The Overlooked Bullpen

Rob Bradford of WEEI posted his Major League bullpen power rankings this morning. Naturally, the first thing I did was a Ctrl + F for “Phillies”. Much to my surprise, they were in the top one-third, but the more I thought about it, the more it made sense. Lost in the talk about the Phillies’ four aces and how many innings they ought to accrue, the bullpen is actually quite formidable.

Faithful readers of the blog know how highly I think of Ryan Madson, but even outside of him, there are quality pitchers capable of getting outs reliably. But let’s start with Madson and get him out of the way.

Madson is awesome. It aggravated me to no end to see him shunned for an apparent inability to close games despite never really getting a fair shot. He was not given the benefit of the doubt when a small sample size can produce a wide variety of results irrespective of skill. During the 2010 season, he posted a 10.9 K/9 and a 2.2 BB/9. His 4.9 strikeout-to-walk ratio was ninth-best in the Majors among relief pitchers, sandwiched by Luke Gregerson and Billy Wagner, two very highly-regarded bullpen arms. And among relievers with at least 50 innings in Phillies franchise history, Madson joined Doug Jones as the only ones to post a 4.9 or better strikeout-to-walk ratio. His 2.49 SIERA was 12th-best in baseball. His change-up is arguably the best in the Majors — even better than Cole Hamels‘. If there’s a Phillies reliever you don’t want to face, it’s Madson.

How about Brad Lidge? He is cast aside as a reliever at the end of his career with not much left to offer. As I mentioned yesterday, the Phillies need to decide on what to do with the closer situation, and it likely won’t include Lidge going forward. Still, he is a well above-average reliever. His 3.55 SIERA was right there with highly-respected closer Andrew Bailey and a tenth of a run behind Mariano Rivera.

Lidge’s 2010 season turned around after Ryan Zimmerman of the Washington Nationals hit a walk-off home run on July 31. To that point, Lidge had a 5.57 ERA with a BB/9 of an even 6.0. After the game, Zimmerman hinted that Lidge was tipping his pitches. From August 1 through the rest of the season, Lidge corrected his flaws and posted a 0.73 ERA with a 3.6 BB/9. Baseball is a game of constant adjustments. Lidge appeared to make the correct adjustments in the final two months last year, and he can only learn from that and benefit going forward. He is not by any means a washed up reliever; in fact, he is still among the most feared in baseball.

Jose Contreras finished 2010 with a 3.19 SIERA, certainly a great accomplishment as it was his first full season as a reliever. He averaged just over a strikeout per inning and was stingy with the free passes as his 2.5 BB/9 indicated. He even induced a good amount of ground balls (45 percent) to help limit the damage. Because he is not in a glorified position with its own title, such as “set-up” or “closer”, Contreras is assumed to be just another reliever, but he is good enough to close on quite a few teams. He may have worn down towards the end of the season as his second-half ERA was more than a full run higher than his first-half ERA, but now that he is going into his second full season as a reliever, he can make the proper adjustments in terms of preparation and conditioning.

J.C. Romero is responsible for the poorly-groomed fingernails of many a Philadelphian, but I insist it’s not his fault. As mentioned various times here on the blog, Romero should be used strictly against left-handed hitters. His xFIP against lefties is a paltry 3.61, but against right-handers, it rises to an unacceptable 5.34. Charlie Manuel has limited Romero’s use to lefties more and more, but never entirely. It would behoove Ruben Amaro to demand that Romero never be used against a right-handed hitter in a meaningful situation. When used properly, Romero is a remarkably effective lefty.

Antonio Bastardo will likely be the team’s second lefty out of the bullpen. He seems to have been given this reputation as a good-but-not-quite-good-enough reliever, but in limited action, he posted a 2.90 SIERA. He can attribute that to an incredible 12.5 K/9 (in about 19 innings), which is not a fluke. Over his Minor League career, he posted a 10.3 K/9. Bastardo’s goal should be to limit his walks, but he is right now an MLB-capable reliever. Consider that his 4.3 BB/9 last year is comparable to Lidge’s 4.1 career average. Those walk rates are not good by any means, but you accept it with the amount of swings-and-misses they induce.

