Front Page: What Happened to Eddie Rosario’s Defense?

[Andrew Thares]

With all of the questions the Minnesota Twins have had surrounding most of their team over the past few years, one area of the team that has been steady has been the outfield. Sure, Byron Buxton has had his fair share of issues with the bat and staying healthy, but there has never been a question about him being the Twins center fielder, when healthy, thanks to his seemingly superhuman abilities to chase down flyballs. Additionally, the play of both Eddie Rosario and Max Kepler in the corners has given the Twins great stability. One of the hallmarks of this trio has been the great defense that they play collectively. However, while Buxton and Kepler have been able to maintain their high levels of defensive play, Eddie Rosario has taken a big step backward this year.I will start this article by saying, I am only referencing fielding metrics generated by Statcast that are available on Baseball Savant. While many of the defensive metrics on sites like Fangraphs and Baseball Reference could be helpful, they are, at least for outfielders, outdated, and nowhere near as accurate for measuring an outfielder’s ability to track down flyballs as Statcast is. The reason for this is that Statcast actually tracks player and ball movements to quantify the time and distance that an outfielder has to track down a flyball, and calculates catch expectancy from that. Other metrics, like UZR and DRS, rely on the eye test from humans, which is a much less precise process. If you would like to read more about how the Statcast fielding metrics are calculated, you can do so here.

 

That being said, let’s start diving into the numbers. We will start by looking at Eddie Rosario’s expected catch percentage vs. his actual catch percentage. So far this year Rosario has an expected catch percentage of 86 percent. That is to say, an average outfielder would catch 86 percent of the fly balls hit to Eddie Rosario this year. However, Rosario’s actual catch percentage stands at just 81 percent, giving Rosario a catch percent added of -5 percent. While this might not sound like a lot, it actually is. Among the 97 outfielders who have fielded at least 50 fly balls this year, only four outfielders have an expected catch percent added worse than Rosario's. This is a severe drop off from the +2 percent, -1 percent and -1 percent, that Rosario averaged in 2016, 2017 and 2018 respectively.

 

Using these metrics, we can calculate Eddie Rosario’s outs above average. How this metric works is, if you have a fly ball hit to you that has a 75 percent catch probability, and you catch it, you get credit for 0.25 outs above average on that opportunity. However, if you fail to catch it, you get credit for -0.75 outs above average on that opportunity. In 2019, Rosario’s outs above average currently stands at -9 outs, which is on pace for 13.5 outs below average. For comparison, Byron Buxton currently has an outs above average of 12, which means Rosario has negated 75 percent of the value Buxton brings in the field defensively with his play in the field.

 

So, what has caused Eddie Rosario to experience such a drastic drop off, from roughly league average outfielder, to one of the worst defensive outfielders in baseball? For that, we will need to dive even deeper. We will start by looking at Rosario’s sprint speed, which is another metric tracked by Statcast. This year, Rosario’s average sprint speed is 27.2 feet per second, which is roughly league average. However, from 2016 through 2018, Rosario averaged 28.0 feet per second. Again, for comparison, Byron Buxton’s sprint speed this year is 30.3 feet per second, so a 0.8 feet per second drop is a significant decline.

 

However, there is more to being a good outfielder than just being able to run fast, though that does help a lot. Getting a good jump on the ball, getting up to speed quickly, and taking a good route to the ball are all important factors in being good at tracking down fly balls. Luckily, thanks to the player movement tracking technology, via Statcast, we can measure each of those factors, using some newly introduced metrics referred to as Reaction, Burst, and Route.

 

We will start by looking at reaction, which measure the amount of ground an outfielder covers in the first 1.5 seconds after the pitch is thrown. From 2016 through 2018, Eddie Rosario measured in at 0.3 feet above the MLB average in reaction. However, in 2019 Rosario is at -0.8 feet below average in that initial reaction stage. The next metric we will look at is burst. Burst measures the feet that an outfielder covers from the 1.6 second mark, to the 3 second mark of the play. In other words, how fast the outfielder is accelerating. Again, looking back from 2016 through 2018, Rosario averaged 0.0 feet above average in burst, or exactly league average. That number has fallen all the way down to 1.2 feet below average in 2019. The final metric that we will look at is route, which measures the outfielder’s route efficiency, and how many extra feet they are needing to run by taking a poor route. From 2016 through 2018, Rosario gained 0.2 feet, per opportunity, on the average MLB outfielder due to taking good routes. In 2019, Rosario is gaining 0.3 feet per route above the average MLB outfielder.

