Jonathan Papelbon’s bad (?) year
Posted by Andy on September 23, 2010
Most Red Sox fans would tell you that Jonathan Papelbon is having a bad year. Some stats to back that up:
- He's got 7 blown saves, a career high, and an 84% save percentage, a career low
- His ERA is 3.86, when his career ERA before 2010 was 1.84.
- His WAR this year is just 0.6, after averaging 3.3 in his 4 seasons previous seasons as closer
However, a closer look at the numbers makes the picture more confusing.
Batters are hitting .218 against Papelbon this year, as compared to .213 last year and .223 in 2008. Those numbers are all basically the same with just a hit or two difference. In his best seasons of 2006 and 2007, Papelbon yielded a BA well under .200, but the point is that not a lot has changed from 2008-9 to 2010. Mariano Rivera has a career BAA of .210, so there's no reason to think that Papelbon's figure of .218 this year is the problem.
In fact let's look at Papelbon's key rate stats of hits, HR, BB, and K per 9 innings pitched.
In 2009, over 68 IP, they looked like this:
7.15 H/9, 0.66 HR/9, 3.18 BB/9, 10.06 K/9
In 2010, over 63 IP, they look like this:
7.29 H/9, 0.86 HR/9, 3.43 BB/9, 10.00 K/9
All of his numbers have gotten slightly worse in 2010 by very small amounts, and they are all still good numbers (with the possible exception of BB/9, which in both 2009 and 2010 have been fairly high, although this year's walk total includes 4 IBB.)
The bottom line is that this is a guy who is still not allowing many hits and striking out a ton of hitters. So what gives? How can Papelbon have basically the same rate stats as last year but much worse ERA, WAR, etc?
The devil is in the details. While he's allowed just one more homer this year than last, 39% of hits hits allowed have gone for extra bases as compared to just 28% last year. Most of them were doubles. In 2009 he allowed 10 doubles and 4 RBI on those doubles. In 2010, he's allowed 14 doubles and 8 RBI on those doubles.
The real issue here is that in the games where he hasn't allowed any runs, he's pitched just like he did last year. But, he's had a few extra games this season where he's allowed some runs. In fact, he's had 3 different games with 4 ER. Having allowed a total of 13 ER more this year than in 2009, those 3 games alone account for most of the difference.
Meanwhile, in his 61 games this year, he's allowed no ER in 49 of them (80%). In his 66 games last year, he allowed no ER in 56 of them (85%). Yes, he was better last year, but it's just a small number of games we're talking about.
Overall, it would seem that Papelbon had a bit of good luck in 2009 and a bit of bad luck in 2010. His numbers were quite similar between the two seasons but one happened to result in a bunch more earned runs and blown saves.
In the larger scheme, I wonder what his 2009 and 2010 numbers taken together say about his long-term closing prospects. Since he became a closer in 1997, Rivera has never had a seasonal walk rate as high as Papelbon's in each of the last 2 years. Rivera has never had a seasonal WHIP as high as Papelbon's in each of the last 2 years. From 1996 to 2009, Trevor Hoffman also had lower WHIPs than Papelbon in 2009 and 2010 and a lower walk rate ever year except for 1996 itself (3.2/9)
Now, falling short of Rivera and Hoffman in a comparison is not a terrible thing. It's like saying a hitter isn't as good as Babe Ruth--well if he's not as good as Babe Ruth but he is as good as Willie Mays, he's still doing really well.
I think from looking at Papelbon's numbers, we can conclude that he's probably not going to be among the handful of very best closers of all time, but he still will probably be very, very good.
September 23rd, 2010 at 10:10 am
As he is not as good as he was at the beginning of his career, I am sure when he hits free agency the Mets will overpay to get him.
September 23rd, 2010 at 10:21 am
Then he can follow in the footsteps of previous great Mets acquisitions and punch out his fiance's father in front of his family!
September 23rd, 2010 at 10:22 am
The kid is just 29; not quite ready for us to be thinking of him as over the hill. {Bearing in mind that this is coming from a man who still can't believe that Truman actually beat Dewey}>
September 23rd, 2010 at 10:42 am
Is he entering games with closer scores than he did in past years? Allow two runs in a three-run game, and you still get a save. Allow two runs in a 1-run game and you get a blown save and a loss.