The bullpen will round out with a slop-throwing long reliever (probably Kyle Kendrick) and perhaps a young arm. Scott Mathieson is one such candidate whom I feel hasn’t been given a fair shot at keeping a job as a Major League reliever. He had all of one and two-thirds innings in 2010 including one appearance in mid-June in which he struggled and was immediately jettisoned to Triple-A Lehigh Valley. The kid, fully recovered from Tommy John surgery, has done just about everything he can possibly do in the Minors — he’s thrown 566 innings in the Minors with a 9.0 K/9, 2.9 BB/9 and 3.82 ERA — so it’s time to use him or lose him.

Another name to keep in mind is Justin De Fratus (he’s on Twitter!). The 23-year-old has been nothing short of impressive in his brief professional career. Last year, between Single-A Clearwater and Double-A Reading, he posted a 1.94 ERA with a 9.8 K/9 and 2.2 BB/9. ESPN’s Jason Grey is very high on De Fratus as well:

De Fratus absolutely looks like a big league bullpen arm, with a 92-95 mph fastball that touched 96 mph once when I saw him, and good tilt on an 83-85 mph slider that can get strikeouts. A strong pitcher with a good frame who uses his lower half well, De Fratus does a good job finishing his pitches. De Fratus isn’t far from a call-up and looks like he could fill a seventh- or eighth-inning role at the big league level, and from there anything can happen. Then again, he’s also in an organization that left Scott Mathieson behind to dominate Triple-A hitters all year.

The Phillies are expected to have a formidable offense and baseball’s best starting rotation, but don’t forget about the bullpen. Their role may have been condensed, but they still have a say on exactly how successful the Phillies will be in the regular season and how far they will advance in the playoffs.

Leave a Reply



  1. Matty B.

    February 02, 2011 11:02 AM

    Given that Mathieson is out of options, you have to figure he’ll get a fair chance at making the roster out of camp this time around.

  2. Bill Baer

    February 02, 2011 11:06 AM

    Yup, the Phillies absolutely have to make use of him or risk losing him to another organization. Can’t imagine a reason why they can’t find a use for him especially when they were giving bullpen innings to David Herndon and seem content with using Kyle Kendrick out of the bullpen this year.

    I’d much rather lose a Herndon to Rule 5 (which won’t be the case this year) or keep Kendrick stowed away in Triple-A than lose Mathieson.

  3. Matty B.

    February 02, 2011 11:36 AM

    They’re paying Kendrick a ton of money, so I expect he will be with the team. The only realistic alternative would be if they move Blanton before Opening Day — which appears unlikely — and give Kendrick the #5 job.

    Lidge, Madson, Contreras, Romero, Kendrick . . . that’s five virtual locks for the bullpen already.

    That leaves two spots for Baez, Bastardo, and Mathieson.

    Baez looks likely to be in the bullpen, considering his high salary.

    So really, we’re looking at one spot for either Bastardo or Mathieson.

    They usually like to carry two lefties in the bullpen, so Bastardo has the inside track on that spot.

    Mathieson might be left hoping for an injury or Baez pitching terribly in the spring. I’d say there’s a decent chance at least one of those scenarios plays out, and he makes the roster.

  4. Bill Baer

    February 02, 2011 11:39 AM

    I wouldn’t be surprised to see Baez cut and the Phillies absorb his salary. They’ve shown in the past the willingness to cut dead weight with Adam Eaton and Geoff Jenkins.

  5. Aaron H

    February 02, 2011 11:42 AM

    I wish RAJ (and other GMs, for that matter) read some basic economic texts about “sunk costs.” The logic that you have to pitch Baez due to his salary is faulty-you’re already paying the guy no matter what happens, so his salary shouldn’t impact the evaluation of his contribution (or lack thereof) to the team.

  6. Dan P

    February 02, 2011 11:53 AM

    Don’t forget about Matt Anderson!

  7. awh

    February 02, 2011 12:16 PM

    I agree on Baez. He’s a sunk cost. He also has a history of up then down years. If he has a good ST he might even be traded to a team that would pick up more than the MLB minimum of his salary – saving them the trouble of cutting him.