 

It is apparent that Rosario has taken a considerable step backward in not just sprint speed, but in his all-around athletic skills as well. So, the question is, how much has this hurt Rosario? The table below breaks down how many feet Rosario is losing, on a per fly ball basis, from where he was in the three previous seasons.

 

Download attachment: Eddie Rosario Fielding Metrics.PNG

 

The first thing to note from this table is the Feet Gained for the Feet Per Second category is multiplied by two, and there is a reason for this. For starters, the average fly ball is in the air for just shy of five seconds (4.97 seconds, to be exact), hence, the feet per second drop off is amplified the longer the ball is in the air. However, feet per second is measured while a player is at max speed, not while he is accelerating. Since it typically takes about three seconds between the release of the pitch, and the moment the fielder reaches max speed, the feet per second is not factored in until after that point. Therefore, the average fly ball hang time of five seconds, minus the three seconds it takes to reach max speed, gives us an average of two seconds at max speed per fly ball.

 

After taking that into consideration, we can calculate the rest of the factors, and see that Eddie Rosario’s ability to track down a fly ball has gone down by an average of four feet per fly ball. While on most fly balls, four feet won’t make much of a difference, over the course of a season, the number of fly balls where the extra four feet would make the difference starts to add up.

 

Knowing this, the Twins need to ask themselves, is this something they can work with Eddie Rosario on? Has he simply lost a step athletically that he might not recover, or is Rosario simply having an off season in the field? Whatever the answer is, it is important for the Twins to work with Rosario, so they can either help him improve, or start to factor in this decline for their future evaluations of Rosario as an all-around player.

 

Click here to view the article

[rgarfinkle]

Thanks for this deep-dive into the numbers.  I don't pretend to understand all of that, but I will say it confirms what my eyes tell me.  I wonder how much of this drop-off is post-ankle sprain vs. pre-ankle sprain, although his early-season defense was also, by my eye, not as robust as previous years.  He's also so casual about one-handed catches, side-saddle catches, etc.; I wish he would play it straighter and safer more of the time.  He has to fix this; giving away 3/4 of Buxton's benefit is not a winning formula.

[Doomtints]

The concern is the number of throwing errors in this Twins infield, not Rosario. It's like watching the four stooges out there.

[ashbury]

The concern is the number of throwing errors in this Twins infield, not Rosario. It's like watching the four stooges out there.

Some days: "Errors are a worthless statistic. The Official Scorer never calls them."

Other days: "Oh noes! We are tied for 10th in the majors for most errors."

 

I'm far from considering infield defense a strength for the team, but the number of throwing errors isn't enough to warrant name-calling.

[Doomtints]

Take a gander at the number of errors by the Twins outfield starters v. Twins infield starters. This is a major problem.

[stringer bell]

Since none of the Twins infielders is noted for their range, I think a high number of errors would be a red flag.

[AceWrigley]

Nice analysis. Just my observation, but Eddie seems slower off the crack of the bat, and his strides seem short. On replays of balls hit to his defensive area he doesn't seem to have any closing speed to reach balls adequately. Might be due to injury or something else but he's not playing the greatest defense this year.

 

And . . . I'm ticked off at Buxton for again running into the wall for no good reason. Dude, you gotta stay off the IL man. Play a little deeper so you can get to the wall before the ball does or something. You've got the speed and ability to race in and grab the flares and short liners. I'd rather a few balls drop in for a hit here and there if it means you staying in the lineup. Rant over.

[ashbury]

Perhaps Andrew will write a follow-up article about defensive metrics for the infield.

[Andrew Thares]

Take a gander at the number of errors by the Twins outfield starters v. Twins infield starters. This is a major problem.

