The Most Astonishing Early Statistic

When I wrote about Max Muncy, I pointed out how strange a season it’s been for the Dodgers. Many of the best players have been hurt, or they’ve underperformed. Meanwhile, they’ve been supported by surprises like Muncy, Ross Stripling, and Matt Kemp. Muncy has been one of the team’s better hitters. Stripling has been the team’s most valuable pitcher. And Kemp has been the team’s most valuable overall player. I mean that in terms of WAR — Kemp has the highest figure on the club. At +1.8, he’s already done more than he did in any of the previous three years.

I skipped right by that when I was writing about Muncy, because I was writing about Muncy. But it’s just as big a deal. Remember, nobody wanted Kemp. Even the Dodgers didn’t want Kemp. It appeared he’d declined into being a valueless player. When he was traded, he wasn’t traded because of his skills; he was traded because of his contract. Now we’re more than two months into 2018, and Kemp’s a major reason why the Dodgers are even alive. His present WAR alone might be the most astonishing early statistic.

I want to drill a little deeper. Kemp’s WAR is surprising. One component of his WAR is even more surprising.

Matt Kemp has been a tremendously productive hitter. With a wRC+ of 160, he looks like the guy he was in his prime. Kemp hitting well doesn’t come out of nowhere, however. Through this point just last season, Kemp’s wRC+ was 145. When he’s been healthy, and when he’s been focused, Kemp has remained dangerous. There aren’t many pitchers he isn’t capable of hurting.

But while Kemp has swung a good bat for much of his entire career, he’s also been a case study in why WAR is so significant. Used to be, you’d look at Kemp’s offense, and you’d figure, welp, he must be pretty solid. Yet in truth, in other areas, Kemp has given value away. He hasn’t been a good baserunner in a very long time. Far more important than that, he hasn’t been a good defender in a very long time. Anyone who’s read FanGraphs for any number of years knows that Kemp established himself as a major defensive liability.

You Aren't a FanGraphs Member
It looks like you aren't yet a FanGraphs Member (or aren't logged in). We aren't mad, just disappointed.
We get it. You want to read this article. But before we let you get back to it, we'd like to point out a few of the good reasons why you should become a Member.
1. Ad Free viewing! We won't bug you with this ad, or any other.
2. Unlimited articles! Non-Members only get to read 10 free articles a month. Members never get cut off.
3. Dark mode and Classic mode!
4. Custom player page dashboards! Choose the player cards you want, in the order you want them.
5. One-click data exports! Export our projections and leaderboards for your personal projects.
6. Remove the photos on the home page! (Honestly, this doesn't sound so great to us, but some people wanted it, and we like to give our Members what they want.)
7. Even more Steamer projections! We have handedness, percentile, and context neutral projections available for Members only.
8. Get FanGraphs Walk-Off, a customized year end review! Find out exactly how you used FanGraphs this year, and how that compares to other Members. Don't be a victim of FOMO.
9. A weekly mailbag column, exclusively for Members.
10. Help support FanGraphs and our entire staff! Our Members provide us with critical resources to improve the site and deliver new features!
We hope you'll consider a Membership today, for yourself or as a gift! And we realize this has been an awfully long sales pitch, so we've also removed all the other ads in this article. We didn't want to overdo it.

For that reason, this past offseason, Kemp was looked at as a glorified DH. That’s why there wasn’t a market for his services. But as I’m writing this, Kemp has a Defensive Runs Saved mark of 0, as a corner outfielder. Looking at Ultimate Zone Rating, he’s at 0.1. Let’s fold in the positional adjustment and examine Kemp’s entire big-league career, viewing his defensive values per 1,000 innings. That puts everything over a common denominator.

Matt Kemp is almost 34 years old. He has a fairly long track record of injuries. Between 2014 – 2017, Kemp was one of baseball’s very least valuable defenders. Now, a third of the way through the year, Kemp’s turned the clock back a half-decade. Kemp hasn’t been a liability by these numbers, and there are more numbers we can throw onto the pile. According to Statcast’s Outs Above Average metric, in 2016, Kemp finished in last place, at -23. In 2017, he again finished in last place, at -18. In 2018, he’s at 0. That’s 0 outs above average, which is the same thing as 0 outs below average. When you look at Kemp in the leaderboard, you find him right next to guys like Jason Heyward and Jarrod Dyson. This is our incredible early reality.

