• Phil Hughes ’10 No Better Than Joba Chamberlain ’09

    Posted by on August 20th, 2010 · Comments (29)

    Have you seen “The Three Faces Of Phil Hughes”? Check out these splits for Phil Hughes, to date, this season:

    His first 6 starts: 1.38 ERA (in 39 IP) and .165/.243/.203 BA/OBA/SLG allowed (148 BF)
    His next 9 starts: 5.56 ERA (in 55 IP) and .286/.329/.484 BA/OBA/SLG allowed (235 BF)
    His next 8 starts: 4.05 ERA (in 46.6 IP) and .261/.295/.418 BA/OBA/SLG allowed (193 BF)

    If this were a Western, it would be called “The Great, The Ugly and The Average.”

    What’s really interesting is when you compare Hughes, to date, against what Joba Chamberlain did in his first 23 starts of last season:

    Hughes: 23 GS, 140.6 IP, 3.90 ERA, .247/.296/.391 BA/OBA/SLG allowed (576 BF) – Yanks went 17-6 in his starts
    Chamberlain: 23 GS, 126.6 IP, 3.98 ERA, .257/.353/.415 BA/OBA/SLG allowed (560 BF) – Yanks went 16-7 in his starts

    Could those numbers be any closer?

    So, what do you think? Has Hughes, as a starter in 2010, been any better than what Chamberlain did, as a starter in 2009, when you look at them after their first 23 starts of the season?

    Comments on Phil Hughes ’10 No Better Than Joba Chamberlain ’09

    1. August 20th, 2010 | 2:57 pm

      Probably not, but what’s the point? Though, like most young starters, he has struggled with consistency, the Yankees seemingly foresee Hughes as a future starter. Is there anything more to be said than just that?

    2. Corey Italiano
      August 20th, 2010 | 3:00 pm

      The big key tell, to me, for the future is the OBP. Look at how many more free passes Joba gave vs. Phil. Eventually, that catches up to ya.

    3. August 20th, 2010 | 3:10 pm

      A 57 point gap in OBP means they are not that close at all. That’s a very large baserunner gap, as Corey noted.

    4. MJ Recanati
      August 20th, 2010 | 3:46 pm

      You could’ve also grouped them as follows:

      First six starts: 1.38 ERA (in 39 IP) and .165/.243/.203 BA/OBA/SLG allowed (148 BF)

      Next eleven starts: 5.51 ERA (in 67 IP) and .291/.333/.494 BA/OBA/SLG allowed (286 BF)

      Final six starts: 3.63 ERA (in 34.2 IP) and .243/.275/.375 BA/OBA/SLG allowed (142 BF)

      All this to show that, apart from periods of extreme success/ineffectiveness, Hughes has been a better-than-average #5 starter in baseball and, given his age, shows much promise for the years ahead.

    5. Corey Italiano
      August 20th, 2010 | 3:55 pm

      MJ Recanati wrote:

      Final six starts: 3.63 ERA (in 34.2 IP) and .243/.275/.375 BA/OBA/SLG allowed (142 BF)

      with a 23-6 K/BB ratio (It was in my post for yesterdays game)

    6. August 20th, 2010 | 4:04 pm

      Granted, yes, there’s a difference in the OBA allowed for Hughes and Joba here. But, bottom line, they each allowed the same earned run rate and the Yankees W/L record in their starts was just about the same. Therefore, in the net, Hughes has been no better for the Yankees, this year, than Joba was in his 1st 23 starts, last year.

      Does the OBA difference mean that Hughes will not regress the way Joba did? Perhaps. Time will tell.

    7. Evan3457
      August 20th, 2010 | 4:05 pm

      A .687 OPS against is the same as a .768 OPS against. Really?

      .687 is Gordon Beckham/Franklin Gutierrez, or more broadly, Erick Aybar, Kevin Kouzmanoff, Johnny Peralta and Kurt Suzuki. .768 is Alexi Ramirez/Adam Jones, or more broadly, Ty Wigginton, Hideki Matsui, Johnny Damon and Mike Cuddyer.

