The Case for Jacoby Ellsbury

Today, we’re writing up affirmative arguments for each of the three primary MVP candidates, and I get to make the case for Jacoby Ellsbury‘s candidacy. On the surface, arguing for Ellsbury is something easy case to make – after all, he led the majors in WAR, and even though it is not a precise measure of value that should be taken as gospel, the guy with the best WAR in baseball is obviously a serious candidate for the award. That said, you don’t have to buy into WAR to see that Ellsbury was probably as good as anyone else who played the sport this year.

Let’s start with his offense. Ellsbury’s offensive production was outstanding across the board. His .321 batting average was fifth in the AL and tops among center fielders, his OBP of .376 was tied for 11th in the AL and tops among center fielders, his .552 slugging percentage tied him for fifth in slugging with Curtis Granderson.

The knock on Ellsbury coming into 2011 was his durability. He took a lot of heat from the media and his teammates in 2010 for playing in only 18 games due to a variety of rib and trunk injuries. In 2011, Ellsbury quieted the critics who labeled him “soft” by playing in 158 games and accumulating a MLB best 732 plate appearances. His high rate stats combined with durability to produce a great set of counting stats. Ellsbury hit 32 home runs (fifth in AL), stole 39 bases (fourth in AL), scored 119 runs (3rd in AL), and even drove in 105 runs (t-6th in AL) despite hitting leadoff for most the year. Ellsbury was also the only player in either league the AL to hit 30 or home runs, steal more than 30 bases, and accumulate more than 100 RBIs.

Moving onto more advanced offensive statistics we more evidence of Ellsbury’s dominance, he was fifth in the league with a .402 wOBA and 150 wRC — in both cases finishing behind Jose Bautista and three 1B/DH types (Miguel Cabrera, David Ortiz, and Adrian Gonzalez).

Ellsbury hit like an All-Star level corner player, and did so while playing center field — one of the most demanding defensive positions on the diamond. Ellsbury not only played center field, he played it at a very high level no matter how you measure defense. He made 0 errors on the year for a perfect 1.000 fielding percentage and led the AL with 388 putouts. Single year advanced fielding metrics can bounce around some, but Ellsbury 2011 UZR of 15.6 was fourth best in the AL and only the Yankees’ Brett Gardner rated as a better defensive outfielder in 2011.

Ellsbury’s 2011 does not look to be a fluke performance either. His BABIP of .336 was only slightly above career average of .325. Hit saw a slight increase in his line drive percentage, but by far the biggest change in his performance is that he saw his power stroke develop, which is not unusual for a player in his prime. Ellsbury’s flyball rate was unchanged in 2011, but more of them left the building (16.7) compared to previous high of 10.3% in his rookie season. Defensively, he has made 2 errors in over 4200 innings and has a UZR/150 of 9.6.

How does Ellsbury’s 2011 compare to the other top contenders? A good case can be made for either Justin Verlander or Jose Bautista, but I think Ellsbury’s case is the strongest. Verlander was clearly the best pitcher in the AL last year, but he only played in 21% of his team’s games. Verlander no doubt affected other games by taking pressure off the Tiger bullpen during his starts, but I prefer MVPs candidates to play more than 34 games.

That leaves us with Jose Bautista. Bautista continued to be one of the most dangerous hitters in MLB. His .302/.447/.608 triple slash line is outstanding and he led MLB with a .441 wOBA and 181 wRC. While his offensive performance was slightly better than Ellsbury’s, he could not match Ellsbury in terms of durability or glove work. He played in 9 fewer games than Ellsbury and had 77 fewer plate appearances. In the field, he made 6 errors and continued to be a below average defender, compiling a -4.8 UZR to give him -46.2 for his career. Ellsbury more than makes up the offensive difference between himself and Bautista with his durability and his above average defense at a premium position.

In total, Ellsbury’s 2011 was worth almost 2.5 more wins that the second best AL center fielder (Curtis Granderson) and more than doubled the WAR output of the third place finisher (Peter Bourjos). Ellsbury’s 2011 season was the best performance by a center fielder since Ken Griffey Jr.’s 1997 season (also 9.4 WAR)—-a season that ended with Griffey Jr. unanimously winning the AL MVP award.

Red Sox fans will forever have a bitter taste in their mouths at the way the 2011 season ended, but perhaps seeing Ellsbury win the MVP will remind them that not all was lost as their home-grown centerfielder developed into one of the best players in baseball.