There are a lot of astonishing early numbers. Don’t get me wrong, you could make an argument for several of them. Nothing astonishes me, personally, more than this. More than Matt Kemp being an adequate defensive outfielder. I know we don’t usually celebrate basic adequacy, but given where Kemp has been in the recent past, this counts as something major.

How might one explain what Kemp has accomplished? It matters, of course, that so far in 2018, Kemp has been healthy. Just last year, he had problems with both of his legs, and it’s no surprise that could take a toll. But you might also remember that, in spring training, Kemp reported to camp down something like 40 pounds. Ordinarily, we dismiss those stories. Players are supposed to show up in great shape. But we can revisit something Mike Petriello wrote about in April. At least as far as sprint speed is concerned, Kemp has gotten faster. He’s gotten faster by 1.4 feet per second.

Compared to last year, that’s the third-greatest increase in baseball. Since 2015, it would be the fourth-greatest year-to-year increase on record. No, sprint speed doesn’t capture everything about running, but one’s top speed is important. Kemp seems to be moving better, which would follow from putting a lesser burden on his legs.

As far as Kemp’s positioning is concerned, I don’t see a major difference. I don’t think that’s the explanation. Part of this might be that Kemp has faced easier opportunities. In 2016, Kemp’s expected catch rate ranked in the 30th percentile. The next year, the 45th percentile. This year, the 96th percentile. Kemp hasn’t yet been challenged like he was before, and that could be helping him out. At the same time, the advanced defensive metrics are built to fold that in. All things considered, he’s been better than before.

Rest could help us out further, here. As we try to understand why Kemp is where he is, we can look at how the Dodgers have used him. In 2015, Kemp finished 87% of the games he started. In 2016, he finished 78%. In 2017, he finished 71%. In 2018, he’s finished 49%. This is probably a reflection of two different things. One, the Dodgers don’t think Kemp is a defensive plus. But two, the Dodgers are trying to preserve his health. The less often Kemp has to run around, the less likely he is to wear down.

One can learn only so much from a few carefully selected defensive videos, but here are four — two missed plays from 2017, and two fairly similar made plays from 2018. The present-day Kemp does appear more willing to take a chance, and that could just be the product of confidence. As Kemp moves around and feels more athletic, he might be more willing to put that athleticism to greater use.

A 2017 miss:

Another 2017 miss:

A 2018 catch:

Another 2018 catch:

I’ll remind you again, it’s not like Kemp is suddenly a Gold Glove candidate. He hasn’t improved to the point where his defense is a strength. He’s improved to the point where his defense just isn’t a weakness. From time to time, balls have gotten down that better outfielders would’ve caught:

But I don’t think the Dodgers much care that Kemp’s defense isn’t elite. Kemp’s defense was never elite. It’s important enough that he’s been about average, because average defense, with that bat, makes for a valuable player. A valuable player who, a few months ago, the Dodgers were trying to dump. They didn’t expect to end up in this position, but, wouldn’t you know it; early Matt Kemp has helped the 2018 Dodgers weather a terrible storm.

Because of Kemp’s history, he’s a higher-than-average risk to get injured. And as Petriello wrote in April, historically, Kemp has been at his fastest in the early parts of the season. He might end up hurt, or he might end up fatigued. Older players don’t often get the luxury of playing at 100%. But remember that this is a slimmed-down Kemp, and a motivated Kemp, a Kemp eager to make a better impression on a competitive club. And even if things go sideways in the near-term future, Kemp has kind of already done enough to further the Dodgers’ cause. He’s already been more than the team expected him to be. This is one time when the front office doesn’t deserve too much credit. Back before spring training, the Dodgers didn’t want Matt Kemp. Through two months and change, the Dodgers have needed him.





Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.

38 Comments
Oldest
Newest Most Voted
southie
8 years ago

Kemp looks like Billy Hamilton running the bases compared to Miggy. Oof marone he looks terrible.

RonnieDobbs
8 years ago

That’s crazy to lump all corner OF together. LF is as low stakes as it gets and the same can not be said of RF. I would also expect RF to be better athletes in general. Would you lump middle infielders together?