      Hughes is also going 1/2 inning deeper per start, and that’s a significant difference. Joba had all too many starts when he had trouble making it through 5. Hughes hasn’t had one start less than 5 innings. Joba had 4 starts of 7 innings or more; Phil’s had 8.

      Phil’s WPA for his 23 starts: +1.77; Joba’s WPA for his 23: -0.90 (Yanks went 16-7 in Joba’s starts; 17-6 in Hughes’ starts. Phil has had much better run support, 7.2 runs/game vs. 5.8 for Joba.)

      ============================================

      Baseball HQ has what it calls a Pure Quality Start rating system for breaking down each start into peripheral components. It awards up to 5 points, one point for each of the following:

      1. Completing 6 innings gets a point. (Any start of less than 5 innings gets 0 points, no matter what else the pitcher does; I think they do this because Baseball HQ is a fantasy site, and you have to go 5 innings to get a win.)
      2. Allowing no more than the same number of hits as innings pitched (i.e., 6 hits in 6 innings gets a point, 8 hits in 7.1 innings doesn’t)
      3. Allowing no more than 1 HR in a start gets a point.
      4. Striking out no fewer than 2 less batters than innings pitched gets a point (3 K’s in 5 innings gets a point; 5 K’s in 8 IP doesn’t)
      5. If the pitcher’s K/BB ratio in a start is 2/1 or better he gets a point.

      For example Hughes start on 4/21 vs. the A’s is an obvious 5: 7.1 IP, 3 H, 2 BB, 10 K, 0 HR. But so is yesterday’s start: 6-4-0-6-1. Joba’s start vs. the Tigers 4/29 last year is a 5: 7-3-3-6-0, but his start in Tampa on 7/29 is only a 4: 8-3-2-5-0.

      The purpose of PQS is not to set up bright line separation between types of starts on a one-game basis. Joba’s start vs. Tampa is clearly better than Hughes’ start yesterday, but doesn’t score as well. The purpose of PQS is to look at groups of starts in the aggregate, but also more especially to spot trends. That is to say, to see if a pitcher’s hit streak is driven by his peripherals; his true pitching ability, or if it’s driven by luck with balls in play, or keeping runners on base from scoring.

      PQS breaks the scores into three categories: Dominant (4 or 5), Disaster (0 or 1), Other (2 or 3).

      In Joba’s 23 starts last year, he had six 5’s, five 4’s, two 3’s, one 2, three 1’s, and six 0’s. That 11 dominant starts, and 9 disasters.

      In Hughes’ 23 starts this year, he has nine 5’s, five 4’s, three 3’s, three 2’s, two 1’s and one 0. That’s 14 dominant starts and 1 disaster.

      By this measure, PQS shows Hughes has had the much better year. Now, one could argue that PQS is biased against Chamberlain’s type of pitcher (high BB and relatively low HR) and in favor of Hughes’ type (low BB and relatively high HR). There is some truth in this. But the margins are clear. Hughes’ start have been dominant/comfortable/proficient more often that Joba.

      ===================================
      Let’s go back and run at this from a more personal direction. I took a look at Joba’s 23 starts and I asked myself a question…in how many starts did Joba pitch a “comfortable” game. That is, a good game, regardless of the score, where I felt he had reasonable control of the opposition hitters. I came up with 8 games out of 23. For Phil, I came up with 11 out of 23.

      Then I asked, OK, in how many starts did Joba pitch at least an OK game. Not so great but a presentable start. I came up with 11, including the eight above. For Phil, I got 14, including the 11 above.

      ====================================
      In short, though the superficial numbers look the same and in fact, Joba may have had a lower OPS+ for his 1st 23 starts this year than Phil has had in his 23, I draw the conclusion that, on a start by start basis, Hughes has been the better pitcher by a moderate but measurable amount.

    8. August 20th, 2010 | 4:05 pm

      MJ Recanati wrote:

      Hughes has been a better-than-average #5 starter in baseball and, given his age, shows much promise for the years ahead.

      Could you have not said the same exact thing about Joba after his 12st 23 starts last year? If not, why? Based on what metrics?

