Addressing Howard and RBI

I have seen the “Ryan Howard drives in a lot of runners” argument used as a justification for his contract extension quite frequently on the Internets lately. I would like to once again demonstrate that the RBI statistic is just about meaningless.

You have probably heard that RBI is dependent on the on-base percentage of the hitters in front of the player in question. This is very true. Ryan Howard will get more RBI opportunities with Chase Utley (career .380 OBP) in front of him than with Brandon Phillips (.312). Additionally, Howard will get more plate appearances hitting fourth in a high-powered lineup (weak offenses don’t rack up as much PA for obvious reasons).

Most people using RBI recently have been mindful of these flaws, but there is a more glaring mis-use still making the rounds — percentage of RBI opportunities. I don’t mean to pick on David Murphy, but he has the most recent and easiest-to-access example.

Howard has driven in a higher percentage of men on base than Pujols has in each of the last three seasons.

Now, part of that is prob due to pitchers treating Albert more carefully (if you are a lefty, you are going after Howard) w/ ROB

Only point is that, in this case, Howard’s RBIs are actually a good reflection of his ability to drive in runs

There’s information missing that is absolutely necessary to make RBI in any way meaningful in this discussion: how many runners are on base for Pujols and Howard respectively? Presumably, the “percentage of men on base” counts runners on second and third the same as a runner on second.

As you can see, Pujols has stepped to the plate with just a runner on first (the worst probability for an RBI) seven percent more often than Howard has. The other percentages are similar, with Howard holding one-to-three percentage points over Pujols. Using counting stats, we should expect Howard to out-produce Pujols in RBI every season.

The following graph will show each hitter’s percentage of RBI by base-state.

Howard has driven in more runners from first base which has a lot to do with A) the speed in front of him and B) his HR/PA rate being nearly double that of Pujols, 9.4% to 5.5%. Pujols makes more contact, however: .327 AVG, .403 OBP to Howard’s .306 AVG, .386 OBP.

Howard is also better with a runner just on third base and with runners on first and third. Again, this is due to a higher HR/PA rate (7% to 4%) and that Howard has been intentionally walked only 16 times to Pujols’ 39 with a runner on third.

Finally, while it is true that Howard has driven in a lot of runs, the real question that needs to be asked is: do we expect him to continue driving in those runs. Some people, such as Matt Swartz, expect Howard to age gracefully. Others expect the duration of his extension to be a turbulent ride. For reasons stated on Monday, I believe Howard will not age gracefully:

Already, Howard has shown signs of decline as his walk rate has declined every year since 2007 and sits at a paltry 3.6% thus far in 2010. His BABIP has been lower as more and more teams have employed an infield shift against him. Opposing teams have also been bringing in more left-handed relievers to face Howard and his production against them has swiftly dropped. His strikeout rate has declined gradually but so has his isolated power. Using FanGraphs’ pitch type linear weights, Howard’s production against the fastball has dropped every year since 2006. He has swung at more and more pitches outside of the strike zone every year since he came into the Majors. Finally, his whiff rate (swinging strike percentage) has increased every year since 2006.

Simply put, comparing Howard to Pujols is a fruitless endeavor. Pujols is, literally, twice as valuable as Howard. And the “Howard is a run producer” argument completely misses the point about projecting his talent through his age 36 season. He may have driven in 136+ runners in each of the past four seasons, but that doesn’t mean he is going to continue to do so throughout the duration of his contract — especially when stalwarts like Jimmy Rollins, Shane Victorino, and Chase Utley won’t be around forever.

Leave a Reply

*

26 comments

  1. phatti

    April 28, 2010 12:54 PM

    RBIs are not meaningless. Their importance is overrated, absolutely. They absolutely do not deserve their place as one of the big three of AVG, HR, RBI. But to call them “all but meaningless” is an overcorrection.

    In the 1987 Baseball Abstract, Bill James offered three ways to assess a statistic.

    1. Does it correlate with winning?
    2. Does it reflect a player’s abilities or is it subject to influences out of the player’s control?
    3. Is it easily understandable by the average fan?

