Elegy for ’18 – Los Angeles Dodgers

Clayton Kershaw wasn’t his peak self. But not-peak Kershaw is still pretty great.
(Photo: Arturo Pardavila III)

Though undoubtedly a successful franchise, World Series victory again eluded the Dodgers in 2018. After finishing as the runner-up, or first loser, depending on how inclined you are to glasses being empty or full, the Dodgers have now gone 30 years without winning the Fall Classic. It’s hard to weep too much at the funeral dirge of a team coming off six consecutive first-place finishes, but it’s been a disappointing run of not being able to close the deal.

The Setup

The Dodgers are an organization that represents, in some ways, the worst fears of the analytical community of 15 years ago. It’s one thing to tell small-market teams to be smarter and not have the Cam Bonifays or Chuck LaMars or Dave Littlefields making decisions. But what would happen when, one day, a very rich team also puts together an extremely progressive, highly competent front office?

That’s not to say the Dodgers were a backwards organization; the team was run by the extremely competent Dan Evans and then by one of Billy Beane’s chief paladins in Paul DePodesta. But what those two did not have was an organizational commitment to put together a bleeding-edge unit with a unified, top-to-bottom purpose like those we see in organizations like today’s Astros or Indians.

Ned Colletti did an excellent job of keeping the farm system running at top-notch efficiency, but he also wasn’t the one to fully leverage the team’s assets in a modern way. To do that, team president Stan Kasten saw the opportunity to bring in Andrew Friedman from the Rays, who in turn hired Farhan Zaidi and Josh Byrnes.

The Dodgers didn’t suddenly become the Rays or anything, but there was a definite avoidance of aggressive spending in free agency back before thrift was cool. We’re starting the fifth year of the Friedman era and the Dodgers still haven’t signed a player who wasn’t most recently a Dodger to a $50 million contract, instead focusing on bringing back incumbent Dodgers such as Justin Turner, Kenley Jansen, and, most importantly and recently, keeping Clayton Kershaw from hitting the open market.

No one was completely irreplaceable; the Dodgers would have preferred to hang onto Zack Greinke, but the cost proved too great. Instead of giant commitments, the Dodgers have been more interested in shorter, more targeted pushes, bringing in Rich Hill and then Manny Machado for stretch drives. But even then, the team will frequently say no when the price is too high, being willing to see pitchers like Walker Buehler, or earlier, Julio Urias, as real crunch-time contributors, not just as assets to flip for veterans.

The winter of 2017 was especially quiet for the Dodgers, with the team joining its large-market brethren in a quest to stay under the luxury tax threshold for 2018. Los Angeles’ biggest signing was the return of Chase Utley, a bit of a gamble for a team that seemed to be falling off a cliff for much of the final month of the 2017 season.

Nor did the team make any big trades. Yes, they flipped Adrian Gonzalez, Scott Kazmir, Brandon McCarthy, and Charlie Culberson to the Braves in return for Matt Kemp, but 2012’s blockbuster was simply 2017’s accounting stratagem.

The Projection

Coming into the season, ZiPS projected the Dodgers tying with the Houston Astros for the best mean projection in baseball, at 95 wins. That represented a significant drop — nine wins — from the team’s 104-58 2017 season, but as anyone who pays attention to these sort of things can guess, 104 wins needs some things to go right; it’s literally no team’s projection.

Even with that fairly yummy forecast, ZiPS saw some divisional risk for the Dodgers, giving the NL West a one-in-five chance of being conquered by a team that doesn’t play in Chavez Ravine. ZiPS saw Arizona, San Francisco, and Colorado as all having winning records in 2018, and it would only take one of the three teams hitting their upside to end LA’s divisional streak.

The Results

The projection didn’t end up too badly, with the Dodgers winning 91 games before seasonal overtime, just four wins below the projection; that isn’t a bad miss as these things go. But the Dodgers certainly didn’t feel like a top team for much of the 2018 season.

For a time, it seemed like all three of the Dodgers’ non-Padre rivals could best the team in the West. Los Angeles was in fourth place as late as June 6 with a 30-31 record, and unlike their early 2017 struggle to get some space from the rest of the division, the team actually just wasn’t playing all that well.

