Yoenis Cespedes’ Run-Saving Right Arm

Yoenis Cespedes‘ defense currently ranks third in the MLB, according to UZR/150. Here’s how the third-most valuable fielder in baseball sometimes likes to play routine outfield grounders:



How in the world is this guy third-best defensively in anything? Because this is how he makes up for those gaffes:




Those throws happened in back-to-back games on Tuesday and Wednesday night. Cespedes is now the league leader in outfield assists, with nine. Nobody else in the MLB even has eight. What’s more impressive is that eight of Cespedes’ nine outfield assists have come in the last three weeks. Yoenis Cespedes has more outfield assists in the last three weeks than any other outfielder has all year.

And four of them have come against the Angels. They’ve had enough:

Although it took Cespedes until the middle of May to register an outfield assist, his strong arm should not come as a surprise to anyone. Looking back through scouting reports from 2012 when Cespedes defected from Cuba, you can see the arm strength repeatedly touted as one of his best tools:

From Baseball Prospect Nation:

Shows solid arm strength that could support right field. Can take a while to get rid of the ball and will need to quicken release and improve accuracy. Might be a half tick better than average. Grade – 50/50

And from Baseball America:

Cespedes has a 60 arm, which would be a weapon in center field and plenty to play right field if he loses a step or for a team that wants to sign him but already has a plus defensive center fielder. His arm stroke isn’t fluid, as it’s shorter than most outfielders and gives his throwing mechanics some funkiness. Regardless of how he does it, his throws have plenty of carry and scouts have generally been pleased with his accuracy.

“I don’t care how it looks,” said one scout, “as long as it gets there and gets guys out.”

Well, it’s getting there, and it’s getting guys out.

As previously mentioned, Cespedes is third in the MLB in UZR/150, trailing only Alex Gordon and Jason Heyward, with an outlandish total of 30.1. For reference, Manny Machado‘s amazing 2013 defensive rookie campaign carried a UZR/150 of 31.8. Cespedes has already amassed more defensive value in left field this season (6.2 UZR) than he did his first two seasons combined (4.4 UZR). And it’s come almost entirely from his arm.

Ultimate Zone Rating consists of three defensive components for grading outfielders. A range component, an error component, and an arm component. Take a look at how Cespedes has graded out this season:

Player Range Error Arm UZR
Yoenis Cespedes 1.1 -0.4 5.5 6.2

Cespedes’ range has been slightly above average. The same scouts who lauded Cespedes for his arm also mentioned that he takes less-than-optimal routes but that his speed helps make up for it. His one fielding error puts him just slightly below average and it’s obvious in the first two GIFs leading this piece that Cespedes can be a bit clumsy in the field. But, like his speed making up for his route running, his arm strength can make up for his misplays in the field. His arm alone has already been worth five and a half runs this season, tops in the league and equal to over half a win in WAR.

So, how else, other than those two throws against the Angels, has Cespedes accumulated his basically arm-only top-three UZR total?

To the GIFs!

5/14 – Gordon Beckham


The tomfoolery begins with a little help from Eric Sogard. This is the only one of Cespedes’ nine outfield assists in which a relay throw was necessary. Conor Gillaspie smacks a two-out double off the wall in left field. Beckham tries to score from first to tie the game, but Cespedes plays his home wall in Oakland perfectly and delivers a dart to Sogard, who in turn delivers a dart to catcher Derek Norris. One down.

5/23 – Brett Lawrie


Lawrie leads off the fourth inning with a single down the left field line. Cespedes actually takes a pretty clean route to this ball, but the most impressive thing is his plant. Cespedes saves himself a split second – the difference here between Lawrie being safe and out – with an efficient one-step plant and throw. He puts it right on target to second base and Nick Punto makes a nice tag to erase Lawrie’s leadoff hit.

5/25 – Jose Reyes


Two days later, in the same series against Toronto, Cespedes guns down one of the fastest players in the MLB in Jose Reyes. The bases are loaded with no outs here and Jose Bautista singles to left field. One run already scores on the hit and Edwin Encarnacion is on deck. Jose Reyes really has no incentive to test Cespedes’ arm, but does so anyway. Mistake.