    I realize a lot has changed since then, but these criteria still have value. RBIs docorrelate with winning–you can’t win games unless you knock in runs. Not perfectly of course, but there’s a correlation. They are easily understandable by fans.

    But the biggest knock on RBIs is the second one–that they are dependent on context and circumstance. I agree. But not entirely.

    If you put Juan Castro in the 4th spot of the Phillies line up, how many RBI would he have in a year? A lot fewer than Howard.

    Why is the top 20 in career RBI list filled with some of the greatest players in baseball history? Because knocking in runs is (only) one characteristic of great players.

    Again citing something James wrote, Players that knock in a lot runs if they

    1. Come to bat with a lot of runners on base and in scoring position.
    2. Hit well, and with power.
    3. Hit well with men in scoring position.

    Ryan Howard benefits from #1, but also from #2 and #3. I know you’re going to say that hitting in the clutch is not an ability, but even if he’s just lucky, there’s value there.

    RBIs are not the most useful statistic out there–not even close. But they’re not completely irrelevant. And Howard deserves a little bit of credit for leading the league in RBI for three of the last four years.

  2. Bill Baer

    April 28, 2010 12:59 PM

    Fair points. However, I think a pertinent question is: “What does this statistic tell us that other statistics cannot (or cannot do as well)?”

    RBI comes up limping after that question.

    Why is the top 20 in career RBI list filled with some of the greatest players in baseball history?

    Survivor bias. Better hitters tend to have longer careers. I would bet RBI rate (as demonstrated above) would include some flash-in-the-pan types.

  3. Shawn

    April 28, 2010 01:28 PM

    A cursory glance at their RBI stats would seem to indicate that Pujols and Howard are roughly equivalent in driving in runners, with Howard having a slight edge because he hits more home runs. Your detailed graphs and analysis seem to indicate that Pujols and Howard are roughly equivalent in driving in runners, with Howard having a slight edge because he hits more home runs. I’m not sure how that “once again demonstrate[s] that the RBI statistic is just about meaningless.”

    You also repeat the oft-made assertion that Howard drives in a lot of runners because of the batters in front of him. Let’s see… those batters generally include a leadoff man who hits a unusually high number of homers for a leadoff man (hence no RBI opportunity) and who also tends to have an unusually low OBP for a leadoff hitter. Then (until this season) we have a #2 hitter who also hits a lot of homers for a #2 and doesn’t have a particularly high OBP. Then we have a superb #3 hitter who also hits a lot of homers and even when he doesn’t is also excellent at clearing the bases. Great players, all. But I’m not sure that that gives the cleanup hitter some sort of overwheming advantage in creating RBI.

  4. Bill Baer

    April 28, 2010 01:52 PM

    I’m not sure how that “once again demonstrate[s] that the RBI statistic is just about meaningless.

    People are using Howard’s RBI to compare him to Pujols, in most cases making them equivalent or ranking Howard higher. The fact is that Pujols is A) a much better offensive player and B) a much, much better overall player.

    Using batting runs on FanGraphs, Pujols has averaged 64 batting runs per season since 2002. Howard has averaged just over 37 batting runs from 2006-09 (his only full seasons).

    Using WAR from FanGraphs (which excludes base running, another aspect Pujols has over Howard), Pujols has averaged 8.1 WAR since 2002. Howard has averaged 4.8 from 2006-09.

    Fair point about the hitters in front of Howard but the facts remain as posted above — more hitters are on base per opportunity for Howard, hence why he routinely knocks in more runners.

  5. Shawn

    April 28, 2010 02:14 PM

    Fair enough. Using RBI to attempt to argue that Howard is anywhere near Pujols as a total player is pretty silly. I think it’s fair to compare them as power hitters, but the fact that Pujols is significantly superior in every other aspect of the game is indisputable.

  6. Bill Baer

    April 28, 2010 02:20 PM

    Yep.

    Also, to clarify with the batting runs described above, roughly ten equate to a win. So Pujols’ batting runs equate to about 6.4 WAR per season and Howard’s about 3.7 WAR.