Matt Kemp, basically acquired as a luxury tax MacGuffin, actually proved to be fairly necessary early; the team’s .707 OPS through May ranked just 23rd in baseball. Cody Bellinger wasn’t hitting, Yasiel Puig wasn’t hitting, Joc Pederson wasn’t hitting, Corey Seager was out for the season, and Justin Turner, recently returned from a broken wrist, wasn’t hitting yet.

That was a lot of talent not performing well and it would have been far worse without the contributions the Dodgers received in unexpected places. Kemp finished May with a 150 wRC+ and was playing defense like an actual major leaguer, and while that wouldn’t persist throughout the rest of the season, it was very welcome at the time. Max Muncy, a 27-year-old who entered the season with a .195/.290/.321 line, had a 133 wRC+ through the end of May and nearly made the All-Star Game (he did make the Home Run Derby).

Then, everything just kind of worked out. Ten Dodgers got at least 200 plate appearances after May and the worst wRC+ of the group was Matt Kemp’s 107, with eight better than 120, and four eclipsing 140. Clayton Kershaw stayed healthy in the second-half, and while he appeared to be the post-peak Kershaw version of himself, 90% of peak Kershaw is a damn good pitcher. The team added to their riches by closing a deal for Manny Machado, the best player realistically available at the deadline.

From that June 6 line, the Dodgers went 62-40, the best in the National League. And after all that, it almost wasn’t enough. The Colorado Rockies hung on to force a 163rd game.

What Comes Next?

The Dodgers haven’t closed any deals with Bryce Harper or Manny Machado yet, but the expectation remains that they are going to be much freer with the pocketbook than in recent years. The team has moved $60 million off their payroll from several years ago. At one point, the Dodgers had more than half a billion in payroll baked into the cake; now their guarantees are only half of that, with none of the deals looking like bleeding disasters, and only Kenta Maeda’s dirt-cheap contract stretching past a third season into the future.

That’s not to say the Dodgers are going to splash cash. The team cleared some payroll commitments by trading Yasiel Puig, Alex Wood, and Matt Kemp to the Reds, a deal in which the only notable salary they brought back was Homer Bailey’s. I won’t venture a guess as to the exact savings, since I think you need a forensic accountant to figure out who is paying Matt Kemp what in 2019.

Even if they stood pat — and I don’t think they will — the Dodgers will almost certainly enter the 2018 as the favorite, and perhaps by a better projected margin than last season despite having a worse year. The Giants are a lot less threatening than they appeared to be entering 2018, Arizona is rebuilding, and the Rockies are an organization that, well, still thinks that Ian Desmond is a major leaguer.

ZiPS Projection – Clayton Kershaw

I mostly wanted to see how Kershaw’s recent durability issues and minor downturn would affect with career projections.

ZiPS Projection – Clayton Kershaw
2019 11 6 2.91 26 26 164.0 138 53 18 27 170 139 4.1
2020 10 5 2.97 24 24 151.3 131 50 17 25 155 136 3.7
2021 9 5 3.10 23 23 145.0 128 50 17 24 143 130 3.3
2022 9 5 3.11 21 21 133.0 118 46 16 22 132 130 3.0
2023 8 5 3.21 20 20 123.3 110 44 15 21 122 126 2.7
2024 7 5 3.27 18 18 113.0 103 41 15 20 111 123 2.4
2025 6 4 3.34 17 17 102.3 94 38 14 19 100 121 2.1
2026 6 4 3.49 15 15 90.3 86 35 13 17 87 116 1.7
2027 5 4 3.65 13 13 79.0 77 32 12 15 75 111 1.3
2028 4 3 3.84 11 11 68.0 68 29 11 14 63 105 1.0
2029 3 3 4.08 10 10 57.3 59 26 10 12 52 99 0.6
Past 153 69 2.39 318 316 2096.3 1570 557 145 536 2275 159 61.6
Future 78 49 3.26 199 198 1226.7 1113 444 158 217 1208 125 26.0
Total 231 118 2.71 517 514 3323.0 2683 1001 303 753 3483 143 87.6

ZiPS is clearly worried about Kershaw’s inning totals, which is really putting the kibosh on his final win total, the projection for which peaked as high as 270 at one point. Kershaw hasn’t had any major injuries of the Tommy John surgery-requiring variety, but the truth is that he’s failed to throw 200 innings for four of the last five seasons, and failed to qualify for the ERA title in two of the last three (you can’t add an infinite inning the same way you can add a hitless at-bat for those who fall short of 502 plate appearances).