I am not sure that this isn’t a case study in why WAR is not significant or the flaws of the metric. The idea that it was correct about Kemp in the past is… well questionable. A legit bat is far more valuable than below average LF defense, which is also not something that you can measure accurately. I am not sure that this isn’t largely about the lack of accuracy of defensive metrics and flawed weights in WAR – actually that is exactly what I think it is. This an opportunity for improvement if you want…or an opportunity to rationalize a flawed tool and its usage.

nenrightMember since 2020
8 years ago
Reply to  RonnieDobbs

He’s rated better in RF than LF by both defensive metrics

RonnieDobbs
8 years ago
Reply to  nenright

That’s fun! Defense is tough to measure. I have dug into specific balls hit and probabilities and I have no faith in the accuracy. I have seen a ball that I thought should be caught that was not even classified in a defensive player’s zone. Thanks for that tidbit though – I don’t think of him as RF at all. His rating is going to be very dependent on the balls hit to him – in a quick look at inside edge it looks like he has pretty much had a bunch of routine balls hit to him – hard to fathom how that grades out as plus… but it does.

burts_beadsMember since 2025
8 years ago
Reply to  RonnieDobbs

Are you serious? Look at anybody’s inside edge, the vast majority will be routine. Then a fair amount of impossible. Then a handful of everything in between.

mikejuntMember
8 years ago
Reply to  RonnieDobbs

Kemp has been getting RF starts over Pederson when Puig is out of the line up due to his arm. He even got RF starts with Alex verdugo in LF when Puig was on the DL.

Charles
8 years ago
Reply to  RonnieDobbs

You might want to read up on positional adjustments: https://www.fangraphs.com/library/misc/war/positional-adjustment/

No, you would not lump middle infielders together.

The positional adjustments are based on actual measurements of players who switched between positions. LF and RF were calculated as historically equal on the scale at -7.5 per full season.

Someone took another look at it at a later point in time over the course of a decade and found a 2-3 run season advantage for RF over LF during that particular time frame, which is not a big deal. Here is a link to a Tom Tango blog post discussing the updated results: http://tangotiger.com/index.php/site/article/positional-adjustments-any-updates

And here is another more recent examination by Jeff Zimmerman, which found no difference between LF and RF: https://www.fangraphs.com/tht/re-examining-wars-defensive-spectrum/

RonnieDobbs
8 years ago
Reply to  Charles

I think this should be regarded as pretty concrete proof that the metrics are inaccurate as opposed to facts. When metrics aren’t measuring something accurately or telling you anything useful they might not be doing their jobs. In any case, thanks for the links I appreciate you trying to help me – even if I am beyond help! I think defensive metrics are poor at their jobs. I think people generally agree, but they don’t like to say it like that – they usually spin it more like they are improving. Please consider that the flawed methods of the past were worshiped at points in history as well. Defense is not and never will be measurable like hitting outcomes as it is all subjective. I think it is a noble pursuit but I am not sure that it is improving the quality of the game or the understanding – in fact, I think it promotes a lack of understanding. How else would you describe inaccurate, overly simplistic analysis?

RonnieDobbs
8 years ago
Reply to  Charles

I read those articles and I don’t understand the methodology. Thank you again for helping me though. As much as I am critical of this work, I do also want to understand what other people are doing. I think I just disagree with the underlying premises and I can’t forget distance myself far away from my knowledge to follow along. What would switching positions have to do with anything? Elite defenders don’t switch positions, so what are we even looking at? I look at InsideEdge data and Hit Probabilities through BaseballSavant and I just don’t see much value there. I don’t understand where the arbitrary metric of “runs” comes from either any more than “wins”. Real games are won with runs, but that isn’t what those metrics measure which I think is problematic… don’t get me started on “barrels” lol

bjsguess
8 years ago
Reply to  RonnieDobbs

I appreciate the tone of your skepticism. That said, modern fielding metrics are much more precise (and tell an important story) than you give them credit for. Conversely, many offensive metrics carry very little value and effectively distort the message they are trying to convey.