    Tampa Bay, where he was a successful closer for 2 years, just might be willing. They’re going with a closer by committee this season.

    OTOH, the signing of Cliff Lee indicates the FO is going “all-in” while thise core group of players can still produce. If Baez pitches well in ST they just may elect to keep him.

    Bill, does Vance Worley, who you didn’t mention, have a chance to make the team?

  8. Bill Baer

    February 02, 2011 12:22 PM

    I highly doubt it, unless the Phillies trade Joe Blanton and are supremely underwhelmed by Kyle Kendrick (and if they haven’t been already, they’re not likely to any time in the future). Worley will be the insurance policy behind Kendrick in the event a starter gets injured.

  9. sean

    February 02, 2011 01:02 PM

    @awh i highly doubt tampa is looking at relievers like baez most likely they would be using jake mcgee as their closer or jp howell. If he has a good spring training then why trade him, use him!

    the thing i have with Mathieson when he was called up last year the velocity of his fastball was impressive (95.2), but it was straight (at least to my eye, and i can’t tell looking at the pitch f/x) and he didn’t have command of any secondary stuff. The twins just timed his fastball and he was toast after that. he needs another pitch to be able to survive

  10. Sloth

    February 02, 2011 01:41 PM

    Kyle Kendrick, at this point, is like that itch that you need to scratch…but you just can’t quite get it. He shouldn’t be given a bullpen spot over Mathieson by any means, Kendrick does not have a consistent “out” pitch and at this point that simply isn’t cutting it, whether a starter or reliever.

    Mathieson on the other hand is a perfect example of what you want in the pen’, a hard thrower, and it’s not as if he couldn’t be the long-man on this team since he will be rarely used in that role, and the longest he will go is probably 3 or 4 innings.

  11. hk

    February 02, 2011 02:03 PM


    You’re right. He should have been non-tendered. Unfortunately, since the team not only offered him arbitration, but settled at $2.45M for the upcoming season, it seems likely that you’re going to have that itch for at least one more season. We can only hope that the starters will go 6 or more IP most nights and that the 7th through the 9th will be left to Romero vs. LHB’s, Contreras, Madson and Lidge and that KK will only pitch in low leverage situations.

  12. Sloth

    February 02, 2011 03:32 PM


    Yeah when I found out that they gave him 2.45 million I was extremely annoyed, Mathieson, De Fratus and Worley could get the job done, a better job no less, for a much cheaper price.

  13. hk

    February 02, 2011 03:54 PM

    It seems as though bullpen construction and spending is one area where some use of sabremetric statistics could really benefit the Phils.

  14. Bill Baer

    February 02, 2011 04:04 PM

    There was a discussion a while ago the Phillies being Saber-savvy. The consensus seemed to be that they’re not, but I found it odd that the team’s success has coincided with the application of Sabermetric concepts such as prioritizing on-base percentage (led the NL in 2007) and not getting thrown out needlessly on the bases.

    But hk brings up an interesting point: the Phillies have not managed the bullpen well despite their success. Charlie Manuel is by no means a Saberist, but it strikes me as odd that no one in the Phillies’ front office noticed J.C. Romero’s splits and passed it on to the powers that be.

    There are other instances where Manuel uses his better relievers in lower-leverage situations as well.

    So, I do agree that applying some Sabermetric principles could make the bullpen even more potent.

  15. hk

    February 02, 2011 04:28 PM


    I’ve often wondered about the practicality of using your best relievers in the highest leverage situations. While I agree that it makes sense to do this in theory, baseball’s obviously not like basketball where you bring someone off the bench and he’s immediately ready to play, so I wonder how many warm-up pitches a team would have their best relievers waste in the bullpen during a season to actually practice this. For instance, every time the Phils are up by 2 runs and maybe even 3 runs after 6 innings, Charlie would have to get Madson up in the pen at the beginning of the 7th to have him ready in case the other team gets a man or two on base. Some times, the Phils starter will pitch a 1-2-3 inning and the Phils will tack on a couple of runs in the bottom of the 7th, enabling Charlie to get through the game without using Madson at all. In those cases, Charlie will have had him (unnecessarily in hindsight) throw those warm-up pitches. Over the course of a regular season, would / could situations like this add up to wasting too many of your best reliever’s innings in the bullpen?