Every team has a lot more infield errors than outfield errors, that’s just how things work. Also just because the infield defense is a problem, doesn’t mean Eddie’s defense isn’t also a problem. If the bullpen is struggling, but so is the starting rotation, you wouldn’t say the bullpen isn’t an issue just because the rotation is a bigger problem.

[hybridbear]

Great article! I’d like to see the same analysis for Kepler & Buxton, and maybe also Gonzalez since he’s also played outfield a fair amount with Buxton hurt.

[Shaitan]

 

Nice analysis. Just my observation, but Eddie seems slower off the crack of the bat, and his strides seem short. On replays of balls hit to his defensive area he doesn't seem to have any closing speed to reach balls adequately. Might be due to injury or something else but he's not playing the greatest defense this year.

 

And . . . I'm ticked off at Buxton for again running into the wall for no good reason. Dude, you gotta stay off the IL man. Play a little deeper so you can get to the wall before the ball does or something. You've got the speed and ability to race in and grab the flares and short liners. I'd rather a few balls drop in for a hit here and there if it means you staying in the lineup. Rant over.

I love that wall collisions are part of his game. But there's a way to do it. Watching the monthly highlights on ESPN today, I saw Hamilton plant his feet and climb. Buxton just runs straight into it. It's reckless. He makes great plays and it's a game changer, but he need to find a sustainable way to do it. Hunter, for example, ran into walls too. But he only really hit the IL for it once and that was partly because Fenway is...old.

 

[yarnivek1972]

Sabremetric gurus that Falvey and Levine are, I am sure they are aware of it. I could see the Twins trying to move Rosario for pitching in the offseason. I don’t think it is at all likely he receives an extension offer he is willing to sign. He’s a good, not great, player. He’s the kind of guy you play until he gets too expensive. Then you replace him with someone cheaper. Mr. Kirilloff for example.

[mlhouse]

The real problem with these types of measures is that their measurement variances are huge.  Did Rosario go from being a decent outfielder to a bad defender in the course of a season?   Or are such measures going to swing back and forth for many players?

[Richard Swerdlick]

It seems obvious to me that in watching Eddie run the bases he is not fully healed from his ankle injury. That must be affecting his play in the outfield as well.

[JoshDungan1]

I honestly think that if the Twins could actually string a couple healthy weeks together for Kepler and Buxton while also having Cave do literally anything other than just look overmatched at the plate, they might have sent Rosario to the DL to try and fully get over whatever ankle issues he is currently dealing with after that sprain he suffered earlier in the year.

 

Because yeah, he just isn't hustling very much right now, I see him grimacing in what appears to be pain a fair amount both out in the field and on the basepaths, and his home-run hitting ability hasn't kept up from what it was earlier in the season. I just want a healthy Eddie to go out, enjoy the game like he has even when the Twins were s***ty enough to be losing 100 games in a season, and be a fully productive #bomba masher again.

[Thrylos]

Based on that data looks like Rosario is slower to react to fly balls than last season.  Would love to see how that changes with Buxton at CF vs Kepler at CF.   My eyeballing says that when Buxton is playing at CF the other OFs wait to see if he has a jump on it because he is all over the place.

 

Another change I saw at UZR is his "arm" rating, which makes sense because runners have stopped challenging him.

 

So I think that this is as much Buxton as the others at the OF...

[highlander]

I think your stats are pointing to a bum ankle. If route to the ball was off that would be bad. With no solid push off reaction and burst are down. Your stats and my eyes agree. Your stats are approved.

[Doomtints]

 

Every team has a lot more infield errors than outfield errors, that’s just how things work. .

 

Take a look anyway. You will be surprised.

[curt1965]

Great article! I’ve noticed this problem much earlier than Rosario’s ankle problems. His method of “hot dogging” every catch has bugged me for a long time. It has cost the Twins at least one game. I also noticed today that he certainly didn’t hustle out of the box on his “single” on the Twins last RBI.