September 23rd, 2010 at 10:49 am
Chris, we can get the answer to that question here:
http://www.baseball-reference.com/players/p/papeljo01-pitch.shtml#pitching_win_probability
His average leverage index has been really high each of the last 2 years, which may have something to do with why his 2009 and 2010 numbers are not nearly as good as 2007 and 2008. He's been put into a lot more high-pressure situations, meaning that other teams may have brought in key pinch-hitters more. His high leverage appearances have been 36 in 2009 and a whopping 42 in 2010, as compared to just 32 and 33 in 2007-8. Beyond numbers, this may have worn him down some, too, by giving him fewer easy saves, more pressure pitches, etc.
I note too that his WPA+ this year of 10.1 is the 3rd highest of his career, while his WPA- is the highest (i.e. most negative.) That means that, yeah, he's had a collection of bad games but also has pitched quite well in all the rest of them.
September 23rd, 2010 at 10:59 am
A small amount of games makes a huge amount of difference with a closer. Last season, he blew three saves. If he duplicated that, that's three more victories this season (I think they won one game where he blew a save). That means the Red Sox would have entered last night 2.5 games behind Tampa Bay in the wild card. Instead they just have been playing out the string for a couple weeks.
September 23rd, 2010 at 11:52 am
Stats are all well and good, but anyone who has seen Pap pitch this year, knows his body language isnt the same. He is very easily frustrated and often times looks disinterested in even being involved in the game. I personally hope they let him go in the offseason, and focus on the offense and building a bullpen to bridge from the starters to Bard.
September 23rd, 2010 at 12:00 pm
With closers, the blown saves are always the indicator for how a fan "feels" about him. It seems like he is having a bad year because of so many (compared to previous years) blown saves. Even if he had been giving up many more (or less) hits/BB than last year, it still would come down to blown saves.
Maybe it would seem like it was more "interesting" (i.e. like John Wetteland) if he loaded the bases every game before getting the save, but if he converted the saves at the same rate as last year, fans wouldn't think of it as a bad year (even though his rates could be significantly worse).
And since we all know the difference between a save or a BS can often times be very slight, a closer can easily blow 4 extra saves with just a few extra BBs/Hits/HR each from the previous season - which shows up in his rate stats as you have outlined.
September 23rd, 2010 at 12:12 pm
@7,
As a Yankee fan, I know I do not like to have to face Papelbon. He might just be having a bad year, I don't know how a team could let a closer that good go. If he truly is disinterested, then the Sox need to work with him to figure that out, but I think it is more that he is frustrated and hasn't really had to deal with sub-par performance in his career. MANY players have bounced back from bad years and you really can't fool around with the closer position when you have a guy that has proven he can do it (as long as he is not injured). I can't believe that someone who does a jig on the mound, can be all that disinterested.
September 23rd, 2010 at 12:35 pm
Andy,
"meaning that other teams may have brought in key pinch hitters more." No, leverage is not set up to measure the quality of the opposition, only the score, inning, and base-out state when the pitcher entered the game are relevant. His average Leverage Index is down from last year. 2.0 in 2010, 2.2 in 2009.
September 23rd, 2010 at 12:37 pm
As a Yankee fan, you should be well aware how teams can let good, proven closers go.
September 23rd, 2010 at 12:42 pm
only the score, inning, and base-out state when the pitcher entered the game are relevant.
Actually, it's those elements throughout the pitcher's entire appearance which are counted on B-R. Sometimes, I do only care about the LI when the pitcher actually entered the game. (The pitcher's own performance will change the LI after he has come in.) B-R should show that but until they do you can check Fangraphs. Papelbon's avg LI when entering the game is 1.91, 2nd highest of his career. Over the last three seasons was about 1.76 each year. (This could be slightly affected by the lower run-scoring this season.)
September 23rd, 2010 at 12:50 pm
kds, sorry I skipped a step in my explanation. I meant that given it was a higher leverage situation, it makes sense that that might have been a key spot where the other team used a pinch hitter due to the high leverage, not that the leverage was high due to the pinch hitter.
September 23rd, 2010 at 12:52 pm
As a Sox fan who has watched Pab for his entire career (though not every game), I can say that his pitching style definitely belies future concerns, either in quality or mentality. He basically only throws one pitch, his fastball, and avoids anything else. As the post indicates, the impact on his numbers has been minor. But there is still a downward trend and little reason to believe it will reverse course any time soon. I think it's likely that the Sox let him go after this year; not necessarily because he's incapable of remaining a competent closer for the foreseeable future, but because the salary he'll command will far outweigh his value.
