Ryan Howard’s RBI Opportunities

At Beyond the Box Score, Jacob Peterson (@JunkStats) did some outstanding work digging into Ryan Howard‘s RBI opportunities. Howard, of course, has been the subject of debate for the last year and a half since he signed his five-year, $125 million contract and the RBI stat has been caught in the crossfire. Peterson finds out where Howard stands amongst his peers.

As for the remaining 10 or so RBIs above average, Howard has also almost certainly benefitted from fortunate sequencing this season. His OPS+ is 30% higher than average when he hits with runners in scoring position, but that is not sustainable.

For his career, Howard’s OPS+ is 13% higher with RISP, markedly lower than this year’s 30% mark. Once you take away Howard’s huge number of intentional walks in these situations (almost half his total walks), his career OPS+ drops to only 7% better than normal with RISP. That’s almost identical to the leaguewide figures, which this year are 6% higher with RISP.

So Howard does not seem to have the ability to hit notably better with RISP, indicating that this year’s performance in that situation has been a fortunate fluke.

Leave a Reply

*

38 comments

  1. Cam

    September 12, 2011 07:46 AM

    Does the league-wide stat also discount IBBs?

  2. olo567

    September 12, 2011 08:28 AM

    Yes, why discount IBB? They matter.

  3. Bill Baer

    September 12, 2011 08:35 AM

    They don’t tell us anything about hitter skill. They tell us about some combination of first base being unoccupied, a preferable match-up behind Howard (e.g. RHP vs. RHB), and other factors.

  4. HW

    September 12, 2011 09:06 AM

    1) Don’t see how one can conclude that Howard does not hit better with runners in scoring position when his OPS is 13% higher over the course of his career. Moreover, his RISP batting average is 18 points higher than his average with the bases empty, so the higher OPS is not just intentional walks. His slugging percentage is also 38 points higher with RISP versus bases empty. The sample is large enough that you can’t write this off as a fluke – it’s a long-term outcome over the course of his career. I agree that his 30% highe OPS this season is unrepresentative.

    2) I think that the concern regarding intentional walks is that you would need to also discount the intentional walks of other batters in order to make a fair comparison. In addition to the factors that you cite as explanations of IBBs, isn’t it possible that he get more IBBs because managers fear his hitting skill (regardless of whether they should)? If you believe that his career difference in hitting in RISP versus bases empty situations has ocurred over a large enough sample to accept that it is not simply random, IBB decisions might reasonably reflect concerns about Howard’s hitting skill (again, regardless of whether the IBB is the optimal strategic play). Surely you don’t think that IBBs are “purely” situational.

  5. charles

    September 12, 2011 09:23 AM

    two points: isn’t it still a “skill” that Howard hits better when there is no shift?

    i was checking howard’s fangraphs page and noticed an anomaly – he destroyed sliders that year, as opposed to every other year. seemingly, he’s dropped that tactic – if it was a tactic – of trying to lock in on sliders. i think it’s unfortunate, since 2009 is his 2nd best year with the bat

  6. Mike B.

    September 12, 2011 09:23 AM

    People knock Howard’s RBIs (and I agree that raw RBI numbers paint an incomplete picture), but it seems to me that a better analysis (rather than simply discounting them entirely as the product of the hitters in front of him) would be to look at the percentage of available runners knocked in vs. the league average, or some “top” players.

    I remember reading something along those lines a year or two ago that pointed out that Howard knocks in a higher percentage of “available” runners than anyone else. This chart seems to support that finding. IMO, there is a lot of value in that, and it might be a way to differentiate the value provided by Howard from the value provided by the guys in front of him.

  7. HW

    September 12, 2011 09:27 AM

    I think the shift is a reasonable explanation, but it doesn’t change the fact that the performance is meaningfully different in those situations (and I’d love to see the spray charts on this). I’m not claiming “clutchness,” just that you can’t look at these data and not find a significant long-term difference in performance. The statement was that he doesn’t hit better in RISP situations. The data indicate that he does; if the explanation is that the defense has to play him differently with men on base, it doesn’t matter from the perspective of thinking about his productivity in those situations.