Print This Post



I am political science professor at the University of North Carolina. I grew up watching the Braves on TBS and acquired Red Sox fandom during the 1986 World Series. My other hobbies include cooking, good red wine, curing meats, and obsessing over Alabama football---Roll Tide! Follow me on Twitter @ProfJRoberts.

46 Responses to “The Case for Jacoby Ellsbury”

You can follow any responses to this entry through the
RSS 2.0 feed.
  1. Flharfh says:

    ” Ellsbury was also the only player in either league to hit 30 or home runs, steal more than 30 bases, and accumulate more than 100 RBIs.”

    *best Rodney Dangerfield impression* “Ryan Braun don’t get no respect…”

    Vote -1 Vote +1

  2. Thanks for pointing that out…fixed now.

    Vote -1 Vote +1

  3. jim says:

    don’t forget that ellsbury kept right on chugging through september during the collapse

    Vote -1 Vote +1

  4. Anon says:

    While his offensive performance was slightly better than Ellsbury’s, he could not match Ellsbury in terms of durability or glove work. He played in 9 fewer games than Ellsbury and had 77 fewer plate appearances.

    PA to help judge durability? Really? Batting position and teammate offense are out of a player’s control and have a large effect on PA. You might as well take away points because his team didn’t make the playoffs.

    Vote -1 Vote +1

    • I don’t disagree on plate appearance, but 9 games is 5.5% of the season. PA just shows that his games played was not inflated by a bunch of late inning defensive work or pinch hitting.

      Vote -1 Vote +1

      • Anon says:

        Ellsbury averaged 4.6 PA per game. He could have started 16 less games (i.e. 10% of the season) than he did and still have a PA advantage over Bautista.

        Look at the fielding stats on baseball-reference, which includes games started, innings played, and complete games (though CG may not reflect durability well depending on manager and team). Fangraphs has similar stats but does not list number of starts at DH.

        Vote -1 Vote +1

      • keyser soze says:

        So what you are saying is Bautista was able to create the same value in less actual playing time? Sounds like he is the one to go with then, especially when you consider his supporting cast vs. Ellsbury’s.

        Pretty hard to criticize Bautista’s durability for missing 9 games for being drilled in the head by a pitch and having a baby. If you want to be honest Ellsbury is the one had not been the picture of health until this season.

        Vote -1 Vote +1

      • LOSH says:

        I wish jose had 4.6 PA per game… damn!

        Vote -1 Vote +1

      • Jonathan says:

        @keyser soze

        “If you want to be honest Ellsbury is the one had not been the picture of health until this season.”

        2010 aside, Ellsbury’s actually had been the picture of health. He played in 145 games in his rookie year (When he was supposed to be platooning with Crisp) and 153 games in 2010. 2010 is pretty clearly the outlier, not 2011.

        Vote -1 Vote +1

  5. Daniel says:

    “Verlander was clearly the best pitcher in the AL last year, but he only played in 21% of his team’s games. Verlander no doubt affected other games by taking pressure off the Tiger bullpen during his starts, but I prefer MVPs candidates to play more than 34 games. ”

    Seriously? We’re back to this pitchers can’t be MVPs argument? Or are you saying that SPs can’t, but closers, sure! Or maybe closers don’t pitch in enough games, but middle relievers…they should win the MVP

    Vote -1 Vote +1

    • I personally don’t think pitchers should win the MVP, but reasonable people can disagree.

      Vote -1 Vote +1

      • BDF says:

        No, they can’t. The instructions for MVP voting explicitly state that pitchers are eligible. That is the context in which reasonable exists. You may *wish* that the MVP award was only for hitters, and you may *believe* that it takes an extraordinary performance the likes of which we have not yet seen to merit a pitcher’s winning the MVP, but ruling it out prima facie qualifies as unreasonable in this context.

        Vote -1 Vote +1

  6. craigtyle says:

    The argument against Verlander on the basis of his only appearing in 34 games should not appear on a serious website.

    Ellsbury 2011 Plate Appearances: 729
    Verlander 2011 Batters Faced: 969

    +9 Vote -1 Vote +1

    • cole0271 says:

      Couldn’t agree more. If you think pitchers should be ineligible for MVP by *definition*, that’s one thing; but the argument that they impact the season less than position players has been thoroughly debunked for years now.

      Vote -1 Vote +1

    • adohaj says:

      The argument isn’t as foolish as you make it seem. Verlander may have contributed to more plate appearances, but Verlander’s contributions are concentrated over fewer games. So Verlander can affect less W/L outcomes.

      Vote -1 Vote +1

      • cole0271 says:

        He was third in the league in WAR. MVP case aside (I’m a Bautista man myself), the argument that he can’t be as valuable as position players is silly on its face.