5/28 – Miguel Cabrera


Here’s Miguel Cabrera, decidedly not one of the fastest players in the MLB. Cespedes tries to play the bounce off the fence, but instead the ball dies when it hits the wall. Cespedes makes up for this by barehanding it, saving himself a split second in the process. He hits Sogard on one bounce and nails Cabrera.

There were two outs in the top of the ninth inning when this happened. Cabrera tested Cespedes’ arm for the sake of potentially manufacturing an insurance run. Instead, he failed and the inning was over. Josh Donaldson hit a three-run homer in the bottom of the ninth and the Athletics won.

5/31 – Chris Iannetta


Here’s where things really start to get interesting for Yoenis Cespedes. This is his fourth outfield assist in eight days. The score was tied 0-0 with no outs in the top of the second. Again, I’m not sure why Cespedes’ opponent tried to test his arm so early in the game with no outs. Had Iannetta stayed at third, the bases would have been loaded with the top of the Angels order coming up and no outs. Instead, Iannetta went for it and didn’t get it.

Pay attention to the date, score, and more specifically, the inning of this next clip.

5/31 – Kole Calhoun


Yup. Yoenis Cespedes’ fifth outfield assist in eight days was also his second in three batters. This one makes a little more sense, given there were two outs, but the circumstances don’t matter. Test Yoenis Cespedes and you lose.

The Angels sent five batters to the plate this inning. Four singled, one struck out. Nobody scored, thanks to Yoenis Cespedes. As Mark Simon of ESPN pointed out, Cespedes was worth five Defensive Runs Saved in this game alone. Only 13 outfielders have accumulated more than five DRS all year.

6/6 – Chris Davis


Something must have been wrong with Yoenis Cespedes. He nearly went a whole week without an outfield assist! One thing worth noting in all of this: Cespedes’ teammates have done a great job helping him out with athletic tags.

There were two outs in the bottom of the eighth here and the score was tied 3-3. When you consider the situation, it’s not a bad idea for Davis to try to stretch this single. When you consider who’s playing left field, it starts to become a bad idea.

Yoenis Cespedes has been one of the most valuable defensive outfielders in baseball this season. Let me rephrase that: Yoenis Cespedes’ right arm has been one of the most valuable defensive outfielders in baseball this season. He has gunned nine runners on the basepaths. Five have come at home plate. He has more outfield assists in the last three weeks than any other player has all year.

The Angels and their Twitter account may have had enough of him, but the rest of us surely haven’t. Keep on firing, Yoenis.

Print This Post

August is an associate reporter for MLB.com, covering the Cleveland Indians. He previously covered the Indians, Browns and Cavs for the Akron Beacon Journal and ohio.com. He tweets often about the Indians, assorted nerdy baseball things and also other stuff, too. He'd like it if you followed him on Twitter @AugustF_MLB, but you don't have to.

49 Responses to “Yoenis Cespedes’ Run-Saving Right Arm”

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

    It’s hard to blame players for taking extra bases when there’s no chance of getting thrown out. Someone needs to start tracking Cespedes’ outfield-grass-to-catcher’s-glove time.

    Vote -1 Vote +1

  2. Dave says:

    That last one showcases some great range, too. Actually, a lot of them do. Whether he’s charging in hard or cutting balls off laterally, he often uses his legs to put himself into a good position to showcase his arm.

    Vote -1 Vote +1

  3. Nick D. says:

    I think his arm has been worth more than 5.5 runs. The reason is looking at the base/out matrix, if his throw to home plate the other night wasn’t so precise, the Angels would have scored 1 run and have a runner on 3rd with 1 out, which gave them an approximate .897 more runs they should typically score. Because of the throw, their was 0 runs scored and a runner on 3rd with 2 out, which gave the Angels an approximate .382 runs they would score. That swing is 1.52 runs. I would love to know what the WPA swing was, as that was also in the 8th inning of a tie game. We know what the WPA would was after the play, what would the WPA have been if the Angels(home team) was leading 2-1 with a runner on 3rd and 1 out in the 8th inning? That might be a half win swing in WPA all by itself. Not to mention all of the other throws he has made this year.