    See the conclusion of the article you linked to:

    “All in all, I think this is a sign that perhaps using purely sabermetric theories which assume that no hitter is significantly better than others at performing in the clutch, and thereby evaluating hitters’ cumulative numbers, is not accurate. This is not quite a testament to the glory of the RBI, but it demonstrates some of the value in looking at situational context hitters as different. I tend to doubt the impact of psychology on performance and mainstream sportswriters frequent use of antiquated ideas of clutchness, but there are probably characteristics that affect hitters’ abilities to succeed in important situations. The characteristic presented here may be one of them.”

  8. John

    September 12, 2011 09:36 AM

    The IBB’s are a reflection of other teams’ fear of pitching to Howard. It’s flawed statistics to discount those from the analysis. Just people who want to bend the stats to make Howard look bad. Bottom line is he is better with RISP. As others said, the shift takes many hits away from him when nobody is on base.

  9. Bill Baer

    September 12, 2011 09:57 AM

    @ charles

    Personally, I’m not a fan of the pitch type linear weights on FanGraphs since they’re based on a relatively new Pitch F/X model. Look at the inconsistencies in Mariano Rivera’s cutter usage, for example:

    www.fangraphs.com/statss.aspx?playerid=844&position=P#pitchtype

    @ Mike B

    www.baseballprospectus.com/sortable/index.php?cid=1010713

    @ HW

    I think the shift is a reasonable explanation, but it doesn’t change the fact that the performance is meaningfully different in those situations

    Under what basis do you determine that it is “meaningfully different”? If you’re using a synonym for “statistical significance“, I would be interested in seeing that laid out.

    if the explanation is that the defense has to play him differently with men on base, it doesn’t matter from the perspective of thinking about his productivity in those situations.

    I hear this all the time, but I don’t see why we should credit Howard for being so predictable that opposing defenses play him a certain way. It’s his flaws that cause them to align that way. If Howard divided up his infield batted balls like a normal MLB player, there wouldn’t be such a disparity.

    Saying he’s “better in RISP situations” is relative. He would have about the same split between RISP and the bases empty if he didn’t pull all his ground balls. Why are we using that to anoint him a RISP master?

    @ John

    The IBB’s are a reflection of other teams’ fear of pitching to Howard.

    That’s why they walked him intentionally 8 times in his great 2009 season (141 OPS+), right?

    He’s been IBB’d 16 times this year with a 128 OPS+. The IBBs reflect the situation, not the player, in this case.

    www.baseball-reference.com/players/h/howarry01.shtml

  10. Matty B.

    September 12, 2011 10:05 AM

    Bill — I don’t think we should credit Howard as having a “skill” that he hits better with RISP, but there does appear to be predictive value in the data nevertheless. So to say that it’s not “sustainable” isn’t quite accurate either.

  11. curtwill

    September 12, 2011 10:17 AM

    @Bill

    You: He’s been IBB’d 16 times this year with a 128 OPS+. The IBBs reflect the situation, not the player, in this case.

    Me: But he will probably be over 130 OPS+(128 is close to that, it’s better than last year believe it or not and he is having his “typical” September so it’s a good bet that he will be 130 or better in OPS+) and not to mention, he has a 159 sOBS+ also with RISP on the same chart you posted so that gives more enlightenment toward the topic.

    The 16 IBBs(3rd in the NL) are because of, as you said situations that might dictate it(in some instances not all), but this also shows that managers would rather pitch to the hitter behind him more often than not. Think about it, especially early in the year, if you’re a manager who would you rather pitch to with RISP, Ryan or Ben Francisco?

    At least now, he has Hunter Pence healthy(he usually hits 5th when the Phils have their whole collection of players) so that gives him more opportunity to hit fastballs. According to Fangraphs, his fastball “value” is in the top 10 so more FBs, the better for him.

  12. Bill Baer

    September 12, 2011 10:20 AM

    @ curtwill

    I actually did research on the “protection” theory that Howard got more fastballs with Hunter Pence behind him, and that wasn’t the case at all. Exactly the same amount of fastballs before and after Pence. I didn’t write an article about it but I did a segment about it on my Phillies 24/7 show a couple weeks ago.

    Here’s what I had on my outline:

    2009-11, Howard has seen about 45-50% fastballs

    Howard saw 48% fastballs before Pence, 48% after

    Before Pence
    vs. RHP: 49%
    vs. LHP: 44%

    After Pence
    vs. RHP: 49%
    vs. LHP: 44%

  13. curtwill

    September 12, 2011 10:24 AM

    Good point@ 2009-11, Howard has seen about 45-50% fastballs
    ? Howard saw 48% fastballs before Pence, 48% after
    ? Before Pence

    And that’s somewhat surprising to me, I figure he would get more fastball than he has, but I think earlier he did get a tad more fastballs when Hunter first came on the team(Philly.com, I think the Inquirer Beat Writer had an article on it).