    9. SG
      August 20th, 2010 | 4:11 pm

      Steve, the problem with this comparison is that ERA is subject to a lot of things that are not in the pitcher’s control. A more apt comparison is this:

      140.7 IP, 132 H, 62 R, 61 ER, 18 HR, 0 HBP, 38 BB, 116 K, 4.02 FIP
      126.3 IP, 123 H, 66 R, 55 ER, 16HR, 11 HBP, 62 BB, 109 K, 4.85 FIP

      The point is, Joba’s peripherals over his first 23 starts did not support his ERA, and as the season continued that became plainly evident as his ERA regressed towards his FIP.

      Hughes’s peripherals are directly in line with his YTD performance, and he’s shown much better command than Joba did in 2009.

      A .80 runs per nine gap is huge. Over 200 innings, it’s around 18 runs.

      You just can’t make the claim that Hughes has been no better than Chamberlain.

    10. Evan3457
      August 20th, 2010 | 4:12 pm

      SG wrote:

      140.7 IP, 132 H, 62 R, 61 ER, 18 HR, 0 HBP, 38 BB, 116 K, 4.02 FIP
      126.3 IP, 123 H, 66 R, 55 ER, 16HR, 11 HBP, 62 BB, 109 K, 4.85 FIP
      The point is, Joba’s peripherals over his first 23 starts did not support his ERA, and as the season continued that became plainly evident as his ERA regressed towards his FIP.

      Dang. Shoulda thought of comparing the FIP myself. That says it all in far less time than I did. Thanks.

    11. Evan3457
      August 20th, 2010 | 4:14 pm

      Evan3457 wrote:

      A .687 OPS against is the same as a .768 OPS against. Really?
      .687 is Gordon Beckham/Franklin Gutierrez, or more broadly, Erick Aybar, Kevin Kouzmanoff, Johnny Peralta and Kurt Suzuki. .768 is Alexi Ramirez/Adam Jones, or more broadly, Ty Wigginton, Hideki Matsui, Johnny Damon and Mike Cuddyer.

      I also shoulda used teams instead of players to make this clearer.

      A .687 OPS is the Oakland A’s.
      A .768 OPS is the Toronto Blue Jays.

      Any questions?

    12. Corey Italiano
      August 20th, 2010 | 4:17 pm

      Steve Lombardi wrote:

      Could you have not said the same exact thing about Joba after his 12st 23 starts last year?

      I would say that Joba was an acceptable 5th starter in 09. That’s why I was mad that he was banished to the pen this year in April.

    13. Garcia
      August 20th, 2010 | 4:27 pm

      Evan makes some excellent points, I knew they weren’t the same – just from what I’ve seen with my own two eyes – and Evan did an excellent job of validating my suspicion.

    14. Garcia
      August 20th, 2010 | 4:30 pm

      @Corey I’d like to see Joba banished from the team and baseball altogether.

    15. Corey Italiano
      August 20th, 2010 | 4:35 pm

      @ Garcia:
      Well sure, now.

    16. Raf
      August 20th, 2010 | 5:22 pm

      Joba has pretty strong peripherals, I wouldn’t give up on him just yet.

    17. MJ Recanati
      August 20th, 2010 | 5:31 pm

      SG wrote:

      Steve, the problem with this comparison is that ERA is subject to a lot of things that are not in the pitcher’s control. A more apt comparison is this:
      140.7 IP, 132 H, 62 R, 61 ER, 18 HR, 0 HBP, 38 BB, 116 K, 4.02 FIP
      126.3 IP, 123 H, 66 R, 55 ER, 16HR, 11 HBP, 62 BB, 109 K, 4.85 FIP
      The point is, Joba’s peripherals over his first 23 starts did not support his ERA, and as the season continued that became plainly evident as his ERA regressed towards his FIP.
      Hughes’s peripherals are directly in line with his YTD performance, and he’s shown much better command than Joba did in 2009.
      A .80 runs per nine gap is huge. Over 200 innings, it’s around 18 runs.
      You just can’t make the claim that Hughes has been no better than Chamberlain.

      Excellent point and very well put.

    18. MJ Recanati
      August 20th, 2010 | 5:32 pm

      Garcia wrote:

      I’d like to see Joba banished from the team

      That might be one of the first times Garcia and I agree on anything. I don’t care where he goes, I just don’t want him on the Yankees.

