The Thing About Josh Donaldson’s Defense
As you by now are well aware, Josh Donaldson was traded to the Toronto Blue Jays over the weekend in a blockbuster deal that sent Brett Lawrie back to Oakland. The Blue Jays gave up Lawrie and a few prospects to immediately get better, because Josh Donaldson is a guy that immediately makes any team better. Over the last two years, only Mike Trout and Andrew McCutchen have a higher WAR than Donaldson, and Donaldson’s been three wins better than the next-best third baseman. Donaldson can hit, he runs pretty well for a third baseman, and he’s good with the glove. Add those up and you’ve got a hell of a player.
But there’s something to that last point — that he’s good with the glove — that’s been on my mind for awhile. It’s something I was going to write about when the Gold Glove winners were announced, but then Donaldson didn’t win, so I saved it for another day. Now that Donaldson is back in the news, today is that day.
Donaldson, by almost all accounts, appears to be elite with the glove. Since the beginning of 2012, only Manny Machado, Nolan Arenado and Juan Uribe have more Defensive Runs Saved at third base. Only Machado, Arenado and Chase Headley have a higher UZR. Only Machado, Adrian Beltre and Evan Longoria have scored higher on the Fans Scouting Report. It’s not a stretch to call Donaldson a top-three defensive third baseman, by the numbers we have. It’s safe to call him top-five.
The reason Donaldson grades so well in the field is his range. He can do this:
And he can do this:
Despite the range, despite the numbers, and despite his main competition in Machado missing most of the season, Donaldson didn’t win a Gold Glove in 2014. When the MLB Network crew were discussing potential Gold Glove winners on September 10, Harold Reynolds said he “would throw up if Josh Donaldson won a Gold Glove.” The reasoning behind these two items are the same: errors.
Donaldson committed 23 errors in 2014, topped by only Ian Desmond and Pedro Alvarez. Errors aren’t something we typically think about because we have far better methods to evaluate defense in this day and age. But 23 is a lot and it’s hard to ignore. And what’s more is how Donaldson racked up these errors.
In 2002, errors began being separated into two buckets: fielding errors and throwing errors. Fielding errors are what many of people take issue with. They’re subjective and they’re circumstantial. Guys with great range are sometimes penalized for almost getting to balls that others wouldn’t get to, while guys with poor range sometimes avoid errors by getting nowhere near potentially playable balls. Throwing errors feel like they hold more weight. They’re less subjective, and arm accuracy is an undeniable skill.
In 2014, Donaldson was right in the middle of the pack with six fielding errors. That leaves 17 of his 23 errors to be of the throwing variety. Only two other third baseman had more than eight. Dating back to 2002, Donaldson’s 17 throwing errors are the third-most in a single season by any player.
With Donaldson, you saw throws like this:
Some like this:
And some like this:
This isn’t an issue that’s new to Donaldson, either. He had 10 throwing errors in a full season’s work in 2013. He had six in just half a season’s work the year prior. It was an issue his first couple years, and this year it was elevated to a full-blown problem. Committing 17 throwing errors in one season is a ton, and it’s a cause for concern. In order to get a sense of what that level of concern is, let’s revisit our leaderboard of single-season throwing errors since 2001. For the guys who had seasons as egregious as Donaldon’s: was there a previous issue? Did it continue? What was their future at third base?
- We’ll start with Mark Reynolds, who had 18 in 2008. He was bad as a rookie in 2007, committing eight in just 100 games, but things got really bad the next year. It never got that bad again, but he also never got over his throwing issues. He committed 27 more throwing errors in his next three seasons and was eventually moved off third base, becoming primarily a first baseman.
- Next we’ve got Chad Tracy. He committed 16 in his rookie year, which led the Diamondbacks to move him off third base. He played primarily first base the next season and even got some time in the outfield before returning to the hot corner for his third season. He committed 13 more throwing errors that year and was primarily a first baseman after that.
- Edwin Encarnacion appears on the list twice, as he committed 16 throwing errors in both 2006 and 2008. His double-inclusion in the top six is evidence enough that he never got over his throwing issues, as are his 11 throwing errors in limited action in 2010 that got him moved off the hot corner permanently.
- Pedro Alvarez is the most extreme case. He committed 16 in 2012, his first full season at third base, followed by 12 the next year and the unbelievable 24 from this most recent season that got him moved off third and onto first base.