    On my last statement, I should have said, “should give him more Fastball”, my fault.

  14. Richard

    September 12, 2011 10:30 AM

    Actually the beat writer specifically said the opposite. Howard had seen like one or two more fastballs, which is hardly worth mentioning.

  15. Rob

    September 12, 2011 10:52 AM

    Think discounting IBBs is poor logic because it would seem to lead to an adverse selection that would negatively impact Howard’s productivity in the at-bats where he does get pitched to. Specifically, he is far more likely to get an IBB from a righthanded pitcher. So, basically you say that IBB doesn’t count, but ignore the fact that it will almost always come in a situation that is more favorable to Howard (hitting off a RH pitcher), leaving him with a greater number of less advantageous situations (hitting off a LH pitcher). Does someone think an IBB is somehow less valuable to the team than an “unintentional intentional” walk, which you are including? This specific part of the approach feels like manipulating statistics to make a pre-determined point.

  16. HW

    September 12, 2011 10:54 AM

    Bill,

    I intentionally avoided asserting a statistically significant difference in RISP versus bases empty because I haven’t done the math. Given the number of observations and the size of the differences, I would bet a hefty amount that the difference is statistically significant.

    Really, you seem to acknowledge that there is a significant difference, but you don’t want to attirbute any value to it (or give “credit” to Howard) because you think that the effectiveness of the shift is a sign of Howard’s limitations. I don’t disagree that the effect of the shift on him shows his limitations, but you can’t go from there to say that the difference in his stats in RISP situations is 1) not meaningful or 2) not sustainable.

    It’s meaningful and sustainable precisely because the shift is effective. This doesn’t mean that he is clutch (or great), but that given a conventional fielding configuration (which he gets when men are on base) he is predictably more productive.

    I would think that this is the kind of analysis you would favor. We have seemingly clear data and a testable hypothesis regarding Howard productivity differential that doesn’t rely on soft claims about psychology or clutchness. He performs better without the shift, and that results in RBI production that’s unexpected given his overall numbers.

    You wrote: “Saying he’s “better in RISP situations” is relative. He would have about the same split between RISP and the bases empty if he didn’t pull all his ground balls. Why are we using that to anoint him a RISP master?”

    I only bring up the RISP numbers because the article asserts that “Howard does not seem to have the ability to hit notably better with RISP.” The data (and your shift explanation) indicate that this claim is false.

  17. Rob

    September 12, 2011 11:01 AM

    HW I agree with you that the ability to hit well without a shift in place is a skill, and that skill is repeatable specifically because it has a consistent cause. This point was raised in several comments in response to last week’s Howard posting, and evidently it was not persuasive to the author.

    BTW – did you notice the Brewer’s shifted Howard even with RISP in the 8th inning yesterday? He grounded to 2nd, wonder if that will start a trend…

  18. Spree75

    September 12, 2011 12:21 PM

    Personally I think there has been a language problem with last couple of articles between the words “better hitter” and “more productive hitter”. The former would indicate a change in skill while the later would inidcate a change in circumstance.

    I don’t think Howard’s skill level is changing when there are runners on the base paths. All the statistical numbers I have seen seem to say that his contact rate by pitch type, fb/gb%, location of hit/out don’t change. This makes sense to me. To get any of these numbers to change would suggest that the player is altering mechanics to get an improved result. If this was true then I would be wondering why the altered mechanics aren’t used at ALL times.

    A changed circumstance though makes perfect sense for Howard with runners on base. Several people (me included) have pointed out that without the shift that Ryan’s batting numbers would look better. Runners on base tend to negate a shift. Hence Howard has a higher stat line with runners on base. Higher stat line with runners on base equals RBIs.

    To sum up a rambling post:

    Ryan Howard is a more productive hitter with runners on base. To say he is more skilled is stretching reality.

  19. Mike B.

    September 12, 2011 12:40 PM

    @Bill: thanks for the link; that’s exactly the stat I was thinking of.

    Spree75 said:
    “A changed circumstance though makes perfect sense for Howard with runners on base. Several people (me included) have pointed out that without the shift that Ryan’s batting numbers would look better. Runners on base tend to negate a shift. Hence Howard has a higher stat line with runners on base. Higher stat line with runners on base equals RBIs.”