    19. Evan3457
      August 20th, 2010 | 6:30 pm

      Raf wrote:

      Joba has pretty strong peripherals, I wouldn’t give up on him just yet.

      I keep hoping, almost certainly in vain, that they make him the #5 starter next year. But that ship has sailed.

    20. Raf
      August 20th, 2010 | 6:44 pm

      Evan3457 wrote:

      I keep hoping, almost certainly in vain, that they make him the #5 starter next year. But that ship has sailed.

      I dunno about that, it’s possible that they return him to the rotation next season.

    21. MJ Recanati
      August 20th, 2010 | 7:43 pm

      Raf wrote:

      I dunno about that, it’s possible that they return him to the rotation next season.

      If they did, they should all be fired. You can’t turn a starter into a reliever, then a starter, then a reliever, then a starter, then watch him flounder, then limit his innings to where he starts once every eight days for only three innings, then turn him into a reliever, then have him prepare for the season as a starter only to turn him into a reliever again AND THEN say that he’ll be a starter again next year.

      The decision to turn him into a reliever was wrong but it would be just as wrong to continue this schizophrenia.

    22. August 20th, 2010 | 7:47 pm

      SG wrote:

      Steve, the problem with this comparison is that ERA is subject to a lot of things that are not in the pitcher’s control. A more apt comparison is this:140.7 IP, 132 H, 62 R, 61 ER, 18 HR, 0 HBP, 38 BB, 116 K, 4.02 FIP
      126.3 IP, 123 H, 66 R, 55 ER, 16HR, 11 HBP, 62 BB, 109 K, 4.85 FIPThe point is, Joba’s peripherals over his first 23 starts did not support his ERA, and as the season continued that became plainly evident as his ERA regressed towards his FIP.Hughes’s peripherals are directly in line with his YTD performance, and he’s shown much better command than Joba did in 2009. A .80 runs per nine gap is huge. Over 200 innings, it’s around 18 runs.You just can’t make the claim that Hughes has been no better than Chamberlain.

      Take 162 games and divide them by five. That’s 32 GS per SP. So, those 18 runs, over 200 IP, breaks down to about a 1/2 run per game. Is a 1/2 run per game, all that much huge, given the Yankees offense? Probably not – considering that the Yankees posted just about the same W/L record with Joba, last year, in his first 23 games, as they did with Hughes this year.

      So, to the question of “Has Hughes, as a starter in 2010, been any better than what Chamberlain did, as a starter in 2009, when you look at them after their first 23 starts of the season?”…

      If you want to say Hughes, with his inside numbers, pitched a 1/2 run better per game, that’s fine. But, in the bottom line, in terms of how each benefited the Yankees, it’s six of one, half dozen of the others.

      Plus, one thing to consider if you want to use the inside numbers, when I try to explain to some that the Yankees W/L record is inflated by their schedule – and getting to beat up on the O’s, Indians, Astros and D’backs – basically looking at their inside numbers – many come back to me and say “A win is a win, not matter who it comes against.”

      So, if that’s true, if a “W” is the thing, no matter what you did, who it came against, etc., how can we ignore that the Yanks went 17-6 in Hughes starts and went 16-7 in Joba’s? Again, bottom line, Hughes in the rotation this year was no better than Joba, at the 23 start mark, in terms of the net result.

    23. MJ Recanati
      August 20th, 2010 | 8:12 pm

      @ Steve Lombardi:
      Not that your methodology makes any sense whatsoever but, yes, 0.5 runs per game is clearly a difference.

      For instance, it’s approximately the difference (on a per-game basis) between the Yankees league-leading offense (5.32 R/G) and the White Sox, the sixth-most prolific offense in the AL at 4.67 R/G. There’s a 78-run difference between the two teams through 121 games which translates to about a 0.65 run differential.

      How can you argue that 0.5 runs is an insignificant difference between Joba ’09 and Phil ’10 (assuming that your methodology even makes sense, which I dispute)?

      In any case, do you not believe in FIP or OBA or OPS? By arguing the way you’re arguing, it seems that you do not.

    24. MJ Recanati
      August 20th, 2010 | 8:15 pm

      Steve Lombardi wrote:

      if a “W” is the thing, no matter what you did, who it came against, etc., how can we ignore that the Yanks went 17-6 in Hughes starts and went 16-7 in Joba’s?