Andy-
A question... you noted his BAA and talked about his BB/9 and the extra base hits. Why not just use opponents OBP and SLUG? We realize these to be far superior stats to BA for hitters... wouldn't the same hold true for pitchers? Or am I missing something? As you show, his BAA compares favorably to elite closers but his OBP (judging by his BB/9) clearly doesn't and it sounds like his SLUG is up, at least compared to himself. Are these numbers readily available? Is there a flaw in this reasoning I'm missing?
September 23rd, 2010 at 12:57 pm
I'm not greatly informed on how leverage works so I might be wrong but isn't it true that his high leverage could be due to the fact that he's putting himself in more pressure situations? His relatively high average leverage over the last two years has coincided with a relatively high WHIP as well.
September 23rd, 2010 at 12:57 pm
Oops it was already brought up
September 23rd, 2010 at 2:41 pm
The consensus I got from Red Sox fans before the season about Papelbon was:
- while he's been an excellent closer overall, he's been in slow but noticeable decline the past few years
- he's going to want a big contract, in line with what the game's other top closers get
- he's not going to give the Red Sox any "hometown discount" on his contract
- Daniel Bard looks pretty good, but he isn't ready to became the closer (yet)
Now that this season's almost over, the consensus seems to be:
- Papelbon's still a decent closer, but he's no longer amongst the elite
- the Red Sox should NOT offer him a long-term contract, and let him declare free agency
- Bard has proven worthy of taking over the closer role
I see the Red Sox offering him arbitration, him declining, and Bard moving into the closer's rule.
#14/BSK - I have also wondered why OBP against and SLG against are not more frequently used,instead of "Batting average against".
September 23rd, 2010 at 2:47 pm
No question OBP and SLG against are better stats, but I always find BAA interesting with closers. I am surprised that Rivera gives up any hits--so many of them are infield dribblers and/or broken bat jobs.
Check out how similar the career slash lines are for Papelbom and Rivera:
Rivera .210/.263/.290
Papelbon .202/.265/.309
The difference, of course, is that Papelbon's numbers the last couple of years have been well above his average. He's had an OBPA of nearly .300 and a SLGA over .330. (Can you believe we're talking about a .330 SLG against as being bad?)
September 23rd, 2010 at 2:49 pm
BSK, you are right. Pitchers' OBP and SLG allowed have not always been easily available, so people have probably gotten in the habit of quoting the other numbers. But since we do have them now, they are worth looking at. (OBP is very similar to WHIP, of course.)
Papelbon is at .218/.296/.355 this season, highest OBP and SLG of his career. .202/.265/.309 for his career. Applying a very basic runs-created formula to his numbers (OBP * TB), he should have allowed 24 runs, and has allowed 30. This implies there has been some clustering of hits/walks and indeed he has been getting shelled with RISP. That may just be chance but it doesn't make his actual performance this season any better.
September 23rd, 2010 at 3:00 pm
Are OBPA/SLUGA available on BR? I don't see it on the player pages, but I may be looking in the wrong place. An interesting stat that could come of this would be OPSA+.
JT-
I wonder if the clustering of hits/walks is luck or has something to do with the anecdotal evidence I offered. With batters, we know the issues with randomness and sample size that BARISP offers; but for pitchers, I have to believe there is more of a skill element, at least with regards to the difference in the quality/control/velocity of their stuff from the stretch as compared to the wind-up. Papelbon may be worse from the stretch (though a runner on 3rd might negate the need to pitch from it). Additionally, the overreliance on the fastball may have an impact on the SLUGA, as a fastball is generally easier to hit for power than a split-finger (his #2 pitch). I'd be curious to see what the pitchf/x data says about him and whether my amateur observations are accurate.
September 23rd, 2010 at 4:02 pm
Click on More Stats to get the batting lines against each pitcher.
Papelbon was excellent with runners on last season, although he was not pitching as well overall as he previously had, and hardly threw his splitter last year. This is the first season he's had problems in such situations.
September 23rd, 2010 at 4:18 pm
Last year Papelbon stranded 89.3% of the runners he allowed. This year he's only stranded 68.5%, despite putting up better peripherals when runners are on than with the bases empty. This is where most of the runs come from.