My grandson would be pissed with my comments-Rosario is his favorite player. And if he is playing hurt-that’s one thing. But if he’s just not playing hard, that’s another. I would hope someone in the clubhouse would address that. I’m not sure that Rocco is the type of manager to hold him accountable.

[Otwins]

Not to pile on Eddie because he is one of my favorite players but not only has it affected his fielding and base running but also his hitting.  He is not hitting for as much power as he was.  I thought he made it 5-0 when he pulled that ball today but it hit the limestone and even after Soler fumbled the ball Eddie was not at second base. He has hit three balls off the wall this week that I think would have been home runs earlier in the year. I am wondering like LA Vikes Fan also. Ankle is worse than a sprain.

[Old Twins Cap]

Yeah, maybe the "Eddie" chants have gone to his head.

 

Players gotta realize: their moment in the sun is brief, and they need to make plays to stay relevant.

 

Love Eddie to death, but, to me, as a fan, I seem someone who is not giving it his all.

 

And in baseball at this level, that matters.

 

A lot.

[nokomismod]

To the poster that suggested that the Twins trade Eddie, I think that is a reasonable option considering the options waiting in AA. But he is a fan favorite and a lot folks including people look at him as a confident guy (sometimes too confident when catching fly balls) who gives the team a little moxie. I think the outfield when healthy is on the plus side for defense and finally this year the offense has come to life.

[ashbury]

My working assumption is that Eddie is a gamer and is working through a difficult nagging injury. I'm not personally in favor of inferring anything about his mindset.

[theBOMisthebomb]

Injury not yet healed as many have mentioned. Fewer have mentioned effort and attitude. I love Eddie and his confidence bordering on arrogance. However, I do not approve of the unsound fundamentals in the outfield. Maybe it's because I was in the left field bleachers for the drop against the Mets and booed Eddie that day.

[stringer bell]

What did the data say prior to the injury? I’m willing to bet that his defense was rated subpar pre-All Star game, which would indicate that it is just not the effects of the ankle sprain.

[jokin]

 

What did the data say prior to the injury? I’m willing to bet that his defense was rated subpar pre-All Star game, which would indicate that it is just not the effects of the ankle sprain.

 

Never has been a great defender. But I'm willing to bet via the eye test and the stats, particularly his OF range and his power production, will suggest that he's just not the same player since before the ankle injury.

Edited August 5, 2019 by jokin

[Tomj14]

 

My working assumption is that Eddie is a gamer and is working through a difficult nagging injury. I'm not personally in favor of inferring anything about his mindset.

agreed, he seems to have no top speed right now (due to injury) and/or is afraid to go top speed and try to stop on the ankle.

Having sprained my ankle multiple times getting up to speed usually isn't the problem, the problem is usually stopping or slowing down. From my non-medical eye he seems to be looping to the ball instead of taking the correct angle to cut it off and IMO that is sign of a injury not healed not a lazy or mental issue.

[Andrew Thares]

What did the data say prior to the injury? I’m willing to bet that his defense was rated subpar pre-All Star game, which would indicate that it is just not the effects of the ankle sprain.

This has been an ongoing issue with Rosario all season. I first noticed back in May that Rosario’s defense was down, but wanted to let the season played out more before I made anything of it. Sure the ankle injury has probably amplified the effect, but it was already an issue previously.

[Trov]

I think this is in party why he has not been extended.  His defense has been trending downward, the eyes can tell that, and his bat has been good overall, but not game changing to level you overlook poor defense.  I like the improvement he has made at the plate, and he loves the big moment to do big things, but he is starting to go by way of man ram, not that bad yet, but he clearly is not taking it as serious as his hitting. 

[twins1095]

If there is a guy the Twins should explore trading for pitching it should be Rosario (rather than Kepler or Sano as have been suggested over the years).

 

Rosario is just as streaky as Sano, gets on base less than Kepler/Sano (by getting himself out expanding the zone), hits for less power, is a bigger minus defensively than Sano, has the worst plate discipline, and is the oldest of the three.

 

Good, exciting player who's got good power forsure--but if the Twins were to use one of their guys as a trade piece (especially with Larnach/Rooker/Kirilloff relatively close to the majors)...it should be Rosario.