Dodgers Defense Is Better, But How Much Better?

Hanley to Rollins defensively is significant Bill Streicher-USA TODAY Sports

If we’re going to be objective about the Dodgers situation regarding all of the trades and signings within the past week (something I’ve struggled to be at times), there is no question defense was at or near the top of the list of motivations behind these moves.

Andrew Friedman is well known for fetishizing defense, signing James Loney, trading for Ben Zobrist, starting Desmond Jennings, trading for Wil Myers, extending and developing Evan Longoria, trading for Yunel Escobar -back when he was good at defense-. Friedman also was the first major league GM to go “all in” on the idea of pitch framing, signing 87 year old Jose Molina to a 2 year contract because he did 1 thing well, frame.

Knowing that Friedman puts a lot of importance on these skills, Matt Kemp‘s trade to the Padres for Yasmani Grandal and Dee Gordon‘s trade to the Marlins (for what became useful pieces plus Howie Kendrick) start to make a whole lot more sense, you or I may not like them, but they make sense. Every position that has seen some change (C, 2B, SS, CF/RF), the defense is inarguably better than it was last season.

-the only position where I’d say it’s not “slam dunk” better is 2b with Dee Gordon being young athletic, and rangy while Howie Kendrick is older and widely regarded as just “solid”-

But just how much better is the Dodgers defense this year?

Lets go position by position.

Catcher

A.J. Ellis was, well… bad last season. Just looking at him play defense was painful, the passed balls seemingly twice a week, the difficulty throwing out runners, the increasingly bad receiving issues. The numbers back up all of these perceptions, as he gave up 34 passed balls/wild pitches, only threw out 16 out of 48 base stealers, and being the 2nd worst pitch framer in baseball *take that how you will*. I know, I know that A.J. had serious knee issues, and as a catcher, knee issues are really bad. And granted, Yasmani Grandal (who also tore up his knee as recently as 2013) isn’t the paradigm of defensive catchers, in fact he was quite bad in the passed ball/throwing out runners department, giving up an extra base to baserunners 40 times on passed balls/wild pitches and only throwing out 7 out of 49 runners trying to steal. But remember, the same guy who gave Jose Molina 4.5 million to go out there and turn balls into strikes is running the Dodgers now, and Grandal is among the lead leaders in the framing department, ranking 13th in baseball when it comes to total runs saved from framing.

I’m not advocating framing as presented for the record, there, has to be more research placed into this statistic, because Baseball Prospectus claims that the Dodgers went from -14.2 runs below average to 12.6 runs above average, an addition of 26.8 runs because of 1 skill (!!!) and that idea is an absurd one. But some things about Grandal vs Ellis are true. Both are horrendous at passed balls/wild pitches. Both have had surgeries on their knees in the past 2 season. But only one has entered his prime. When choosing between a soon to be 34 year old catcher coming off of a knee injury in the prior season, and a just turned 26 year old catcher who is further removed from his knee surgery, you take the latter 10 times out of 10. What’s also true about Grandal and Ellis is the fact that a catcher throwing out a baserunner is somewhat dependent on the pitcher he is throwing to, how quick that pitcher is to the plate, how fast the baserunner is, and what pitch is thrown by the pitcher, etc. It isn’t just a matter of “the guy must have a weak arm” because that’s not what a major league scout told Pedro Moura two days ago:

"The knee is the major question,” the scout said. “The surgery is affecting him on defense. I saw him throw pretty good, but I do think his mobility behind the plate was hampered"

Lets assume that it’s possible for improvement defensively because he’s still so young. 26 year old Yasmani Grandal is probably a surefire upgrade over 34 year old A.J. Ellis, Grandal has a pretty strong arm, and now he gets to work with the pitching staff that has at least partially made Ellis into the prolific threat from throwing runners out. I’m not going to pretend that Grandal was that great at blocking balls in the dirt, but even if you don’t buy into framing as is, there isn’t much of a question that Yasmani is a better overall pitch receiver than A.J. Ellis.

At worst, i’ll peg this as a slight upgrade defensively, with a high likelihood of Grandal being a significant upgrade because of the fact that as a younger Grandal gets further away from his knee injury, the higher the chance Yasmani improves throwing runners out and blocking pitches in the dirt.

Second Base

This is the tough one. Mostly because with all of these numbers, we have to remember that next season is what we’re going to be seeing, not 2014, and while Dee Gordon had his rough moments defensively (seemingly everything went wrong during September), given a year to adjust to the position, and the range Gordon has, it’s possible that 2015 Dee Gordon becomes one of the best defensive second baseman. (Or not, and he takes a step backwards like he did when playing SS.