- Ryan Braun is the other member of the 16-and-up club, a feat he committed in his rookie year. He was promptly moved to left field and hasn’t played the infield since then.
- Ryan Zimmerman is another guy who finds himself in the top 10 more than once, and another guy who didn’t last at third. His throwing issues have lasted his entire career, beginning with a 15-throwing error season back in 2007. He’ll be primarily a first baseman this year and moving forward.
Donaldon’s company is concerning. History has shown that racking up 17 throwing errors in a season isn’t just a blip. It’s an identifiable problem. And every third baseman who has had throwing issues as prevalent as Donaldson’s hasn’t lasted much longer at the position.
To be fair, there is a key difference between Donaldson and those guys, and that’s that Donaldson is otherwise an excellent defender. Of the 10 third baseman who have committed more than 15 throwing errors in a single season since 2002, Donaldson is one of just three to have a positive DRS or UZR, and he’s head and shoulders above the rest. Donaldson’s range alone will allow him to remain a plus defender, so long as his throwing problems don’t reach Pedro Alvarez heights. At the same time, his throwing problems aren’t something we can just ignore, and they seem like something that might not be accurately captured by advanced metrics. I think Zimmerman is perhaps Donaldson’s best comp, as Zimmerman had pretty good range in his prime years. But Donaldson is already pushing 30, and if age or injuries impact his range like they did Zimmerman, it’s not unreasonable to conclude that he could face a similar fate and be moved off the hot corner before too long.
But here’s the thing about Josh Donaldson: even without the defense, he’s still a great player. Even without the defense, he’s probably still the best third baseman in the league. Even if you strip away all of the defensive value that advanced metrics have given him the past two seasons, making him just a league average defender, he’s still tied with Beltre atop the WAR leaderboard at third base. Josh Donaldson doesn’t need his defense to be an great player. He’d have been a +5 WAR player the last two years without it. But the defense is what puts him in an elite class above his peers, and the defense comes with a blemish.
August used to cover the Indians for MLB and ohio.com, but now he's here and thinks writing these in the third person is weird. So you can reach me on Twitter @AugustFG_ or e-mail at august.fagerstrom@fangraphs.com.
Alvarez has pretty good range and hands at 3B, too. It’s the throwing that’s a problem.
Alvarez really doesn’t have good range, -3.3 RngR career. He does have better range than you might expect from someone with his body type but he’s still below average.
Might depend on the stat you use; at least last year, DRS and Total Zone ranked Alvarez’s range much higher than UZR did (and Fielding Runs Above Average, insanely, ranked him positive). The stats may have some trouble handling the Pirates’ shifts.
Still don’t see why we should be concerned about Donaldson’s defense. Yes, he has a problem with errors. Yes, most of his errors are throwing errors. Yes, throwing errors are less subjective and more predictive than fielding errors.
If he makes 20 errors a season, but ranges to make 40 plays that your average 3B wouldn’t make… What’s the problem? An out is an out.
Is there any evidence that for a player like Donaldson (great range, bad throwing accuracy), his skillset will decline sooner than other 3Bmen?
I don’t necessarily think we should be concerned this year. But defense peaks early and Donaldson is nearly 30. What happens when he loses a couple steps? What happens if he suffers an injury that limits his range? That’s why I invoked Ryan Zimmerman’s name.
This post isn’t meant to tarnish Donaldson’s defensive ability or value in the year 2015, I just thought it was something worth mentioning. He’s rightfully regarded as an elite defender, but he has a pretty serious problem that other elite defenders don’t have.
I’m not disagreeing with you, I’m just not seeing the connection you seem to make between his unique skillset and the predicted decline. Why do you think that Donaldson’s defensive ability is going to decline sooner or more precipitously than an average skill set? Do infielders who are dependent on range (rather than their arms) decline earlier/steeper on average? It’s not like he’s an outfielder; range for a 3Bman is about quickness and reflexes, not running speed.
I don’t think I implied Donaldson was at risk to decline any sooner than other third basemen his age. At least I didn’t attempt to imply that.
Quoted from the article: “But Donaldson is already pushing 30, and if age or injuries impact his range like they did Zimmerman, it’s not unreasonable to conclude that he could face a similar fate and be moved off the hot corner before too long.”
That’s probably the part that Yirmiyahu took to be implying a different aging curve, and I read it that way as well. Otherwise the same thing could be said about all players around 30, regardless of their defensive profile.