    On the other hand, I totally agree with you that Charlie wastes a lot of Madson’s pitches in the bullpen and on the mound protecting leads (like 4-0 in the 9th vs. the Mets) that Jose Contreras or even Chad Durbin should be able to protect.

  16. Bill Baer

    February 02, 2011 04:58 PM

    I don’t know that teams should be trying to precisely implement their relievers in the situation you depict, but in general, you can see these situations coming.

    For instance, the Phillies and Cardinals are facing off in Philadelphia, Cliff Lee vs. Adam Wainwright. It’s scoreless through six. Lee starts to tire in the seventh, walking the lead-off guy on five pitches, missing his spots badly. I think at that point, you make the call to the bullpen to get a couple guys up, including Madson.

    At that point, you should have a contingency plan: if Lee allows another base runner (not via home run), you bring in Madson; otherwise, you bring in the other guy (e.g. J.C. Romero or Jose Contreras).

    The Phillies’ win probability (assuming both teams are of equivalent talent) with a tie game with runners on first and second and no outs in the top of the seventh is 36 percent. With a runner on first and one out in the top of the seventh, they’re at about 50 percent.

    EDIT: If anyone is wondering, I got the win probability numbers from Tom Tango’s “The Book“.

  17. Aaron H

    February 02, 2011 05:00 PM

    Kendrick actually seems to be an even bigger case for a platoon role than JC Romero-maybe even a ROOGY would be the right word? Check out his career splits against RHB vs. LHB:

    RHB: 3.00 K/BB ratio, OBP .303, SLG .397, OPS .700 (granted, this is with a .273 BABIP)
    LHB: 0.88 K/BB ratio, .382 OBP, .531 SLG (!), .913 OPS (with a .317 BABIP).

    Righties are hitting like Wilson Valdez (should of traded Rolinss) and lefites like Joey Votto off of him. You rarely have a guy used exclusively against RHB, but if they are going to keep KK in the pen, this seems like the optimal way to use him.

  18. Bill Baer

    February 02, 2011 05:03 PM

    Very good point, Aaron. To add to your point, Kendrick has a career 4.18 xFIP against right-handers and a 5.54 xFIP against lefties.

  19. Scott G

    February 02, 2011 05:48 PM

    One of my biggest complaints is always the mismanagement of the bullpen. I think this is the perfect time to suggest that Bill Baer approach the FO or someone within the organization to make the plea for utilizing relievers based on the severity (leverage) of the situation.

    I am entirely serious about this if you have any real way of making this come to pass. Please make this happen. I can deal with Ryan Madson blowing games. I don’t know if I can deal with Ryan Madson pitching in 3-4 run games, however.

  20. Bill Baer

    February 02, 2011 05:55 PM

    Well, one time when I had a press pass, I made eye contact with Ruben Amaro when I was walking to the press box. So maybe he’ll remember that magical split second and heed my advice.

  21. Scott G

    February 02, 2011 05:59 PM

    Oh yeah, and if Romero unjustifiably pitches to one more RHB or switch hitter, I might cry.

  22. hk

    February 02, 2011 06:05 PM


    I admittedly have never pitched at a high level and I don’t know how many warm-up throws it takes for a pitcher like Madson to be ready. Does 20 sound reasonable? In the hypothetical scenario you cite, there’s a good chance that Madson will warm up and not get into the game. If this happens 20 or 25 times during the season, he may leave 400 to 500 pitches in the bullpen. This seems like a lot for a guy who throws about 1,200 game pitches during an entire season.

    It is possible that, by Charlie just giving Madson the 8th inning during the regular season, he actually keeps him fresher later in the season, which has benefits in September and October.

  23. Bill Baer

    February 02, 2011 06:43 PM

    I’ve never pitched at a high level either (though Dickie Noles was my coach for a few years), so I can’t speak from experience more than anyone.