        Vote -1 Vote +1

    • Jaksiel says:

      The argument that plate appearances and batters faced are equivalent should not appear on a serious web site.

      Vote -1 Vote +1

  7. gabriel says:

    Y’all have two good advanced defensive metrics, UZR & DRS. If you substitute DRS for UZR, Bautista and Ellsbury reverse their WAR ratings, so I’m unconvinced that the headline WAR values should leave anyone with a presumption in favour of Ellsbury.

    +7 Vote -1 Vote +1

  8. Evan says:

    I still don’t know why WPA/LI is left out of MVP formulas or debates on this site. Context based stats belong in the MVP discussion.

    Vote -1 Vote +1

  9. Jason says:

    Ugh, WAR must be the single worst baseball statistic. As a concept it is fine. It really would be nice to have a single statistic to measure player value. The problems are that the statistic is poorly documented and poorly calculated. What is worse is that the statistic is abused by the very people that advocate it.

    There are two centerfielders in the AL that might be considered for the award. One is dismissed out of hand on this site because his WAR is significantly lower than the other’s. That seems reasonable until you ask why his WAR is lower. After all both players have essentially identical OPS.

    The difference in the two player’s WAR is accounted for by 40 extra plate appearances for the player who bats leadoff and by a large UZR advantage. WAR is a counting stat. Is there any reason to prefer leadoff hitters over all all other hitters in your MVP consideration? Of course not. Leadoff hitters should get no a priori preference, but that is exactly what you do when you use a stat like WAR to determine value.

    Second, WAR includes UZR. Nobody ought to treat UZR as anything other than a random number generator until the authors of the statistic release the data and release how they calculate it. We, as consumers, need to be able to estimate the error in the statistic before we can use it for comparison purposes. Should we dismiss one player’s MVP candidacy simply because of error in an estimate? Me may very well be doing that. In fact it is very likely.

    WAR really should not be used at all to evaluate MVP candidates. It is a nice concept, it is just flawed in execution and interpretation.

    +5 Vote -1 Vote +1

    • Anon says:

      Well said.

      The PA differences would make a great study. Batting order and teammate offense have large effects on PA, and this cascades to all other counting stats (WAR included).

      Vote -1 Vote +1

    • Ian R. says:

      “Is there any reason to prefer leadoff hitters over all all other hitters in your MVP consideration? Of course not.”

      Uh, yes there is. Leadoff hitters get more plate appearances than anyone else on the team. All other things being equal, that indicates more value. It’s just that most teams bat their best hitter 3rd or 4th, so the players who contribute the most don’t always lead the team in PA.

      “Ugh, WAR must be the single worst baseball statistic.”

      Look, I’m not saying WAR is perfect. But… really? WAR is worse than RBI? Or pitcher wins? It’s a damn good statistic.

      Vote -1 Vote +1

      • Jason says:

        So you believe that given a choice between two players exactly identical in talent and production, the one who bats higher in the lineup ought to win the the MVP. I don’t see how a manager’s decision about how to structure his lineup ought to effect the MVP voting. ….this is especially true given the fact that nobody who actually gets to fill out a lineup card bats their hitters from best to worst.

        RBI and pitcher wins are a strawman. I’m not arguing MVP voting should be based on either. I’m just pointing that WAR has significant limitations when used to compare between players (which is its only purpose). You might say its a “damn good statistic” but I want to know how good it is before I use it to make determinations. It is standard practice to quantify how good your statistic is. I want to know the error so that I can assess whether a UZR of +10 is actually significantly better than a UZR of -10 (I actually doubt it is). I want to know if a WAR of 7 really is better than a WAR of 5. I’m really not asking a lot here!

        Vote -1 Vote +1

    • Salo F says:

      WAR is not the worst stat in baseball but you are right about the PA argument.

      People like me, who like WAR, tend to dismiss RBI because it depends on line up spot. But we dont discuss that Elsbury had more PAs and therefore the chance to accumulate more WAR because of that

      Vote -1 Vote +1

    • Eric M. Van says:

      You’re aware of the fact that MGL cannot legally release the data underlying UZR, aren’t you? And that he actually has explained how it’s calculated, within that limitation?

      You’re also aware, aren’t you, that UZR is not the only available defensive metric? UZR has Ellsbury 15.6, Granderson -5.1. DRS has Ellsbury, 6 to -15. Total Zone has Ellsbury, 4 to -2. BP’s FRAA has Ellsbury, 11.6 to -13.2. So, no, it’s not actually “likely” that the possibility that Granderson was more valuable than Ellsbury was dismissed because of an “error in estimate.”