Isn’t August just saying that for similar defenders Donaldson is a bit unique in that his range is much better than his defense stats and his throwing is much worse – with the two offsetting for very good defense.
He then extends that logic to say that one of these counterweights is static (bad throwing) and one tends to inflect meaningfully. So if you have a large positive that is inflecting meaningfully offset by a smaller negative, pretty soon you go from a net positive to a net negative…. and that is both different AND faster than someone who might have “average” skills in both areas.
“Donaldson led all third baseman in errors this season”… Alvarez had more yet played in 50 less games at 3B
Yeah, Alvarez didn’t up in the leaderboards because he wasn’t qualified at 3B. Oversight on my part, thanks.
WORST TRADE EVAR!
You do realize that Lawrie is an elite glove with the potential to equal Donaldson with his bat right? I was actually pissed the Jays got rid of him.
Lawrie may one day be Donaldson, but Donaldson is already there. Also the Jays have JD for one extra year of control. This is a huge win and upgrade over Lawrie, and I’m speaking as a Jays fan who loves that Redbull chugging monster.
If that “one day” is 2015, and if Graveman continues being unhittable, and if Donaldson suffers the same fate as just about all Jays players and becomes another turf-DL victim then this trade becomes awful for Jays and yet again, Beane looks like a genius.
So you are saying that if Lawrie gets to 7.5 WAR, if Graveman becomes a dominant MLB pitcher and if Donaldson loses 4 or more WAR to injuries that this will be a good trade for the A’s?
Lawrie would have to stay on the field long enough for that to happen.
Lawrie has missed 184 of a possible 486 games over the past 3 years, a 38% clip. Even if he got injured at a lesser rate, playing in 130 games would be a massive improvement. Such a leap seems unlikely given a recurring oblique injury in 3 different seasons, which is particularly troubling given his all-out style of play.
Donaldson played in 158 games each of the past two seasons.
Much of Lawrie’s Donaldsonesque potential emanates from his remarkable 2011 season, when he took a big step forward in fulfilling his toolsy tools everyone spoke of since he was drafted. However, his recent performance from 2012, 2014, and 2014 indicates it is not likely he will become Donaldson. The recurring oblique issue may have altered his swing just enough to result in the poor wRC+ of the past 2yrs. This, despite playing in a park that is friendly to hitters like him. Had he no injuries, Lawrie actually would have projected Paul Molitor-like performance, with more power and less steals (and be better than Donaldson) Unfortunately we have not seen those abilities materialize.
The only way this works for the A’s is if Barreto turns out to be really awesome in 2017 or 2018. I don’t believe Larie will get a lot better than he is now, because of his injuries. Donaldson is a huge upgrade over Lawrie now, and will be over the next 3 seasons as well.
What if Donaldson gets injured after playing on turf for over half his season (lots of games in TB too), and Lawrie reacts to playing on grass positively?
All Lawrie has to do really, is hit 20 HR and the A’s win this trade. That’s not even taking into account how everyone seems to be totally ignoring the season Graveman had, which – wow. We really should be looking at how awesome his 4 seamer is. Why aren’t we?
So if Lawrie has a JP Arencibia batting season Oakland wins?
It should also be noted that if the 4 traded Jays players combine for 24 WAR per season in Oakland and then sign below-market contracts PLUS Donaldson turns out to be that Reggie Jackson gun-toting robot from Naked Gun that this trade will be good for the A’s
Why is everyone writing comments that are a) low probability and b) obvious in their outcome and then BOLDLY stating that obvious outcome?!?!
Pedro Alvarez actually had the most errors by a third baseman this year (as you effectively point out, since Donaldson had 23 errors total and Alvarez had 24 throwing errors).
Also, when you say that only one other player had more than eight throwing errors I think you mean only one other third baseman, but that isn’t true; Alvarez, Lonnie Chisenhall, and Anthony Rendon all had more than eight (as did a bunch of shortstops and catchers). I think you had the minimum innings filter set to 1000.
At least half this comment was already preempted by Ctown!
Yep. Damn qualified leaderboards filter got me. All of this should be fixed now.
Edwin was so bad at 3b they nicknamed him E5.
And I still (and always will) refer to him as E5.
Is E5 better than Elmon Young? (There is no “D” in Delmon Young, hence, Elmon).
this. awesome nickname. glad that edwin and the jays can laugh about it now.
note: the A’s had E5 for a short while before he discovered how to really hit.