I don’t care about UZR’s lack of transparency very much, and we need something more than our eyes to evaluate defense . Inside edge does that pretty nicely. While it has it’s problems, like slightly penalizing guys with range to get to a ball that a baseball player ordinarily wouldn’t and classifying that “missed play” as a play with “even chance of being made” as opposed to a play with an “unlikely chance of being made”, it’s one of my preferred ways of looking at defense, along with my eyes.

So instead of adding more words than need to be added, lets just look at a straight comparison of inside edge fielding between Dee Gordon and Howie Kendrick:

-The numbers underneath what the play is classified as, means that the player made (X) percentage of the type of play. So Dee Gordon made 96.6% of the “routine” type plays this past season. The numbers in parenthesis are the total attempts at the play he had. So Howie Kendrick made 20% of the unlikely chances he had last season- (numbers courtesy of fangraphs dot com)

[table id=15 /]

What sticks out was wooooow, Dee made some amazing plays this past season, and we saw that, for sure this past season. When looking at the data, these two players appear to be polar opposite, at least defensively.

The latter 3 rows (aside from impossible) are all dominated by Gordon, who made 2 more remote plays than Kendrick in the same amount of chances (8.3%-0.0% advantage percentage wise) he was better in the unlikely stage (50.0%-20.0%), as well as the even area (84.6%-55.6%). All in all, when looking at the remote/unlikely/even phases, Gordon undeniably made more plays than Kendrick did, and it wasn’t because Dee had more opportunities. We also know that turning routine and likely plays into outs are probably more important, mainly because those comprise the bulk of chances a fielder has, and Kendrick owns a 1.5% advantage in converting a routine play into an out. That may seem insignificant, but if you give them the same amount of of chances (say 425 opportunities), then that’s approximately 6 plays over the course of a season. In the “likely” department, Kendrick made 4.5% more plays than Gordon did last season.

This data pretty much confirms my suspicion, I imagined that Howie Kendrick would have the surer glove when it came to routine/likely chances. That suspicion comes from the fact that Kendrick 1) has a good glove and that’s one of the reasons he’s stuck as a middle infielder for so long, and 2) he has been playing the second base position professionally since 2002, whereas Dee has focused solely on second base for about a year now.

On the other hand, it’s not as lopsided as many would have you believe, Kendrick’s clear advantage comes from making more likely/routine plays, but Gordon’s range defensively has allowed him to make an absurd amount of even/remote/unlikely plays. I don’t need you to tell me how oversimplified this exercise is, but lets play with the data and give these two players the same amount of chances across the board, with 425 routine chances/ 28 likely chances/ 13 even chances/ 14 unlikely chances/24 remote chances (504 total plays to be made).

Using the percentages they had last season, Kendrick makes 416 of the routine chances/23 of the likely chances/7 of the even chances/3 of the unlikely chances/0 of the remote chances. (449 total plays made)

Gordon makes 411 of the routine chances/ 21 of the likely chances/11 of the even chances/7 of the unlikely chances/ 2 of the remote chances (452 total plays made)

And keep in mind, that these numbers come from 30/31 year old Kendrick, and 25/26 year old Dee Gordon, these numbers do not add in the potential age regression (however significant or insignificant it is) Kendrick might suffer at age 31/32, these numbers do not add in the potential growth a player with amazing range and only one prior full season at the position might enjoy.

It’s not all that hard arguing that Dee Gordon was an equal defender to Howie Kendrick this past season. It’s also not all that difficult envisioning a situation where Gordon takes a step forward defensively and outplays Kendrick next season, despite what UZR might say. I’ll peg this as a wash, optimistically and a very possible downgrade defensively next season.

ShortStop

CLEAR UPGRADE. NEXT.

Just kidding, but pretty much

[table id=16 /]

Did you expect differently? I’m just not going to debate Hanley Ramirez vs Jimmy Rollins. I still don’t believe Ramirez to be unplayable at SS especially if you take into account his offense, but this isn’t about offense, it’s about defense and how much better Rollins is going to be on that side of the ball. Rollins having an 1.7% edge on the routine plays, having an 14.4% edge on the likely plays, having an 8.3% edge on the even plays, having an 6.8% edge on unlikely plays, and no difference on remote/impossible plays is a definitive upgrade from the previous shortstop. Even if you account for Rollins not being quite as good in his age 37 season, it’s almost impossible seeing him worse than Ramirez was last season -doesn’t mean overall he’ll be good, but defensively this was a win for the Dodgers-.

Outfield

I’ll just discuss this as the outfield because there is a rookie entering the fold, and we would be dealing with pretty small numbers for 2014 Puig in CF and 2014 Kemp in RF.