    Vote -1 Vote +1

    • Josh says:

      Interesting. Is it possible to account for all possibilities in a defensive measurement like that? Also, is there any measure for a throw that prevents an advance? Or does the “arm” calculation only take account of outs?

      Vote -1 Vote +1

      • From the FanGraphs UZR Primer:

        “They are based on the speed and location of batted balls to the outfield and how often base runners advance extra bases (advances), don’t advance the extra base (holds), or get thrown out trying to advance (kills). Park factors are used in arm ratings. For example, because the left fielder plays so shallow in Fenway and balls tend to quickly ricochet off the Green Monster, it is difficult to advance an extra base on a hit to LF in Boston. In Colorado, because the OF is so expansive, base runners advance more easily than in an average park. The UZR “arm engine” adjusts for those things.”

        Vote -1 Vote +1

  4. Josh says:

    The first two GIFs are reminiscent of Delmon Young’s Twins days. If only he wasn’t Delmon Young…

    Vote -1 Vote +1

  5. Johnston says:

    I have been watching all of the Athletics’ games this season, and Cespedes’ arm has been absolutely amazing.

    Vote -1 Vote +1

  6. Brian says:

    There is a part of me that is conflicted about UZR and its relation to actual on-field events versus components.

    If I understand it right, Cespedes would have a much worse Arm rating had all of those players decided not to challenge him. By definition, then, the better and more noteworthy and famous your arm strength, the lower your ceiling for this stat.

    Yes, gunning down a player at home is way more valuable than having the reputation to make the player hold up at third. But gunning down a player at home, for instance, once in a month, is so ephemeral compared to five or six players who might next year say “that’s Cespedes, I’m stopping.”

    Pitcher WAR on fangraphs is based off FIP, and componenents, while ERA is what actually happened on the field and is ignored. It confuses me that UZR/Defense/Arm ratings would take a very different tack. Just thinking aloud.

    Vote -1 Vote +1

    • EthanB says:

      Which part of FIP didn’t actually happen on the field? All of WAR is only based on the run values of things that happened.

      Vote -1 Vote +1

      • Brian says:

        FIP awards WAR on the basis of merit/worthiness, not how you performed.

        If you go 6ip, 3er, 0bb, 15K, you earn more WAR than a guy who pitched a shutout with bad peripherals. That’s just stupid.

        That’s like basing offensive WAR on xBABIP rather than actual offensive production.

        Vote -1 Vote +1

        • a eskpert says:

          Because most research indicates that the person with Randy Johnson performance was unlucky, whereas the other person was relatively lucky in how their batted balls turned into outs. The basic idea in DIPS theory is that pitchers don’t have very much control over batted ball outcomes.

          Vote -1 Vote +1

        • Brian says:

          Who cares? WAR isn’t about projecting the future or awarding gold stars.

          If someone shitty like Jaime Garcia or Kyle Lohse or Joel Pineiro puts up a season with a 2.25 ERA and bad peripherals, he helped his team win a LOT of games.

          I don’t care if Strasburg struck out the world if he had a 3.50 ERA.

          WAR is about assigning wins, not brownie points.

          Would I take Strasburg over Lohse going forward? Yes. Every time. But as long as WAR is attached to FIP, its dumb.

          Vote -1 Vote +1

        • AK7007 says:

          Brian, fangraphs WAR uses FIP because ERA is entangling nine players together in it’s assignment of runs and then pinning it all on the pitcher. Garcia/Lohse/Pineiro probably don’t put up great ERAs with bad peripherals unless their defense is helping them out. Put their exact seasons instead in front of, I dunno, this year’s Yankees, and watch out.

          The truth is, pitchers have more control than FIP assumes. So that means that the actual value of a pitcher is somewhere between FIP and the arbitrariness that is ERA. It’s fine to acknowledge the limitations of FIP based WAR, but don’t be a mindless slave to runs against WAR (that’s baseball-reference WAR). There is probably a middle ground truth out there. Not that anybody in the public sphere knows exactly where that is though.

          To sum up then, mindlessly using fielding independent numbers is dumb. Mindlessly using runs against numbers is dumb. Thinking a little bit and figuring out that WAR as a concept isn’t dumb even if both fan graphs and baseball reference aren’t exactly accurate down to the decimal place, and can disagree on players with extreme good or bad luck? That’s smart. Be smart, don’t discount entire schools of thought before you even understand them.