The A’s could have had E5 and Moss in the same lineup. Still shuddering at the 2014 DH production that E5 could have given them, instead of playing Callaspo there *ugh*.
One bounce throw off the turf in rogers centre and he’ll hit first every time. I watched an interview with ozzie smith, and he said he’d take advantage of turf on long throws by making his throws bounce once. The bounce was always true because the ground was rubber.
But at the same time, he only committed 16 errors in 2013 (in more innings) which is still high, but not astronomically so.
“I think Zimmerman is perhaps Donaldson’s best comp, as Zimmerman had pretty good range in his prime years.”
Ryan Zimmerman:
2007: DRS 18; RngR 15.5; UZR 16.0; UZR/150 15.5
2009: DRS 22; RngR 11.5; UZR 11.9; UZR/150 12.2
2010: DRS 11; RngR 15.1; UZR 17.9; UZR/150 17.9
Josh Donaldson:
2013: DRS 11; RngR 10.7; UZR 9.9; UZR/150 11.1
2014: DRS 20; RngR 14.5; UZR 15.5; UZR/150 13.3
In 2012 Donaldson posted a UZR/150 of 10.8 while playing about half a season at third base.
I agree that Zimmerman and Donaldson are good comps. But saying that Zimmerman had “pretty good range” is either a gross understatement or casts doubt on the validity of the same fielding metrics that give great value to Donaldson. Because Zimmerman from 2007-2010 was actually better than Donaldson has been the past three years.
I think you’re nitpicking the difference between “pretty good” and “pretty great” here. They’ve both got good – as opposed to bad – range, is the point.
Mind is blown Zimmerman is only a year older than Donaldson.
I wonder if he might have been saved some of those if Oakland hadn’t been playing sub-par defenders at 1B and SS. 2 of those examples of errors above were at least playable
Great point here. A lot of bad throws are hidden by better than average first basemen. Are there any defensive adjustments (up or down) on defense that’s directly affected by other defenders
No, it is not a great point. Only Josh Donaldson’s mother or his attorney would make such a fatuous argument and attribute fault to someone other than Josh Donaldson for those horrible throws.
I don’t think they’re arguing that Donaldson isn’t responsible for the bad throws; they’re saying some of them don’t show up on the record as errors if a superior 1B is fielding them.
Interesting point. Unfortunately for Donaldson and the Jays, they don’t have a great fielding SS or 1B.
Smoak may not have high ratings defensively but he is excellent at picking bad throws.
He has to play in order to take throws.
The defense was the reason I was shocked at this trade. The loss of Lawrie on the field is something I didn’t think I’d have to deal with, other than the usual injury issues that is.
Time will tell, but AA has to have something else coming if Lawrie cannot be moved over to 2B anymore. I’m not sure Devon Travis is that near term solution for this season.
The A’s have a potential all-star on their hands, I hope for their sake, and Lawrie’s, that his potential will finally become reality.
What if the A’s don’t think that Donaldson is an elite defender? If you look at their team stats since 2012 (when Donaldson starting playing 3B) he hasn’t been the best 3B defender on the team. Last year and in 2013 Alberto Callaspo was better (SSS warning) and in 2012 Brandon Inge was better. So maybe the A’s think that it’s just their positioning that makes him a defensive star?
However even with a move to 1B he’d still be an elite bat – over the past two years his wRC+ ranks 6th at 1B compared to 2nd at 3B.
6th at any position is /elite/?
If it’s 6th after Cabrera, E5, Votto, Goldschidmt, Freeman, that sounds elite enough to me.
“[Throwing problems] seem like something that might not be accurately captured by advanced metrics.”
Why is this? It seems to me like throwing errors would be one of the easiest things for advanced metrics to accurately measure.
This was sort of me thinking out loud. I could be wrong, but advanced metrics, to my knowledge, don’t weight errors very heavily because of their subjectivity, lack of year-to-year correlation, etc. But Donaldson’s repeated throwing problems appear to be a different story, so perhaps their negative value aren’t accurately reflected by the metrics?
Again, someone who knows more about these calculations can feel free to correct me.
Don’t advanced metrics essentially measure plays made against plays that should have been made? It seems to me, and maybe I’m wrong, that throwing errors are going to be relatively easy to determine whether or not the play should have been made. Moreover, it seems relatively easy to determine the run values of those errors as well. My guess is that the current metrics are able to capture these errors pretty well.