  7. phatti

    April 28, 2010 02:28 PM

    Bill–that’s a very good question, but it’s also a high standard to meet. If you have something like WAR, by that standard, you don’t need a lot of other statistics. RBI measures how many runs a player knocked in. There’s no other stat that tells you that. It’s not as relevant as, say, how much has he been on base, but it’s still relevant.

    Good point on the survivor bias, which I guess you could use for almost any career stat. I gave a quick look at the top season totals and asked, did this player have a truly great offensive season, and the first one that I think you could maybe answer that question with a “no” is Vern Stephens, 1949 (and he was no slouch with the bat that year).

    On an important note, let’s hope the boys can beat Lincecum today… starts in 15 minutes!

  8. Mike

    April 28, 2010 03:45 PM

    I get that it is unfair to compare Howard to Albert Pujols, but isn’t it also very unfair to comapre him to Mo Vaughn, Cecil Fielder and David Ortiz. He is in much better shape than any of them were, and his game is improving into his thirties. Isn’t he much more akin to Frank Thomas or Willie Stargell?

  9. Jake

    April 28, 2010 08:02 PM

    Off topic, but I’m always amazed how much Frank Thomas is slept on as one of the all time greats. For an 8 year period he put up some Bonds level numbers (My favorite is the 211 OPS+ he put up in the strike shortened ’94 season.)

  10. E

    April 28, 2010 10:41 PM

    It is unfair to compare howard to Mo.

    Mo Vaughn could hit lefties in his prime.

  11. Bill Baer

    April 29, 2010 05:00 AM

    Using WAR from FanGraphs (which excludes base running, another aspect Pujols has over Howard), Pujols has averaged 8.1 WAR since 2002. Howard has averaged 4.8 from 2006-09.

    Maybe not exactly twice as valuable, but pretty darn close.

  12. The A Team

    April 29, 2010 09:10 AM

    The corollary to that career low walk rate you mentioned is that his contact data has undergone a complete transformation. I wrote up an article a couple weeks back detailing it and the finding still hold true.

    phillieslongdrive.com/ryan-howard-stompin-on-some-tricks/

    I think it’s too soon to determine whether or not this is a change for the better. If he sticks with this approach, I expect a walk rate under 8% simply because he’s hitting a lot more baseballs. I’m also concerned that his power might also take a slight dive although he should be less prone to prolonged droughts.

  13. phatti

    April 29, 2010 01:14 PM

    @Jake

    I think the reasons that Thomas doesn’t get mentioned in the all-time great category are:
    1. He was a tremendous hitter, but that was all. Very little defensive or baserunning ability.
    2. He played in the steroid era, so even though there’s no evidence that he was dirty, he’s still downgraded in people’s minds.
    3. the second half of his career was not as impressive as the first half. It was still really good, but the most recent impression is what stays with people.

  14. Jordan21

    April 29, 2010 02:58 PM

    To say the RBI totals are “meaningless” would allow you to exchange a person with Howard in the lineup.
    (Not taking account into health) I think Howard’s RBI totals are a direct reflection of three things 1.Power 2. Place in lineup and 3.Potency of lineup.

    Take out any one of those three and the RBI totals go down. I think what has made Howard elite in the RBI category is the consistency with which he has hit with amazing power.

    He is the fastest player in major league history!! to 100,150 and 200 home runs. Not surprisingly he also lead the majors in RBIs during that span.
    So to call the RBI totals “meaningless” would be to ignore the fact that they do reflect his elite power combined with a potent lineup.
    All the prognosticators who compare Howard to other players from other eras please stop right now. There has never been a player like Howard so we really cant tell.
    What we do know.
    1.Late start on career AGE 26 first full season.
    2.No major injury history
    3.concerned about the holistic aspect of health and fitness

  15. e

    April 29, 2010 03:12 PM

    Howard makes Adam Dunn’s contract look like one of the best in baseball. Dunn has had a higher OPS+ the past two seasons.