The beauty is that we have several different methodologies and a stack of metrics to help us understand (1) what happened and (2) what is likely to happen in the near future. One shouldn’t be focused on just a single metric or approach to understand this game.

sadtromboneMember since 2020
8 years ago
Reply to  Charles

To be fair, the positional adjustments are kind of a problem. They’re opaque, and even the FG page backtracks a bit saying “sure, we know that these are a little arbitrary, but it’s more important that we have something.” And that’s before we question exactly how good the measurements are for the “easiest” position (1B) and the “hardest” one (catcher), since there’s a ton we can’t measure there. The whole thing needs a much better explanation of methodology.

nenrightMember since 2020
8 years ago

Couldn’t the expected catch probability being so high be a product of his positioning? I feel like such a huge change would be more than just random luck.

Maybe seeing where his catch probability is in relation to other dodgers outfielders could help shed some light?

RonnieDobbs
8 years ago
Reply to  nenright

I would say that random luck sounds completely reasonable. Positioning is relative to the balls hit, which is out of his hands. Look at the underlying data if you are interested – its actually not all that common to not have routine balls hit to you. Over his career, 68% of balls in his zone have been routine – in his entire career he has had less than 500 balls hit at him in LF in nearly 2000 innings – 13% were impossible. Another fun stat is that the number of “impossible” balls is roughly equal to the number of balls that are in every other bucket outside of “routine”. Moral of the story – most balls are routine and there are not as many balls in the sample as you would think. This is all very susceptible to small sample sizes.

Ivdown
8 years ago

Very little has made me happier in 2018 than Matt Kemp’s resurgence. He has looked like the Kemp of old and is an NL MVP candidate to this point. That being said, I definitely do not expect this level of play to continue. However, if Kemp can hit something like .280/.325/.475 with below average defense the rest of the way, he’s still going to be fairly valuable to the team.

sadtromboneMember since 2020
8 years ago
Reply to  Ivdown

It’s hard to believe it has actually happened, much less that it will continue.

Not only does he have to sustain the BABIP luck, but in order to continue being functional in the field (and keep hustling out ground balls) his legs need to stay attached. And that is a really dicey proposition.

He probably needs to hit a little better than that to justify a roster spot given the likely regression to somewhat-below-average defense. What you outlined sounds kind of similar to Didi Gregorius’s line last year, and he put up a wRC+ of 107 last year. But if he can bop a slugging line of .500 for the rest of the way that puts him around Jonathan Schoop, who had a wRC+ of 120. And that’s practically a fairytale ending.

ShowTime98
8 years ago
Reply to  sadtrombone

“He probably needs to hit higher than that to justify a roster spot”. Maybe for the 2019 season. But as far as 2018 goes theres no question of Kemps place on the team

supadhyaya82
8 years ago

Is it fair to say “advanced defensive metrics are built to fold that in”? Not sure they’re built to adjust for very low “N”. Said differently, I’d love to know what his 2018 defensive value would be if he didn’t make the 2 catches in 2018 videos posted, so we can gauge sensitivity of literally just 2 opportunities.

Jonathan
8 years ago

It’s been a great reunion thus far, and ideally the Dodgers continue to pace his play to ensure he doesn’t wear out.

RonnieDobbs
8 years ago

Isn’t it silly that we are talking about Matt Kemp’s defensive value? Matt Kemp has hit very well this year… that is all we really should be talking about.

bosoxforlifeMember since 2016
8 years ago
Reply to  Jeff Sullivan

It was a complete surprise to me and I think virtually every other person who visits this site. i doubt that anyone believed that Matt Kemp was capable of delivering what he has provided to the Dodgers this year. Offensively and defensively he was washed up in the opinion of everybody and his performance has been, to say the least, astonishing

ChippersJonesing
8 years ago
Reply to  bosoxforlife

As a Braves fan I remember Kemp’s hot start to the season last year with the bat. I also remember every ball to left being terrifying.

Free Clay Zavada
8 years ago
Reply to  RonnieDobbs

Isn’t it silly that you’ve posted six comments with the same general skepticism that makes little sense?

Roger McDowell Hot Foot
8 years ago

The power of actually trying!

madfencerMember since 2017
8 years ago

“Part of this might be that Kemp has faced easier opportunities… This year, the 96th percentile. Kemp hasn’t yet been challenged like he was before, and that could be helping him out. At the same time, the advanced defensive metrics are built to fold that in.”