September 23rd, 2010 at 6:15 pm
If I understand the data correctly, it seems as if his success (or relative lack there of) this year is largely a function of luck, with upticks in OBPA and SLUGA (I love that acronym, btw) also contributing. Despite this, his peripherals are trending downward and his approach is, potentially, questionable. Seem like a fair assessment?
Thanks to Andy and all; posts and conversations like these put things in perspective.
September 23rd, 2010 at 6:26 pm
The Sox' defense has been notably poorer this year, especially in the outfield. Was at the game Pap "blew" a couple of Sunday's ago against the White Sox. He would have closed out the ninth if Kalish could have caught one of two well-hit but catchable fly balls to CF. Ellsbury or Cameron would have done so, I am sure.
September 23rd, 2010 at 6:57 pm
Andy, this was one of your best pieces. It didn't require a rocket science blend of calculus to see the story with Pap this year. His year is the story of the curse of stats for closers- you have a solid year spoiled by a handful of really horrible outings. A sharp pencil was able to show what the whole AL knows... Pap is still one of the feared closers in the AL, despite this not being his best year.
September 23rd, 2010 at 7:57 pm
Mark-
That brings up an interesting question... was Papelbon "robbed" because Cameron/Ellsbury was there or was he enjoying a "luxury" when he did have them? If we are getting a sense of his true talent level, we'd want to look at how an average CF would have handled those plays. If we are looking at what the Sox could have reasonably expected, then we could consider the anticipated starting CFs, assuming their was reason to believe they'd remain starters (a bit questionable with Ellsbury and Cameron).
September 23rd, 2010 at 8:40 pm
Great, intelligent discussion all round. Andy, a meticulous, well thought-out and researched post.
@7
Matthew, from where I sit you've got it right. Papelbon appears to lose his cool too easily for a closer. I still remember the bad optics of his failure to cover second base in the short outfield double against the White Sox.
Even worse, he appeared to take no responsibility for it in the post-game media sound bites!
September 23rd, 2010 at 8:59 pm
@22
Will S, so is Papelbon a "victim" of how he is being used this year? It's clearly a manager's call about whether to bring a closer in to start an inning or to clean up another pitcher's mess.
Who has inherited more runners on a per-season basis over their career, Rivera or Papelbon? And is inherited runners scored a true measure of closer effectiveness?
I've always felt that Rivera was "sheltered" for much of his career (if you can use that word for a closer) by only working 9th innings when the Yankees are ahead.
Somewhere, somehow, the definition of a save has to be tweaked.
September 23rd, 2010 at 9:07 pm
http://www.baseball-reference.com/players/p/papeljo01-pitch.shtml#pitching_reliever
In 2009 and 2010, Papelbon has inherited 31 runners and allowed 9 to score. He had 15 outings of more than 3 outs recorded (out of 127 total appearances.)
http://www.baseball-reference.com/players/r/riverma01-pitch.shtml#pitching_reliever
In 2009 and 2010, Rivera has inherited 32 runners and allowed 8 to score. He had 14 outings of more than 3 outs recorded (out of 124 total appearances.)
So, the numbers are remarkably similar.
However, if we go back to Rivera's 4th and 5th years as a closer (equivalent to this year and last year for Papelbon) he inherited 51 runners and allowed 15 to score, and had 30 outings of more than 3 outs out of 132 total appearances. So Rivera was, in theory, taxed a lot more back then than Papelbon has been.
September 23rd, 2010 at 10:04 pm
Andy, spot on... as usual! The similarities between Rivera and Papelbon in terms of usage are remarkable.
I guess Mariano has ice in his veins and Jonathon has fire. Wow, the make up of a closer.
September 23rd, 2010 at 11:19 pm
A few bad outings can really inflate a relief pitcher's ERA, so it's possible he'll bounce back. Yet we can't ignore the obvious, which is most closers do not have long careers, at least as closers. The Rivera's and Hoffman's are the exception. I'm not really sure Papelbon has a specific out pitch that will allow him to be a top-tier closer well into his 30s. We may simply be seeing the beginning of his fall.
September 24th, 2010 at 12:07 am
@27.
I was trying to remember who Pap reminded me of, and couldn't think of Hrabosky until I caught your response here. Thanks.
September 24th, 2010 at 2:55 pm
The Mad Hungarian was far more demonstrative than Pap!