    I can’t reject your hypothesis either. Perhaps during the season, we can keep track of pitchers who warm up and never get into the game, then see what happens.

  24. hk

    February 02, 2011 06:46 PM

    Really, I never would have thought that Dickie Noles would go into coaching.

  25. Bill Baer

    February 02, 2011 06:50 PM

    He was actually a really good coach. I was a terrible hitter, but he helped me raise my LD% from 0% to 1%, which is quite a feat. And he scared me away from throwing curve balls at too young of an age, for which I thank him.

    I believe he is still does drug and alcohol counseling for the Phillies, which is a different form of coaching, I guess.

  26. Css228

    February 03, 2011 01:29 AM

    @aaron amen on raj but hopefully not other GMs

  27. MG

    February 03, 2011 10:36 AM

    Please tell me how a middle-of-the-road bullpen which didn’t make any substantial upgrades this offseason (only lost Durbin) and likely doesn’t have much help for youngsters this season going to be substantially improved even if they just have to give the Phils 390-400 IP this year?

  28. MG

    February 03, 2011 12:17 PM

    How are you defining ‘elite’? A sole metric, a series of metrics, just in the NL? To me that means this unit finishes in the top 2-3 in the NL in just about every major pitching category among bullpens.

    By no objective measure were they ‘elite’ last year. ERA, WHIP, K/BB, K/9, BAA, OPS, % SV PCT were all just about middle of the pack in the NL or slightly below average.

  29. MG

    February 03, 2011 12:18 PM

    How are you defining ‘elite’? A sole metric, a series of metrics, just in the NL? To me that means this unit finishes in the top 2-3 in the NL in just about every major pitching category among bullpens.

    By no objective measure were they ‘elite’ last year. ERA, WHIP, K/BB, K/9, BAA, OPS, % SV PCT were all just about middle of the pack in the NL or slightly below average. Nor will they this year with almost the entire same cast of characters.

  30. hk

    February 03, 2011 01:00 PM

    The pen was 10th among MLB pens in FIP and 9th in xFIP despite not having their best reliever (Madson) for about 1/3 of the season. It seems to me that elite might be an overstatement, but no more so than middle of the road is an understatement.

  31. MG

    February 03, 2011 01:17 PM

    Baer – It selectively lists individual stats here or there but nothing that looks at aggregrate performance vs. competitors.

    hk – Lots of teams have injuries. Your point about FIP and xFIP is a good point that this bullpen might have been above average. I guess I would take 10-15 aggregrate measurements and boil them down via ranking to see bullpens rank. Number and actual criteria would obviously be up for discussion. I could see omitting something like % SV or ERA possibly.

  32. sean

    February 03, 2011 01:28 PM

    MG then what is your definition of an elite bullpen so we can take it to task?

    also kendrick’s contract at 2.45 million isn’t guaranteed, he can be designated for assignment like galarraga and traded but i doubtt he phillies would do such a thing since they need depth just in case and kendrick fits that bill nicely.

    also phillies are heavy on scouting and contract with baseball info solutions for other things.

  33. hk

    February 03, 2011 01:48 PM

    MG: You are right that lots of teams have bullpen injuries, but there were very few relievers with a better FIP than Madson who missed as much time as he missed. Also, when you consider that the Phillies bullpen threw the fewest IP (421) of all MLB pens last year, the impact of losing 25 IP from Madson had an even greater impact on their stats than the impact of losing 25 IP from a top reliever would have on the stats of a pen that threw 500 IP.

  34. awh

    February 03, 2011 01:59 PM

    MG, I was going to respond over at BL, but I’ll do it here:

    First, you raise a good point as to definition of “elite”. You are entitled to your definition. However, my definiton of “elite” is top 25%.

    So, if the bullpen power rankings Bill linked to are accurate (and, of course, you’re free to dispute them), then they would qualify as elite.

    However, I think the error in your analysis is an underlying assumption of mediocrity from certain pitchers, most notably Brad Lidge. You raised an excellent point on BL about 1st vs. 2nd half Lidge. However, I don’t think Lidge needs to be as good as he was for the 2nd half in order to be effective. Somewhere in between and he’ll still be just fine – and feared.