From what I understand about defensive metrics, I would think that a fielder that is consistently getting easy opportunities would naturally get stuck towards a neutral defensive rating. It’s only with difficult chances that a field can rack up positive or negative runs saved (assuming he’s not blowing the easy chances). So I think the fact that he’s in the 96th percentile in catch probability could exactly explain how he’s managed a surprising DRS of 0.

a different brad
8 years ago

I hate to be a wet blanket, but independent of his defense the difference in his bat between this year and last year is a small amount of power and 90 points of BABIP. He still never walks, and the most bullish projection system here only gives him a 112 wRC+ going forward. He’s in better shape and the balls are finding gaps, but there’s no reason to believe he can maintain anything like this.

drew_willyMember since 2020
8 years ago

His statcast numbers leave me wondering if his success at the plate this year is maybe not so easy to wave away on the grounds of a high BABIP – https://baseballsavant.mlb.com/savant-player/matt-kemp-461314?stats=career-r-hitting-mlb. Recall also that his career BABIP is .340 even though projections here call for his BABIP going forward to be more like .320.

To word things better, I guess I’m wondering if his statcast batted ball numbers should maybe cause us to expect his BABIP going forward to be nearer to his career BABIP than steamer or the depth charts projections might say. Maybe (once you account for launch angle, exit velo, and barrels data) going forward a BABIP of, say, .325 or .330 is more probable than the projections might make us think. I think Perpetua’s xStats support my idea that we should be careful to not dismiss his resurgence *too* quickly. He is making much better contact this year.

https://docs.google.com/spreadsheets/d/1k8d972fng34anlhbOJD9aUlVJgkR5mMesjDNNrmZ3Us/edit?usp=drivesdk

Noah BaronMember since 2016
8 years ago

A few things here. One, even in the 2018 videos, he still looks quite slow and immobile, even as he makes the catch. Second, you have to take into account the impact that positioning has on these metrics. I’d imagine that the Dodgers are superior in their positioning than the Braves, which might be part of the reason his metrics have improved. Finally, these metrics take a while to stabilize, so perhaps we should wait more than 2 months before declaring Kemp’s defense to be fixed. Although it is an interesting trend.

carterMember since 2020
8 years ago

I think the weight loss and increased speed is a real thing. I also think that he is trying a lot harder. People mature at different rates, and the way he looks/acts looks entirely different. Before he looked like he wanted to cash a paycheck and date Rhianna, and did well just simply because he was talented. Now he looks like he was tired of being a laughingstock and wants people to see “hey I can still be good. “

Original Greaser Bob
8 years ago
Reply to  carter

I want an article on players pre and post dating Rhianna careers.

PemtlcMember since 2026
8 years ago

She’s like Kim Basinger in The Natural.

'Tungsten Arm" O'Doyle
8 years ago
Reply to  carter

Nothing like a salary dump trade to get under your skin, and make you want to prove something. That, and the increasing signs that Father Time isn’t your biggest fan.

tung_twista
8 years ago

Being traded to the Braves was already an obvious salary dump considering Olivera got released right away.

ArmadilloFury
8 years ago

I vote for Ross Stripling leading all Dodgers pitchers in WAR & FIP to be the most astonishing statistic

London Yank
8 years ago

If almost all of his opportunities have been routine then you have no information with which to measure him against other players. Of course he will come out as neutral. If anything, these data are a cautionary tale for WAR, not an endorsement of it.

insidb
8 years ago

Can someone explain to me why he looks like CC Sabathia in the Braves-Mets game?

TKDCMember since 2016
8 years ago

The Braves were looking to get rid of one of two aging, replacement-ish level corner outfielders, got rid of the one that is now 6th among NL outfielders in WAR, and oh by the way made the right decision. Maybe Kakes and Kemp can patrol the outfield together once again at the ASG?

websoulsurfer
8 years ago

Since the defensive metrics you used have zero value without utilizing 450 games of data to make your performance guesstimate, your cool looking chart has no meaning.
If you want something with worthwhile information, go back and do one in 3 season increments. That will have value.
OR just look at StatCast OAA. You will find Kemp down at -1 so far this season. Below average. Last season you will find him at -18 or dead last. In 2016 you will find him at -23 or dead last.