    This make sense to me. We’re starting to see a lot more “active defenses” employed against all hitters, not just Howard. It stands to reason that the purpose of changing defensive alignments is to counter a given hitter’s strengths. Therefore, it also stands to reason that when the optimal alignment isn’t used (e.g., when runners are on base), the hitter will tend to have a better average outcome because the defense is no longer optimal to account for that hitter. It doesn’t mean the hitter is “better” in such situations, but it means the defense is “worse.”

  20. Mark

    September 12, 2011 12:50 PM

    What I would like to know is how this information stacks up to the likes of Gonzalez, Pujols, Texeira and Fielder. I assume they would be the best ones to equate this info to. Otherwise, it looks to me like he’s doing his job. People can argue whether he’s worth the contract, but I keep reading that he doesn’t match up to the other luminary 1st basemen. Would it be possible to run the other 4 through the same process?

    P.S. Saber guy I’m not, so sorry for not throwing in some stats. :)

  21. Phil Eichinger

    September 12, 2011 03:41 PM

    How many home runs and rbi”s were game winners???????

  22. Jay

    September 12, 2011 04:21 PM

    So Howard is not only the best run producer in the league, he also is able to advance runners at an above average rate. What a beast. #WorthEveryPenny

  23. Jon

    September 12, 2011 06:18 PM

    The whole crux of this argument that I will never understand is why there’s an assumption that Howard’s production against the shift is his “true production”, rather than his production with RISP in a non-shift situation.

    If his non-shift production stays consistent, and his shift production stays consistent (and lower), can’t he both a) consistently outperform his overall numbers and b) be overrated by his RBIs?

  24. Phlipper

    September 12, 2011 07:48 PM

    Of the “elite” hitters I’ve looked at, Howard’s career %OBI (% of other runners on base ahead of him knocked in) is at an elite level.

    If you add that to the fact that he knocks himself in via at a better rate than anyone else in the game, what you wind up with is an elite cleanup hitter, an elite run producers. It really isn’t that complicated an analysis.

    These are simple facts.

    From what I’ve looked at, his career OPS with runners on base and RISP are very comparable if not better than other “elite” hitters. That, along with the fact that he hits HRs at a virtually unprecedented rate explain why his %OBI is so high. The reason why his performance with runners on base and RISP is so good is because that is when the shift isn’t employed.

    You can discredit his performance with runners on base and RISP all you want by claiming that it isn’t sustainable, or by saying that it isn’t a product of his “clutchyness” (which I haven’t seen anyone claim). The problem with that argument, however, is that he has sustained a high level in that metric throughout his career.

    The simple fact is that he produces runs – even when the # of opportunities he has is controlled for – at an elite level. When you have a team that can put runners on base ahead of their cleanup hitter at a pretty good pace, that is EXACTLY the kind of cleanup hitter you want.

  25. Phlipper

    September 12, 2011 07:50 PM

    Sorry – that should read “…he knocks himself in via HRs at a better rate than anyone else in the game…”

  26. thephaithful

    September 12, 2011 08:15 PM

    To truly analyze RBI Opps, people need to stop being lazy and do the work to create expected RBI. See my work below:

    www.thephaithful.com/aeratio

    Every Base Out Situation holds an expected RBI value. Calculate every situation based on the given season’s averages. I can’t do 2011 until retrosheet comes out.

  27. Kaycee

    September 13, 2011 10:56 AM

    Question about the OPS+ and how much higher it is with RISP compared to prior years… I know you have written before that Howard is hitting to the right side of the field more this year than in years prior. Could this have an effect? I was thinking that, if he is hitting more to the right, the shift would be even more effective than usual. Which would cause a larger divide when compared to situations where there is a runner on second and the defense has to spread out to cover third.

    If so, that 30% might be somewhat sustainable, not because Howard is that much better this year with RISP, but because he is worse without.

  28. EHW

    September 13, 2011 11:49 AM

    At the risk of pouring additional gasoline on the pile of crap that is the general Ryan Howard/sabermetrics discussion, Howard currently leads the Phillies in wRC.

    I know that the number by itself only means so much, but considering we were praising Shane for having whatever high percentage of the team’s total wRC he had, Howard deserves at least a bit of credit in that regard.