          Vote -1 Vote +1

        • Bo Jackson says:

          But those aren’t wins that they got. Those are wins that the defence behind them got, or the hitters failed to get.

          Vote -1 Vote +1

        • truth says:

          “Would I take Strasburg over Lohse going forward? Yes”

          Yo genius, that’s exactly why we use fip to gauge a pitcher’s performance, because it tells you who the better pitcher world be going forward.

          Vote -1 Vote +1

        • Za says:

          Hey Brian, please look up “RA9 WAR” which is what you’re looking for. The difference between it and standard fWAR is significant for guys, like Kershaw, who routinely pitch to lower ERAs than one would predict based on FIP and xFIP alone.

          Vote -1 Vote +1

        • John C says:

          The RA9 numbers are what you’re looking for.

          For the vast majority of pitchers, the FIP-based WAR is pretty close to accurate. There are some pitchers, like Matt Cain and Jeremy Guthrie, who seem to have the ability to out-pitch their peripherals year after year, but for most pitchers, an ERA and an FIP that are out of whack is just evidence of good or bad luck.

          Kyle Lohse is a good pitcher, but he needs a good defense behind him or he’ll get killed, because he needs his defense to get the outs. Whereas Stephen Strasburg, striking out an extra four men per 9 innings, asks much less of his defense and therefore has more ability to win on a bad team.

          Vote -1 Vote +1

        • Brian says:

          WAR is a judgment on the past, not the future, just like every other statistic we have that measures something that happened.

          Batting average, WHIP, WPA, wOBA, they all measure things that happened with different levels of predictive usefulness for the future.

          I understand WAR – but @bojackson and @truth are missapprehending my argument.

          @John C thanks for your response. Let me just respond quickly to it:

          “For the vast majority of pitchers, the FIP-based WAR is pretty close to accurate. There are some pitchers, like Matt Cain and Jeremy Guthrie, who seem to have the ability to out-pitch their peripherals year after year, but for most pitchers, an ERA and an FIP that are out of whack is just evidence of good or bad luck.

          Kyle Lohse is a good pitcher, but he needs a good defense behind him or he’ll get killed, because he needs his defense to get the outs. Whereas Stephen Strasburg, striking out an extra four men per 9 innings, asks much less of his defense and therefore has more ability to win on a bad team.”

          Kyle Lohse would probably need a better defense behind him in the long haul in order to put up numbers that would be comparable to Strasburg. This is self-evident.

          What I’m talking about is the fact that WAR measures wins added. If Lohse throws a complete game shutout where every out was a laser line drive right at a fielder, you know what? That’s still a CGSO and extraordinarily valuable to his team.

          Again, my point is that WAR shouldn’t be based on FIP. I love FIP. FIP should exist, as should xFIP and SIERA and all the others. But WAR should be distributed on ERA, just like offensive WAR is distributed on what actually happened, not what should have happened.

          Vote -1 Vote +1

    • neck wattle says:

      I think UZR accounts for bases not attempted due to arm reputation, but I’m not sure how, and I’m also not sure if it ends up being equivalent in value.

      Vote -1 Vote +1

      • neck wattle says:

        Ah, see August’s post above – the average rate for attempting extra bases is computed according to park and players are credited with the deviation from it.

        Vote -1 Vote +1

        • Brian says:

          Right yes- so players NOT running counts + your Arm factor, but not nearly as much as throwing a guy out. It just doesn’t seem rightly balanced between merit/reality.

          Vote -1 Vote +1

        • AK7007 says:

          Brian, that’s because the marginal run value of holding a guy back a base isn’t nearly as valuable an out. It might feel all funny to your gut, but it represents what happened on the field. It’s rightly balanced with regard to the number of runs a fielder prevents. What happens on the field isn’t about merit – it’s about what actually happened. (I know, obvious is obvious)

          Better teams lose to worse teams not because merit, but because reality. Good fielders make bad plays and bad fielders make good plays – again, not about merit. And sometimes, guys with bad arms make good throws, and guys with great arms don’t get challenged. We credit them based on what happened, not “well, he only got challenged and made a great play because his arm strength sucks, so dock that from his defensive numbers.”