I’ve talked about age based regression in this post a lot, generally pre-prime players are at the top of their game defensively. Getting younger mostly is a good idea when discussing defense, and going from Yasiel Puig and Matt Kemp, to Joc Pederson and Yasiel Puig is certainly going to be an upgrade.

For one, take a look at this scouting report from Baseball America

"“He’s probably the best player in the league overall,” one scout said. “He can do everything; He’s a really, really good center fielder; He’s aggressive on the bases. The only question is is the power legit?  … But the bat speed is there. He’ll play center in the majors and make it look easy, he can really go get it.”"

Defensively, that makes me feel really good about Joc Pederson in center field. I remember back in 2013 Pederson having questions about his ability to play center field. He’s still not a slam dunk to be a long term fixture there according to some but there’s no question that he’s the best major league play  the Dodgers have right now. Here’s a list with the top center fielders in the game sorted by DRS (defensive runs saved). This stat is far from perfect, like many defensive metrics are, but it gets the point across, that only the best of the very best can play CF. The eye test can help us out with the evaluation also. Being a top tier center fielder would be difficult to attain for Joc, as he would have to rank among the likes of Juan Lagares (28 DRS), Jackie Bradley Jr. (14 DRS), Lorenzo Cain (14 DRS), and Jarrod Dyson (14DRS). I’d -somewhat arbitrarily- say that Joc should perform a couple tiers below that, not an elite center field talent, but one that provides slightly above average to solidly average defense.

Don’t get caught up in the names too much because relative value drags down what a player actually does, focus on the talent level Pederson has, and it’s not unreasonable for Pederson to outperform a 0-5 DRS rating, especially considering that defense is a young man’s game. The conservative idea is that he’s not in the Lagares/Bradley/Cain range, and that’s not a knock on Joc.

I hate talking about Matt Kemp as a center fielder, because he will never ever play there again, and it drags his entire value down because we still do so. However, the Dodgers last season got about a 1.5 months of Kemp’s defense in center field, and while that month and a half was clearly below average, even if it wasn’t historically bad, you still have to account for it when discussing the CF situation. Here’s his inside edge fielding figures, they weren’t great in center field, moving him out of that spot and into a corner should account for reasonable improvement.

But speaking objectively i’m not sure you can say Matt Kemp would be a defensive plus in right field, he sure looked comfortable in the limited time he saw there last season, but small sample caveat should be taken into account. He isn’t good at getting reads off of baseballs hit off the bat, and relies on his athleticism to cut down on his mistakes. Luckily for Kemp he’s an amazing hitter, truly an elite talent, but for the sake of this write up, he’s an underwhelming defender to say the least.

And hey, even if you assume Kemp’s best case scenario, he shows he can still outrun most of his mistakes, his arm continues to be a cannon, and he doesn’t get injured, there’s still no way I take him over Puig from a defensive standpoint. Puig has an easy elite arm tool, he has incredible range, he’s always seemed like he just needs refinement at the major league level to become a true elite defender. Take a look at his inside edge fielding numbers. Despite the breath holding that accompanies a ball hit to Puig, he makes all of the easy plays, and a good chunk of the plays that he’s given a 1%-60% chance to make to where I consider it a plus to have him manning RF. Yeah there’s a lot to work with there, and the upside is still there, as opposed to Kemp who is 8 years older than Puig.

So the outfield should see a fairly large increase in overall defense if Pederson and Puig are above average defenders. That doesn’t sound like an unreasonable thing to ask from them, and there is some potential upside from them considering how young and toolsy they are.

What does this all mean?

This is a lot of stuff to be written on one subject. But my feelings can be summed up like this:

We are probably never going to see AJ put up good defense again, so despite Grandal’s shortcomings in passed balls, it’s not as if the alternative is any better. However what Grandal does do correctly is so much better than Ellis, that he’s probably a slight upgrade at worst even if you don’t believe in the framing stats as presented.

Howie Kendrick and Dee Gordon are pretty much an underrated wash with Gordon being a better bet going forward to perform defensively.

2042 Jimmy Rollins is a significant upgrade from 2014 Hanley Ramirez defensively, so 2015 Jimmy Rollins is a pretty safe bet to be miles better than Hanley ever was in a Dodger uniform.

Joc Pederson being average in center field is better than Yasiel Puig, was better than Matt Kemp. And Yasiel Puig in RF is better than Matt Kemp in RF, so it’s clear that the outfield has been upgraded fairly significantly.

A team might never significantly upgrade defensively at 3 up the middle positions in span of 24 hours like the Dodgers did by getting Yasmani Grandal, Jimmy Rollins, and clearing space for Joc Pederson.

Clayton Kershaw–Zack Greinke-Hyun-Jin Ryu-Brandon McCarthy-[LOL who cares] are probably going to set the new run prevention record for a staff behind the defense they have.