There’s also no methodological difference that would account for most of this change, so I’m not comparing apples and oranges. Using the latest build of ZiPS, Kershaw’s peak projection would have been 267 wins. Comparing apples and oranges isn’t really weird anyway; when I’m at the grocery store, there’s apple juice and orange juice next to each other, so I’m literally comparing them all the time.

If Kershaw can get through a couple of seasons unscathed, that number should shoot back up. But there’s real risk here.

Dan Szymborski is a senior writer for FanGraphs and the developer of the ZiPS projection system. He was a writer for ESPN.com from 2010-2018, a regular guest on a number of radio shows and podcasts, and a voting BBWAA member. He also maintains a terrible Twitter account at @DSzymborski.

Newest Most Voted
Inline Feedbacks
View all comments
5 years ago

lol at Kershaw’s projection. Sorry but for a guy as proud as Kershaw, absolutely no way he would go that long like that.

5 years ago
Reply to  stever20

The projections for established players always seem kinda’ pointless to me. Players project to be slightly lesser versions of who they have been recently, as they get older. Projected drop-offs are slow and gradual. I know the systems are set up to make the best possible guesses, but “player loses .3-.4 WAR per year for the next ten years” is the most boring projection. A system that saw Matt Muncy’s 2018 coming, or even in the possible outcome set, now that would be interesting. Usually wrong, but interesting at least.

I assume Kershaw’s projection is also pitching only – his hitting was a huge plus in 2018 (for a pitcher), adding a significant chunk to his 2018 value – any guesses on whether he continues that?

Cave Dameron
5 years ago
Reply to  aweb

So you want a projection system that makes wild predictions because a few of them may turn out to be true. That’s not how projections work, that’s called making guesses.

5 years ago
Reply to  Cave Dameron

Luckily for aweb, we have a thing called “The Internet” where people make “Usually wrong, but interesting at least” predictions every day!

5 years ago
Reply to  aweb

Just to directly contradict what you’re saying, ZiPS projects him for more WAR this year than he had last year.

5 years ago
Reply to  stever20

ZiPS is actually projecting Kershaw to still be a league average pitcher in 2029 at age 41, which is actually pretty incredible. That’s a better ERA+ projection for Kershaw in 2029 than ZiPS is projecting for Jake Arriera, Nick Pivetta, or Zach Eflin in 2019. If he can actually still be a significantly better than average big league pitcher into his late 30s, like Dan is projecting, I’m not sure why Kershaw would be too “proud” to do that.

5 years ago
Reply to  BenZobrist4MVP

Only making 66 starts his last 5 years? Sorry but I don’t think there’s a chance that happens. Show me 1 pitcher who has ever had a career end the way that’s projected? Yeah good when he pitches, but only making 10-15 starts a year?

Dave T
5 years ago
Reply to  stever20

These projections – like any decent or better projections for baseball players – are logically some sort of weighted average of a range of outcomes.

For multi-year projections, we’d also expect the actual observed performance in a given future year to correlated with other years because year-to-year health and effectiveness are correlated: in other words 2027 season performance will correlate with 2026 and 2028.

The claim that you are attacking is a strawman that misses the point of the projections.

5 years ago
Reply to  stever20

It’s not a “life-choices” projection, it’s a stats projection. It will project stats for any year you ask it to. I’m sure it will project his 2039 stats and his 2119 stats if you ask it to.

Psychic... Powerless...
5 years ago
Reply to  stever20

So in addition to forecasting on-field performance, ZIPS should also predict when a guy decides to retire?