  16. Jordan21

    April 29, 2010 03:15 PM

    There is no reason to believe that Howard can not keep hitting home runs at the same pace until he is 33 with a 10-25 percent drop at 34-37 and then a gradual drop until 39. With that said Howard’s average overall RBI totals at age 35 should be in the 116-126 range.

  17. Malcolm

    April 30, 2010 01:53 AM

    It ultimately doesn’t matter whether or not Ryan Howard is as good as Albert Pujols. Albert Pujols will either be with St. Louis for the rest of his career or will sign with some team other than the Phillies for a huge amount of money. I’m not going to go as far as to say that any baseball player deserves $25 million per year, but I feel pretty good about having Howard at first base for 7 years.

    I think it’s just as silly to try and predict what will happen to a player’s career for the next ten seasons based simply on statistics the same way it’s silly to value him completely on RBI. Do fangraphs take into account work ethic, metabolic rate, type and duration of workouts, or ability to adjust to pitchers? How can they? Sabermetrics are useful for some things, but they can’t predict the future or tell you exactly how valuable a player is–nothing can do that.

  18. Bill Baer

    April 30, 2010 02:16 AM

    Malcolm,

    If we are to discard Sabermetrics because they can’t measure “work ethic, metabolic rate, type and duration of workouts” then how do you know that they mean so much that Sabermetrics is missing out by not being able to include them?

    Total mind-blower.

  19. aladou

    April 30, 2010 01:09 PM

    Bill – Ok, ok, you win. We’re convinced already. Howard is at best a slightly above average player who is only going to get worse, and the contract is a monumental mistake.

    So what’s your point? What’s next? What do we do with this information? Maybe we should boo Amaro whenever we see him or hear his name. We haven’t had a lot of practice lately, but I’m sure it’ll come back to us. If Amaro isn’t around, can we just boo Howard instead?

  20. hk

    May 02, 2010 09:03 AM

    Malcolm,

    Among other things, people use sabremetrics to attempt to better predict the future – for instance, FIP and xFIP have shown to be better predictors of future ERA than ERA itself – and to try to better assign dollar values to past production. Nobody knows whether Howard will produce enough to justify the contract. However, we have made enough advances beyond the “old boxscore statistics” that we know it is a bad approach to weigh RBI’s too heavily in the process.

  21. Chareth

    May 03, 2010 08:48 AM

    hk, people don’t just use sabermetrics to predict the future. They also use it to start petty arguments on the interwebs.

  22. Jerxton

    May 03, 2010 01:14 PM

    Bill,

    Want to ask you a question about your feelings, and I’m not sure you’d answer honestly…but do you dislike Howard? Are you angry that the Phils paid him so much?

    It seems like you go hellbent on knocking the guy, and while I appreciate your man crush on Pujols you claim to be a philly fan, and sometimes it doesn’t come across in you writings.

  23. KH

    May 03, 2010 03:08 PM

    Sabermetrics has never really accurately predicted the future unless you consider 60%-70% right in the best cases as accurate.

  24. KH

    May 03, 2010 03:10 PM

    Its much better at giving value to things that already happened. Howard is close to just as likely to be Carlos Delgado, Jim Thome, Willie McCovey, or Willie Stargell as Cecil Fielder or Mo Vaughn.

  25. David

    May 06, 2010 01:04 AM

    For a batter, a big chunk of WAR breaks down into A) the ability to get on base and score runs and B) the ability to advance runners (which may result in other players scoring runs), correct? I realize that in Howard’s case a lot of B involves home runs and fly balls rather than singles and bunts (so perhaps B would need to be broken down further), but to what extent does B – or some major subset of B – correlate with RBIs? Even taking players’ abilities to already be on base and players being arranged in ways that take into consideration A and B it seems there should still be some correlation leftover.

    Which would mean that as long as Howard is left in a reasonable spot in the batting order and keeps hitting home runs (and getting those RBIs), he shouldn’t be too bad of a deal.

Next ArticleVideo: Howard, Ibanez, Halladay Q&A