true talent.

        Also, doesn’t it strike anyone as pretty convient that Chan Ho Park went from a pretty awfull pitcher in the AL to roughly league average in the NL? Chan Ho Park’s FIP in the AL was 5.10….just saying…

        Vote -1 Vote +1

      • Wally says:

        Ok, I should have proof read that, please excuse my butchering of “convenient” among some of the other mistakes.

        Vote -1 Vote +1

  6. Ugueth says:

    Breaking news: Derek Jeter is a Hall of Fame caliber hitter. Chan Ho Park is not even close to a Hall of Fame caliber pitcher.

    Vote -1 Vote +1

    • Boomer says:

      This matters why? Does this mean that Cap’n Jetes should NEVER EVER sacrifice to get an insurance run unless its against a HOF pitcher?

      Vote -1 Vote +1

  7. Ugueth says:

    No it doesn’t mean that.

    Jjust like the idea that Chan Ho park is a good pitcher shouldn’t mean that Jeter should drop a bunt.

    Vote -1 Vote +1

    • Ugueth says:

      In other words, you could have easily responded to the original poster’s headline of “Breaking News: Chan Ho Park Is Good” with something like:

      “This matters why? Does this mean that Cap’n Jetes should NEVER EVER swing away to set up a big inning unless its against a poor pitcher?”

      The fact that Chan Ho Park is good isn’t a significant factor in determining whether or not Jeter should have bunted because guess what, Jeter is also good.

      Vote -1 Vote +1

  8. HH says:

    Why is it so hard to believe that Park could be effective out of the bullpen? There is hardly a closer in baseball that could survive as a starter. Almost every one of them would have an ERA over 5.00. Almost every bullpen great was a failed starter.

    Park still has a good fastball and can get the job done as a reliever for a couple of innings. That’s the case with almost every reliever. Put them in the rotation and its bombs away.

    Vote -1 Vote +1

  9. I’m wondering what the change is then. His career K/BB as a reliever is 2.04. Is there a reason we should trust the 50 IP as a reliever this year over 146 IP before this.

    Vote -1 Vote +1

    • I think you skewed the numbers a bit to make your point here. You pointed to his 2009 relief numbers and as I showed he hasn’t always been this good even as a reliever. Then you point to his career splits against righties, but you didn’t show his 2009 splits when he was actually worse against righties by K/BB (2.17 against righties and 2.27 against lefties)

      I’m assuming the 1.5 WAR this year is a fluke for someone with a total of 4.6 WAR in the previous 7 seasons. I still think we’re seeing replacement level pitching in Park.

      Vote -1 Vote +1

  10. bflaff says:

    The thing that is notable this year about Park as a reliever, which may have been true in other years as well (but I wasn’t paying attention) is that when he is on (and he spent much of the year in the ‘pen being on) he wasn’t just striking people out, or getting through innings — he was dominating. In his first appearance back from the DL in the NLCS his fastball was touching 95+ on the gun, and his breaking pitches were lethal. Not sure what explains his success as a reliever and his failure as a starter this year, but he clearly became the favored 7th inning guy by the end of the year. And if he hadn’t been injured in September, just as Manuel was losing faith in Lidge (and watching Madson struggle in the 9th as well), he might be the closer right now.

    That said, he’s not 100%, and his velocity was much less in his second NLCS appearance (game 2, when his results were notably worse), so the confidence in him is not as much right now as it was earlier in the year.

    Also worth mentioning that the perception on Park may be influenced by the CW on the Phillies pen, which says that it is crap.

    Vote -1 Vote +1

  11. a damn fog says:

    So your strategy here is to only point out talent levels that skew the averages when happens to suit your argument? Mariano Rivera and Chan Ho Park are effective pitchers, therefore it’s smart to bunt – absolutely no mention of the fact that Jeter is a very good hitter. Nor any mention of the advantages of resting Rivera, who is in the twilight of his career and has already thrown more two-inning saves this postseason than he probably should have in his entire career, if the Yankees are able to build a sizable lead.

    For a leading analyst on a statistics-oriented site, it’s downright embarrassing for you to add up run expectancies and act as if 2 + 2 = bunt without even considering the fact that the probability of successfully bunting is less than 100%. Especially embarrassing since Jeter even FAILED IN THE SITUATION AT ISSUE. And you can’t remedy that by saying that he shouldn’t have been bunting with two strikes: He easily could have popped up/bunted too hard on one of the first two, and even if he didn’t, he has still put two strikes in the count, and that absolutely affects his chances of ending the at-bat with a hit.

    You are wrong. It was not the right play. You are playing devil’s advocate for no reason other than to hear yourself talk.

    Vote -1 Vote +1

    • vivaelpujols says:

      No where in the article did Dave say that Jeter should have bunted because Park was a good pitcher. I honestly have no idea where you got that from.

      Vote -1 Vote +1

    • JLProck says:

      Dave isn’t saying that Jeter should’ve bunted because Park is good. He’s discounting those who said Jeter should’ve swung because Park sucks, hence the title of the article.

      Vote -1 Vote +1

  12. Jason says:

    Dave,

    You used to argue that FIP isn’t a good metric for relief pitchers. What has caused you to change your mind?

    Vote -1 Vote +1

    • Michael says:

      Why would FIP be bad for relievers? The only problem is that HR, which don’t stabilize as quickly as K’s and BB’s, can skew FIP by a lot in a small sample of innings. Dave accounted for that in the article (he certainly mentioned it).

      Vote -1 Vote +1

      • Jason says:

        **********Why would FIP be bad for relievers? ***********

        Dave has argued in the past it’s because starters and relievers belong to two separate cohorts…

        That’s why I asked Dave about the change of heart….

        Vote -1 Vote +1

      • Dave Cameron says:

        My comments about FIP have never been of the kind that they are “bad for relievers”. I have talked about the differences in BABIP and HR/FB rates for relievers and starters and how ideally we wouldn’t regress them to equal numbers, but it’s a minor issue.

        There’s nothing wrong with using FIP for relievers.

        Vote -1 Vote +1

      • Jason says:

        Thanks for the explanation.

        Vote -1 Vote +1

  13. PhD Brian says:

    Chan Ho Park was considered an “Ace” when he was a Dodger. And frankly he was! In Texas he sucked, but his style of pitching is not one that favors a hitters park. Anyone with a brain new he would probably get shelled in Texas. Old Texas stadium favored left handed hitters and Park has never gotten lefthanders out all that well. Since Texas, he has been very effective against right handed hitters and bad against left handed hitters. For his career, taken alone, his numbers against right handers are good enough for the hall of fame, but his numbers against left handers probably would have trouble succeeding in AA. Knowing that split, Park was almost a sure out against Jeter. Jeter was right to bunt.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

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=""> <strike> <strong>

Current day month ye@r *