      By that logic, Carl Pavano was a great signing because he was 18-8 in 2004 and his team was 22-9 in his 31 starts.

      I fail to see the connection on how two pitchers can be the same just because their records (and their team’s record in games they start) are the same.

    25. August 20th, 2010 | 9:49 pm

      MJ Recanati wrote:

      In any case, do you not believe in FIP or OBA or OPS? By arguing the way you’re arguing, it seems that you do not.

      I’m in favor of looking at component stats, because they do sometimes tell a different story than the conventional stats. That said, at that the end of the day, it’s about reality. That means real runs allowed to score, real wins, etc. Last time I checked, a team never won a game because they had a higher expected batting average than the other team or saved a game because their BABIP was lower than the other team.

    26. Evan3457
      August 20th, 2010 | 10:15 pm

      Steve Lombardi wrote:

      MJ Recanati wrote:
      In any case, do you not believe in FIP or OBA or OPS? By arguing the way you’re arguing, it seems that you do not.
      I’m in favor of looking at component stats, because they do sometimes tell a different story than the conventional stats. That said, at that the end of the day, it’s about reality. That means real runs allowed to score, real wins, etc. Last time I checked, a team never won a game because they had a higher expected batting average than the other team or saved a game because their BABIP was lower than the other team.

      O.K., then, if you’re looking at results as results, then, yes, Hughes’ results so far this season are not very much better than Joba’s 1st 23 starts last season.

      However, if you’re trying to check true effectiveness, then Hughes has been better by a noticeable margin, and projects better down the road.

    27. Raf
      August 21st, 2010 | 10:27 am

      MJ Recanati wrote:

      You can’t turn a starter into a reliever, then a starter, then a reliever, then a starter, then watch him flounder, then limit his innings to where he starts once every eight days for only three innings, then turn him into a reliever, then have him prepare for the season as a starter only to turn him into a reliever again AND THEN say that he’ll be a starter again next year.

      It can be done, it’s a matter of preparation. Chamberlain isn’t the first pitcher to bounce between the rotation and the pen.

    28. MJ Recanati
      August 21st, 2010 | 3:00 pm

      Raf wrote:

      It can be done, it’s a matter of preparation. Chamberlain isn’t the first pitcher to bounce between the rotation and the pen.

      I’m sure it has been done but I’m not sure it’s been handled as poorly in those cases as it has been handled in this case. He’s barely a good enough pitcher to be in the big leagues as a reliever, I don’t see how going back to square one with him as a starter will help matters. Mistakes were made; let’s not make them even worse.

    29. SG
      August 21st, 2010 | 5:35 pm

      Steve Lombardi wrote:

      MJ Recanati wrote:
      In any case, do you not believe in FIP or OBA or OPS? By arguing the way you’re arguing, it seems that you do not.
      I’m in favor of looking at component stats, because they do sometimes tell a different story than the conventional stats. That said, at that the end of the day, it’s about reality. That means real runs allowed to score, real wins, etc. Last time I checked, a team never won a game because they had a higher expected batting average than the other team or saved a game because their BABIP was lower than the other team.

      If you want to say that Hughes has been no more valuable than Joba was over their first 23 starts in 2010/2009 respectively, I wouldn’t disagree.

      But when you use the word better, you are not necessarily just talking about value, you are also at least alluding to ability. And if you are looking at ability, the fact that Hughes has pitched much better than Chamberlain over the subset you’re looking at means that Hughes’s performance is far more likely to be sustainable.

      If you don’t think the fact that Hughes has pitched more innings per start, has allowed 26 fewer base runners over 14 more innings while striking out 7 more people shows a significant difference in how much better Hughes has actually pitched, I’d love to make a bet with you that Hughes will pitch closer to how he’s pitched this year than how Joba pitched over the rest of 2009. If I win, you can pay for my Baseball Reference Phil Hughes sponsorship.

      So my quibble is with your use of the word better. If you weren’t trying to imply that Hughes is going to end the year by tanking just like Joba did in 2009, I have no qualms.

    Leave a reply

    You must be logged in to post a comment.