  29. JC

    September 13, 2011 12:01 PM

    Yea. But I’m not sure his wRC of 87.5 is significantly better than Shane’s 87.4, especially because Shane’s wRAA, wRC+, and wOBA are all higher than Howard’s (29.2 vs 19.6, 144 vs 123, .387 vs .357).

    **note- not taking credit away from Howard here. I like him **

  30. EHW

    September 13, 2011 02:26 PM

    True, which is why the number itself only means so much. And not to discredit Shane either because he’s obviously had an awesome year other than his September struggles. It could just be because wRC is a counting stat, I think, and Howard has played 25 or so more games than Shane.

    It’s just something to note–that we’re praising one guy for what he’s done and shitting on another when their total offensive production, independent of everything else, has been more or less equal. Granted, when you take into account the CF vs. 1B comparison and when you add in defense and running and what not, Shane’s obviously had a better season, but still, I feel like a 4.3 WAR difference between them may be overstating it. (Acknowledgement: I didn’t read most of the recent Howard debates on here, so apologies at the fact that some of this has probably come up before; I’m just making a comment on the general perception of their seasons from Bill and most analysts.)

    In any case, even if his rate stats aren’t great (and they’re not), if the first baseman’s job is to produce offense Howard has pretty much done that. He’s ranked 9th among 1B this year in wRAA, and that sounds about fair as to where I’d rate him.

  31. hk

    September 13, 2011 04:04 PM

    EHW,

    But, even if we only use wRC, what does it tell you about Howard’s year if he has needed 17% more plate appreances than Victorino to post basically the same wRC? In addition, what does it tell you about Howard’s and Victorino’s years relative to their future contracts that Howard is a significantly worse base-runner than Victorino and is a significantly worse fielder (at a less valuable position)?

  32. EHW

    September 13, 2011 06:47 PM

    “But, even if we only use wRC, what does it tell you about Howard’s year if he has needed 17% more plate appreances than Victorino to post basically the same wRC?”

    It says that Shane has been better, has a better triple-slash and wOBA, and has gone on the DL twice. All of which I at least alluded to already.

    “In addition, what does it tell you about Howard’s and Victorino’s years relative to their future contracts”

    I said nothing about their contracts and I wasn’t planning on it. I’m not trying to defend his contract, I’m trying to say he’s not some guy who the Phillies could just replace fairly easily and not someone who deserves to be shat on the way the sabermetric community has enjoyed doing so recently.

    “that Howard is a significantly worse base-runner than Victorino and is a significantly worse fielder (at a less valuable position)?”

    No one’s gonna deny that Shane is BETTER than Howard on the bases or in the field, but I do feel Howard gets a bad rep at those spots. His defense isn’t BAD, at least; Bill himself has said (I think) that UZR has some trouble with first basemen, and he’s at least decent at scooping, catching, and basically anything that doesn’t involve throwing the ball. Certainly not going to be in any Gold Glove conversations, but he’s not a huge liability the way some would like to believe.

    And while his baserunning is pretty poor, there are very few quality hitting first baseman who are even average runners. And, likewise, very few center fielders who are NOT good runners. Qualified 1B average out to -2.1 baserunning runs this year; qualified CF average to about 2.6. Shane is a bit above that average, Howard is below it. That shouldn’t be news to anyone. Yeah, his running is bad, but I don’t think anyone’s really expecting it to be good, and that’s true for most first basemen.

    Don’t get me wrong: I’m not some huge Howard apologist. I didn’t even read any of the other big arguments about Howard on Crashburn or even anywhere. And I don’t think the guy’s flawless by any means and I know that his contract was pretty dumb. But I think he’s a solidly above-average first baseman. Like I said based on his wRAA, 9th seems about right. And I didn’t mean to make this Howard vs. Shane because (a) I love both of them, (b) Shane is awesome, and (c) either way, no one should have to think twice about which one of them has been better this season as a whole. But at the same time I think saying that Shane has been three and a half times the player Ryan has been this year (which is what WAR suggests) is a little overboard.

  33. Phil Eichinger

    September 14, 2011 08:05 AM

    Ted Williams hit against the shift, Stan Musial hit against the shift, how did that turn out ???

  34. Yo Mama

    September 14, 2011 10:53 AM

    So, Phil. Howard is not as good as two of the best ever? Shocking.

  35. Jim

    September 14, 2011 10:17 PM

    This horse is deader than Eohippus.

Next ArticlePhillies: Good at Baseball