          Vote -1 Vote +1

    • The Hashtag Kid says:

      “By definition, then, the better and more noteworthy and famous your arm strength, the lower your ceiling for this stat.”

      Players ran on Jeff Franceour all the time – and seem to continue with guys like Alex Gordon too (those are just examples of guys with large assist numbers that I’ve also had the pleasure to watch/measure myself). I’m not sure a player’s reputation for arm has as much to do with whether they get run on or not as other factors.

      Vote -1 Vote +1

  7. StrikeThree says:

    Great throw! Terrible slide! What’s the in-deck hitter doing? He’s to the right of the plate, waving his right hand “hurry up!” Should be in line with throw, facing runner, to signal where, how to slide. An exaggerated lean to his left, on his knees, hand stretched out would have given the runner to know he needed to make an extra effort to avoid the tag, rather than coming in straight, sitting up, giving his shoulder to the catcher for an uncontested tag. Sliding wide, reaching back with his hand might…might have prevented extra innings.

    Vote -1 Vote +1

  8. Bo Jackson says:

    My arm > Yoenis Cespedes’ arm.

    Vote -1 Vote +1

  9. Tony says:

    The two latest throws are certainly fantastic. However should he really get full credit for throwing out the runners since they wouldn’t have attempted to advance if he hadn’t misplayed the balls so badly? Great throws though. My arm hurts just thinking about throwing the ball that far.

    Vote -1 Vote +1

    • AK7007 says:

      Maybe you could try to dock him the marginal run value of a runner advancing, and then re-credit him for the out. But that would be confusing and need to be done on a manual, case by case basis (i.e., that was an error followed by getting the run back on the throw or something) – I think things are fine as is.

      Vote -1 Vote +1

      • NathanLazarus3 says:

        What Cespedes did In those situations was more valuable than cleanly fielding the ball. By mishandling the ball, he changed, in one case, a clear Pujols double to an attempted triple. Then he changed an attempted triple into an out. In all, his actions totalled a change from a double to an out. Over the course of the season, he’ll likely be docked for fielding mistakes that give runners extra bases, but the plays against the Angels were strongly positive in value. For the same reason, a player who baits runners into trying to advance and then guns them down is more valuable than a guy whose arm is so intimidating that no one tries to advance. If Cespedes fields balls more cleanly in the future and if runners are more intimidated by him then these putouts won’t happen in similar situations in the future, and that will cost him value (though he might add value through intimidation and through putouts on cleanly fielded balls).

        Vote -1 Vote +1

  10. jim fetterolf says:

    The throw in the first gif above was lollipop, lot of arc, little speed. His arm couldn’t carry Jeff Francoeur’s arm’s jockstrap.

    Vote -1 Vote +1

    • Za says:

      It’s been calculated at having left his hand at well over 90 MPH. Wouldn’t exactly call that a “lollipop”.

      Vote -1 Vote +1

      • jim fetterolf says:

        Look at some of Frenchy’s videos. Yo had a lot of elevation compared to Frenchy’s great ones, the best of which came out of the gun at 95mph while he was moving to his left. Hit 3B’s glove on the fly and Peralta slid into the waiting tag.

        Vote -1 Vote +1

    • Wally says:

      It had to have arc, since, ya know, it went >300 feet on the fly. The foul pole is at 340 and he’s maybe 10-15 in front of it…

      And what’s even better about that throw is that Yo is off balance. He picks it up, chucks it and stumbles to his left all in one motion. No crow-hop or even a sold planting of the back foot. This guy could probably throw the ball out of the park from home plate in most ballparks.

      Vote -1 Vote +1

  11. riouxes says:

    Also mishandles a ball here: https://www.youtube.com/watch?v=hMPcfuCyBQM
    …only to still make the play.

    Vote -1 Vote +1

  12. scatterbrian says:

    I feel like Gary DiSarcina shares some of this.

    Vote -1 Vote +1

  13. Lucas says:

    Is there a problem with Cespedes’ DWAR and UZR? His UZR/150 is among the league leaders but his UZR is middle of the pack despite having played a similar amount of innings to those players ranked near the top.

    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>