      Finally, if there were two players with identical performance, one who hit leadoff and one who hit 4th, and they had the same PT but the leadoff guy of course had more PA and hence more value as measured by any metric … yes, the leadoff guy is the MVP. *He had more value.* It’s not an award for “best.” Imagine a guy with slightly better performance per game who was inexplicably benched at random 25 times by an idiot manager … you couldn’t give him the MVP because he *should* have been more valuable had he been given the opportunity. If one manager gives a player more opportunities to contribute value than another player got from his manager, you shouldn’t adjust that away. The value accrued was real.

      This truth doesn’t really give leadoff hitters an unfair edge, because traditionally they require a tool set that makes it extremely unusual for them to match a #3 / 4 hitter in per-game offensive value. There are sound reasons for wanting speed as well as OBP from your #1 and #2 hitters, and if a guy runs well enough to hit leadoff for the average manager and can slug .552 while he’s at it, I don’t see why he doesn’t deserve credit for the extra value accrued in the additional PA.

      Vote -1 Vote +1

      • Jason says:

        Yes, I am aware of how little is publicly known about UZR. The legal issues don’t concern me. I am not willing to “just trust it” and neither should anyone else. The least they could do is estimate the error. You let them off the hook by accepting whatever they give you blindly.

        I am aware that there are other defensive metrics. However, they are not at all independent. So, the fact that they somewhat agree is not evidence against random chance. All the measurements might simply be capturing the same randomness. If you look at independent data sets (both players from past years) you will see that both players have been good and bad. Maybe their talent really does fluctuate wildly from year to year. Maybe rare stochastic events in the data are driving the analysis. If difficult chances are rare (which they are) and if difficult chances disproportionately effect these calculations (which as far as I can tell they do) then perhaps Ellsbury just happened to get a disproportionate share of pseudo-difficult chances (i.e. chances rated difficult but that actually almost every player makes). When you force a continuous distribution to be discrete these types of errors are guaranteed. When the variation within a player is on the order of the variation between the players you are trying to compare there is either no variation between the players (we know this isnt the case) or your measurement sucks (my vote). There is a good chance we are trying to weigh feathers using a bathroom scale…

        I won’t argue about “value”, but my opinion is that MVP voters should not a priori bias their votes by defensive position and by lineup priority. WAR does this so it doesn’t enter my personal thinking about MVP candidates. ….I don’t have a vote so it really doesn’t matter.

        Vote -1 Vote +1

    • Theo says:

      I agree about WAR being the single worst baseball statistic. Great concept, but it’s full of assumptions. It’s a lazy person’s statistic.

      Vote -1 Vote +1

  10. David Carter says:

    How ironic that Youkilis was so vocal about Ellsbury’s “softness” last year. This year, Youk was often injured, and was nowhere the player he once was. It would be no surprise if Youk found himself playing for another team in 2012.

    Vote -1 Vote +1

  11. anon says:

    When should we be expecting the case for Miguel Cabrera?

    Vote -1 Vote +1

  12. Tom says:

    Why is Pedroia not a “prime” candidate? Why is fangraphs playing into the narrative of the media? He’s a full win ahead of Verlander, which I think is enough to at least put them on equal ground.

    Vote -1 Vote +1

    • Jason says:

      In the real world, how could Pedroia ever be a serious candidate when Cano outperforms him on almost every meaningful measure? Second basemen from third place teams aren’t given serious MVP consideration when the two teams finishing ahead of him in the standings both had better second basemen.

      Vote -1 Vote +1

      • Eric M. Van says:

        Why do I get the feeling that if the team you root for had the players with the great fielding stats instead instead the subpar ones, you’d just love UZR and WAR?

        Pedroia and Cano: FG has Pedroia with a 134 to 133 edge in RC+, and a 57.6 to 56.0 run edge in offensive value. B-Ref has it 61 to 51. Only BP has Cano ahead, 55.0 to 53.3. So much for Cano outperforming Pedroia on “significant” measures. As for the ones you apparently think don’t matter, UZR has Pedroia with a 17.9 to -3.0 edge, DRS has Pedroia ahead 13 to 6, Total Zone has Pedroia ahead 9 to -2, and again only BP’s FRAA has Cano with an edge, 6.0 to 2.0. Median or mean of the four measures, that’s a 9 run edge for Pedroia. Functional human eye / brain combinations have Pedroia with a comparable edge as well.

        +5 Vote -1 Vote +1

      • Jonathan says:

        “In the real world, how could Pedroia ever be a serious candidate when Cano outperforms him on almost every meaningful measure?”