    And, frankly, you’re “analysis” of Lidge really is just speculation, not analysis. Sure, teams could catch up to him by making an adjustment to his throwing 60-70% sliders, but didn’t they have at least a month to do that last season? It was apparent by the 3rd week of August Lidge had changed his approach. The fact is, we learned last season how filthy his slider really is. He threw it 60-70% of the time, hitters knew it was coming, and they still couldn’t do anything with it. And he wouldn’t be the first relief pitcher, or closer, to rely on one pitch.

    Also, Madson “should” be around for the whole season, assuming his teammates keep him away from folding chairs.

    So you disagree with Bill that Contreras can pitch well for a whole season, now that he is more familiar with the relief role and pacing himself. OK, you disagree. I guess you don’t think Contreras is capable of learning anything from last season.

    Also, your focus exclusively on last season makes me want to scream “SAMPLE SIZE”, especially where relievers are concerned. Should we toss out the results these relievers have had in past seasons, not just 2010? Suppose Baez has a bounce back year. He’ s been up and down before. Why not now?

    Also, the addition of Lee and the rotation depth probably means less innings for the bullpen. If that’s the case then Romero and Bastardo may see only LHH. If that’s the case it’s not good news for opposing LHH. Bastardo struck out 14 of the 41 LHH he faced last season, and had a BAbip against RHH of .429. Do you think that .429 is “sustainable” for RHH when they face him? Also, he gave up only 1 HR in 18 IP. Sure, call me on sample size, but he did improve in 2010. Bastardo was a starter until mid-way in 2009. Doesn’t he get a chance to learn to adjust to the bullpen?

    Besides, isn’t using the 2010 results for the pitchers in this bullpen also using a small sample size?

    So, while I’m not sure I would place the Phils’ bullpen at #7, I think they have a chance to be a little better than a middle of the road ‘pen.

  35. awh

    February 03, 2011 02:11 PM

    Oh, and speaking of injuries, wasn’t Bastardo derailed after being lights out earlier in the season in 2010?

  36. MG

    February 03, 2011 02:26 PM

    awh – ‘Elite’ at 75th percentile? I would either define it at 80th to 85th percentile as a minimum floor and possibly higher. It would largely depending on what your are observing and how the data is distributed.

    Since you have 30 MLB teams and 16 NL teams, I would say ‘elite’ (80th percentile) would be top 6 among MLB teams and top 3 in the NL.

    The metrics you could use would be up for debate. Things that are no-brainers include:


    Others I would include that shouldn’t be that controversial:


    Up for debate but I would include:

    % SV PCT (although I could see easily tossing this one too)

    That gives a base criteria of 10 rankings. Maybe you want to include others like WAR or WPA. The only question is do you provide any criteria with a higher weight because it is something the reliever has more direct control of.

  37. MG

    February 03, 2011 02:40 PM

    awh – Yeah Bastardo was set back by injuries last year but he was the year before too. Injuries & health are something that is actuaully a bigger impediment to his MLB career than his control issues.

    Ditto on Contreras. He is a guy who has had a ton of injury problems in the previous few years prior to last year, went throw a dead arm period last year, and is ancient at this point.

    Maybe I am wrong on Lidge and that his amazing Aug/Sept largely carries over to this season. I actually don’t think he will be terrible like he was in ’09 and the first half of ’10. Just not quite as dominant as Aug/Sept. Either way, he is the key to this bullpen yet again because he has such a likelihood of having such a wide variance in performance. That right knee though has had 3 major injuries in the past few years and can’t imagine that it is in great shape at this point in his career.

    Yeah getting Lee makes a difference but this was already a bullpen that threw the fewest IP in the NL at 421 which was the lowest total an NL bullpen threw since the ’05 Cards (397 2/3 IP).

    If everything breaks there way with the rotation, they likely will be under 400 IP and push the 387 2/3 IP that the DBacks’ bullpen threw in ’02. That would be a big difference. If they did pitch that few IP, you likely will really limit IP by the lesser talents and a good chance for this bullpen to be ‘elite’ based upon aggregrate measures.

Next ArticleThat Mighty Fine Starting Rotation