I seem to recall Lawrie had a wild arm when he first started playing 3rd as well. Lawrie has a 9.56 TE per 162*9 innings while Donaldson is at 14.46. They seem to be trending in different directions though. If you scratch out 2014 we get 10.587 for Lawrie and 11.625 for Donaldson. What are the chances this is a one year blip for both of them?
Aren’t throwing errors subjective and circumstantial as well? Donaldson is going to have a lot more throwing errors if he has Mo Vaughn at first base than if he has Keith Hernandez because of the latter’s ability to pick balls out of the dirt.
To an extent, yes. What I meant by that, though, is there’s a big gray area when a player fails to get to a batted ball on whether or not it’s an error. When a third baseman doesn’t get his throw to the first baseman, there’s no wiggle room – it’s gonna be an error.
But yes, the first baseman matters and not all throwing errors are 100% on the third baseman.
I do not have the time to go through all the Donaldson plays from last year but I would guess that 3b who have great range at third may have many more throwing errors because of their ability to get to balls that need rushed throws, such as was exhibited in the first gif when Donaldson threw out the base runner from the seat of his pants.
This is me just rambling, but out of curiosity, how would Donaldson play at 2nd? His range seems good enough for it. And while his issue is more arm accuracy than arm strength, I’d still imagine that being closer to first base, and having more time to make the throw would help. The Blue Jays do stink as 2B too…
He doesn’t look like a 2B, but it’s really not about big vs. small when a SS (which JD isn’t) is moving to 2B or 3B, but rather some skills that are mildly correlated with those size. It’s about which is their weakness: range or throwing.
Thing about that is Jose Reyes is probably going to be the one moving to 2B if there have to be any positional shifts due to flagging defensive abilities. So if Donaldson does move to 1st you could have a mirror image of the 2015 infield in… 2018? 2019? Depends on Reyes I guess. And if this move even happens at all.
A useful stat, if it exists, would be throwing accuracy whether it’s an error or not. You can’t just look at one error an say the 1st baseman should/could have handled it because there may also be some that the 1st baseman shouldn’t have handled but somehow came up with the ball. It works both ways.
It doesn’t need to be taken to the level of Reynolds’ digestive distress, but it shouldn’t be ignored either.
There’s a “scoops” stat that tracks how many throws a 1st baseman scoops up in a year. I suppose if you could split these scoops to the infielder making the throw you could then look at (throwing errors + scoops)/(throws to 1st) as a rough measure of throwing accuracy.
http://www.fangraphs.com/leaders.aspx?pos=all&stats=fld&lg=all&qual=0&type=0&season=2014&month=0&season1=2014&ind=0&team=0,ts&rost=0&age=0&filter=&players=0&sort=14,d
Throwing errors might also depend on range. Higher range equals more opportunities of throwing from awkward positions and longer throws. It was an interesting finding but the study would be more complete with an analysis (if it is possible) of the correlation between higher-range third basemen and throwing error rates.
For a fielder to be in “awkward positions” that would likely mean they are at the edge of their range. Doesn’t every fielder have an edge of their range? Wouldn’t they too end up in the same “awkward positions”? The bottom line is Segar and JD had almost the same amount of assist opportunities and Segar made 1 throwing error to JD’s 17. That doesn’t pass the smell test.
Every fielder has his range but a “range” is mostly circular (I guess). Larger range=longer edge of range. I’m thinking along the lines of ernesto: more range/better hands is more throws is more errors.
Even if I give you that the edge would be larger… 17X the number of throwing errors of Seager?
Assuming that an assist is a ground ball opportunity to throw out a runner…
In 2014 JD led the majors with 328 Assists. The next 4 Seager 327, Carpenter 288, Sandoval 282 and Arenado 281 had a combined 1,178 Assists. All four combined had 17 throwing errors.
Chase Headly’s RZR is the 2nd highest behind JD (.770 to .769). In 234 Assists, Headly only had 3 throwing errors.
17 is a lot no matter how you slice it.
Makes u think, if someone got a super tall or long-armed 1B with good hands that might be a great pairing with a big-but-wild-armed 3B or SS.
“History has shown that racking up 17 throwing errors in a season isn’t just a blip. It’s an identifiable problem. And every third baseman who has had throwing issues as prevalent as Donaldson’s hasn’t lasted much longer at the position.”
“…and they [throwing errors] seem like something that might not be accurately captured by advanced metrics.”