        So slugging percentage is the only meaningful measure? Because that’s about all Cano had on Pedroia this season.

        Pedroia: OBP: .387, wOBA: .377, OPS+: 131, K% 11.6%
        Cano:OBP: .349, wOBA .375, OPS+: 129, K% 14.1%

        The only thing Cano definitively outclasses Pedroia in in 2011 is power numbers. Pedroia hit for better average, got on base more, struck out less, stole more bases and played better defense by just about every metric available. Even throwing out the nearly 2 1/2 win WAR advantage Pedroia has, it’s really hard to argue that Cano was better than Pedroia this year unless you base your entire argument around seven extra home runs and nine extra doubles being more valuable than a nearly forty point difference in OBP, a better K rate (Eleven less strikeouts in over fifty more plate appearances), elite defense and better baserunning.

        Vote -1 Vote +1

      • Jason says:

        Eric M. Van,

        You are wrong about that. I reject WAR and UZR on scientific grounds. I never look at them of my own accord, so I don’t even know where the players on my team of preference rank.

        I’m willing to bet that if you polled 30 big league managers all 30 would choose Cano over Pedroia. Cano is a hitter no manager wants to face. He is feared as the best hitter in one of baseball’s best lineups. Pedroia is a really good player but he isn’t feared. On his own team Gonzalez, Youkilis, Ortiz and now Ellsbury are the worries. If you swapped them it would be a similar situation. Pedroia would be behind Teixiera, Arod Granderson and perhaps even Swisher. Cano, however, would be right there just after Adrian Gonzalez.

        Vote -1 Vote +1

      • Jonathan says:

        “I’m willing to bet that if you polled 30 big league managers all 30 would choose Cano over Pedroia.”

        And then they’d go give Derek Jeter a seventh Gold Glove, despite the fact he’s a terrible fielder by every metric imaginable and every scouting report you could find. You overestimate the intelligence of managers.

        You’re operating off of rationales that even Joe Morgan would be embarrassed using.

        Vote -1 Vote +1

      • Jason says:

        Jonathan,

        I have no problem accepting that managers no a little bit about baseball. …I also think that if managers ignore defensive metrics it is probably because they intuitively know better and they may very well be right! I often times like to listen to the perspectives of players and managers. They have played and watched enough baseball that they see things and understand things that we don’t. I don’t mean to imply that there is some voodoo going on that can’t be quantified. To the contrary, we ought to be able to do a better job with quantification. And the best way to get new insights in to how and what to quantify is to listen to the people that really know the game.

        Vote -1 Vote +1

      • Jonathan says:

        So you choose to stand pat on an argument of “I don’t care that there’s quantifiable evidence that Cano was not as good as Pedroia in 2011, I’m going to use conjecture to conclude that Cano was better than Pedroia.”

        You’re basically supporting conjecture with MORE conjecture. Either I’m being meticulously trolled or you just have no idea what you’re talking about.

        Vote -1 Vote +1

      • Jason says:

        Jonathan,

        When I look at the two player’s batting lines I see a huge power advantage for Cano. In absolute terms Cano outpowered Pedroia across the board despite 50 fewer plate appearances. That is a huge advantage for Cano. Pedroia’s only real advantage is walking more. I will take all the extra doubles, triples and homeruns over the walks. They lead to more runs.

        Ask yourself how Cano was able to score more runs than Pedroia despite fewer plate appearances, batting lower in the lineup, and a lower on base percentage. I think it is because of all those extra base hits that Cano produces and Pedroia doesn’t. ….he’s also a damn good baserunner despite people’s perceptions of him.

        Vote -1 Vote +1

  13. Tom says:

    “In the real world, how could Pedroia ever be a serious candidate when Cano outperforms him on almost every meaningful measure? Second basemen from third place teams aren’t given serious MVP consideration when the two teams finishing ahead of him in the standings both had better second basemen.”

    lol

    how do you feel about the fact that Pedroia just had a season greater than any Derek Jeter has ever had

    Vote -1 Vote +1

    • Theo says:

      There’s no way to validate WAR or UZR. You must be “Drinking the Kool-Aid” of the cult known as Fangraphs.

      Vote -1 Vote +1

    • Jason says:

      I get the impression that good seasons for Jeter end with a ring on his finger and not an historic collapse. ….by the numbers, this comment is just silly.

      Vote -1 Vote +1

  14. Tom says:

    No, I’m just giving the troll a taste of his own medicine. Thanks, though.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current day month ye@r *