This casts a bit of doubt over the reliability and ( and usefullness) of advanced fielding metrics.
When I saw the 2nd GIF, I thought that this was going to be an article about how playing in Oakland with all that foul territory inflates Donaldson’s UZR. Do we have any idea whether this is the case? Obviously that play in the 2nd GIF is outstanding, but Donaldson would not have had the opportunity to make that play in most MLB parks.
UZR is park-adjusted. To what extent, though, I do not know.
Looking at the second GIF, that’s a routine foul ball in any other stadium. I know defensive stats are supposed to account for ballpark, but that’s a play he won’t have a chance to make in Toronto. Seems to be like leaving Oakland will hurt his value simply because he won’t have the chance to show off his range as much.
Somewhat new to trying to understand and figure out advanced defensive metrics. In GIF 1, while he did make an outstanding play, if you look the shortstop was in excellent position and would have probably made the play. Not sure how that is addressed. In GIF 2, again great play but more park driven perhaps than player driven. Other 3b at other parks never get the chance to make that play. Not sure how park-adjusted deals with this. Lot of fielding stats remind me of RBI’s in that they are as much a function of opportunity as they are in individual skill. Not everyone gets their turn to make these plays.
here we go, as soon as a player is traded from Oakland, fangraphs starts trashing him.
here we go, as soon as something isn’t gushingly good news about Toronto, everdiso starts whining.
Perhaps you should have someone read and explain the article to you next time. If you don’t understand something, feel free to ask an adult for help.
Break out the diaper and the binky.
Many of these throwing errors can be attributed to 1. Jed Lowrie being to lazy to reach for a ball. 2. Callaspo/Sogard/Punto not being able to catch a fastball 3. Whoever at first not being able to pick a ball in the dirt like an average defensive first baseman.
Clip 1. Legitimate mistake trying to hurry up a throw to get Altuve.
Clip 2. Looks like Punto. Nuf said
Clip 3. Vogt at first is a result of needing to put a hot bat in the lineup. ANY major league first baseman should be able to pick that short hop.
Donaldson is an average catcher turned into an ELITE third baseman. Most of his mistakes are him trying to do too much and I would not hesitate to run him out every game.
Also, he was battling injury toward the end of the year and I suspect he injured his knee during tarp catch 2. Dude is a beast and will become the most valuable all around player Canada has ever had. He plays 100% at all times and every at bat.
He WAS a beast…he’s now on turf. Lets check out how that worked for Lawrie, Bautista, Lind etc etc every single Blue Jays player ends up on the DL due to turf – and also playing the Rays in TB a lot too.
Really unfortunate that every team that plays on grass gets the advantage of never having players go on the DL.
“He plays 100% at all times and every at bat.” – again, exactly like Lawrie.
I wonder if Jays fans would rather have Dickey, Reyes and Donaldson, or all the players they traded for them?
I thought I had a going problem, but my doctor said I had a throwing problem.
I did a little digging on this myself because I was astounded by JD’s defensive numbers but intrigued by the number of errors. He led 3Bs in errors and assists which made me think about this as a % as a more important metric. I believe the stat would be more worthwhile if i could find the number of throws he made but i only have assists so thats what i used (if someone knows how to find that stat i would be grateful)
what i have is the number of TE by the number of Assists made
2014
Lonnie Chisenhall= 0.054945055
Josh Donaldson = 0.051829268
Anthony Rendon = 0.046808511
2013
Ryan Zimmerman = 0.057692308
Pablo Sandoval = 0.053398058
Conor Gillaspie = 0.0450
Josh Donaldson = 0.039215686
2012
Pedro Alvarez = .060606061
Ryan Zimmerman = 0.042253521
Miguel Cabrera = 0.032921811
I didnt get much further on this because i waste enough time at work as it is. this is my first time posting so i would be more then happy to have any pointers or feedback.
Thanks for posting this, I was thinking along the same lines. I.e., if his range is much greater than that of most other 3B men, so he gets to balls that otherwise would go for hits, it stands to reason he makes more throws as well. So even if he has average throwing accuracy, one would expect more errors.
An additional possible factor is that when he ranges far to reach a ball, he may be in an awkward position to make a throw, hence more likely a bad throw. That first gif shows such an instance when the throw was on the money, but how often does he throw the ball away when he’s in that position? Maybe sometimes the scorekeeper recognizes the difficulty and awards the batter a hit, but it seems to me sometimes he would likely be charged with an error.