keyboard_arrow_uptop

It’s hot stove time, when we shift our focus from the season that was to the season on the horizon. Tires are kicked, trades are proposed, and free agents are considered. During it all, metrics like WARP allow us to summarize the performance of individual players. But what about the overall core of a team’s major-league roster? How can we say, objectively, whether a team has built a core of ongoing contributors? Or, by contrast, whether it has been overly reliant on transient (e.g., departing) assets?

A productive core consists of two things: good players under long-term control, and good players who are not too old. Players under long-term control allow a team to be patient and avoid expensive, volatile solutions (except in those cases when the players under long-term control become those expensive, volatile solutions); players who are younger also tend to play more games, be more productive, and remain productive for longer.

To evaluate each team’s production from core contributors, I created a profiling system called Core Value. The system debuted at the conclusion of the 2013 season, and we used PECOTA before the 2014 season to project possible improvements. Now, with the 2014 season concluded, we can see what actual progress was made.

Methodology

The method is as follows:

1. Tabulate and combine all batting (by BWARP) and pitching (by PWARP) contributions by each player in the major leagues this year.

2. Create a Control Index by weighting the years of remaining team control covering each player. A maximum of five years of team control is credited, and the variable is centered at two years. So, for example, players under five years of team control have a Control Index of 2.5 (5/2); players in a contract year have a Control Index of .5 (1/2).

3. Create an Age Index by weighting the proximity of each player to a peak baseball age, or more precisely, the age at which we start seeing serious and continued decline. As with the previous studies, I’ve used a peak age of 27 for all players and positions, although there are other (very good) opinions on that issue. [1] This means that a 30-year-old player will be penalized at a rate of .9 (27/30), whereas a 22-year-old player will experience a bonus rate of 1.23 (27/22).

4. We calculate so-called core “wins” or “value” by multiplying each player’s total WARP times that player’s Age Index and Control Index. The aggregate core wins for a team is tabulated, and the original aggregate WARP is subtracted. The resulting figure, which I call Core Win Differential, reflects the extent to which the team’s contributions were provided from younger, controlled players.

5. Although roster depth is important, teams strongly benefit from having their stars also be a part of their core. For that reason, we also track so-called “core players” — stars who amass 5 or more core wins over the season. [2] We tabulate both the total number of core players on a team, as well as the number of pitchers who are core players, as good rosters tend to have both.

6. Finally, we rank each team's comparative performance, from 1 to 30, in each of these three categories, and average the three scores to come up with a final, composite rank of what we call Team Core Value.

Before we provide the results, there are two caveats. First, this system is descriptive, not predictive. We are telling you how each team did this particular year. This does not necessarily mean the team will do the same or better next year, or that there is no other path to success.

Second, for a variety of reasons—injury, distractions, bad luck, etc.—a player’s performance in a given year might not be representative of their true talent. Some players have off years, and others have fluky good years. Again, we are telling you how the players, and their teams in the aggregate, actually did; you, the reader, can decide whether you think a particular team’s performance is representative of future results or not.

Results

Here are the team rankings from 2014:

TEAM

Core Wins Differential

Total Core Pitchers

Total Core Position Players

2014 Team Core Value Rank

2013 Team Core Value Rank

2013-2014 Change

Dodgers

48

3

6

1

13

+12

Indians

46

2

3

2

4

+2

Pirates

41

1

5

3

6

+3

Astros

43

1

3

4

29

+25

Mets

37

1

4

6

14

+8

Angels

42

1

3

5

8

+3

Cardinals

41

3

2

7

1

-6

Mariners

39

1

4

7

21

+14

Braves

33

1

5

7

1

-6

White Sox

39

2

2

10

23

+13

Giants

31

1

3

11

10

-1

Tigers

26

2

3

12

19

+7

Rockies

30

1

3

13

3

-10

Rays

31

2

2

14

12

-2

Cubs

37

1

2

15

26

+11

Marlins

40

0

4

16

15

-1

Yankees

25

2

2

17

30

+13

Twins

29

1

2

18

28

+10

Athletics

27

1

2

19

17

-2

Brewers

27

0

5

20

11

-9

Nationals

25

0

4

21

5

-16

Orioles

16

1

2

22

17

-5

Diamondbacks

33

0

2

23

15

-8

Royals

25

1

1

24

21

-3

Reds

24

0

3

24

9

-15

Rangers

22

1

1

26

7

-19

Phillies

13

1

1

27

24

-3

Blue Jays

15

1

0

28

20

-8

Red Sox

17

0

2

28

25

-3

Padres

21

0

1

30

27

-3

Don’t get too excited about a difference of a few spots between teams. In general, the meaningful locations are (1) the upper quartile (the top eight teams), (2) the bottom quartile (the bottom eight teams), and the remaining teams in the middle (the interquartile range). Slight changes in methodology, such as lowering the peak age by a year or two, or using WAR numbers from a different method, can move teams within brackets a bit, but the upper and lower quartiles, particularly at the extremes, tend to be fairly stable. This is encouraging and suggests some robustness to the approach.

What do we learn from this year’s rankings, particularly as compared to last year? We could talk at length about every team, but a few developments jump out:

It’s Nice to Have Lots of Money . . .
Money seems to buy fewer wins than ever, but as in real life, more money makes things easier. The Dodgers are this year’s Core Value champs, and they can largely thank their ability to trade for contracts like that of Adrian Gonzalez, extend Clayton Kershaw, sign Zack Greinke off the open market, and recruit Yasiel Puig as a young international free agent. At the same time, the Dodgers have not merely purchased their success: Scott Van Slyke was a late-blossoming draftee, Kenley Jansen was an undrafted free agent, and the Dodgers did draft Clayton Kershaw to begin with. Overall, the Dodgers have a stable core of roster assets on which to rely in 2015, provided they can agree on who gets to play each day.

But Money Does Not Guarantee Success.
One subtext of the 2013 World Series was how differently the two teams involved had been constructed: The Cardinals ranked first in our measures last year, whereas the Red Sox, thriving on production from short-term free agency moves, ranked near the bottom. The Red Sox of course triumphed, showing that a young roster core does not automatically translate to championships. Acknowledging this, we nonetheless noted the following about the Red Sox:

Their roster generated more WARP than any other club in the 2013 regular season. However, a great deal of that production is heading out the door, and the aging remnants should struggle to replace it from within. . . . Last season may have been do or die year for the Red Sox. They registered only three Core Players, and none of them was a pitcher. Of their top 10 contributors in 2013, five of them are on the open market. The Core Wins formula discounts the contributions of the departing free agents accordingly, and the Red Sox now face a predicament heading into 2014. With a top-rated farm system and substantial financial resources, they have more ability to replenish than most, but they have a lot of work ahead of them.

This assessment proved prophetic. Jacoby Ellsbury left, other players on the wrong side of 27 regressed, and the young players counted on to fill the gap—like Jackie Bradley Jr. and Xander Bogaerts — were not yet up to the task. Of course, thanks in part to that disappointment, the Red Sox now have a protected draft pick in addition to their fine farm system and deep pockets, so no one is feeling sorry for them.

By contrast, while the Red Sox floundered early and often, the Cardinals, despite injuries, similar pitching regression, and an overall slow start, managed to grind out 90 wins in a tougher division and returned to the postseason.

Good News in Cleveland
Cleveland did extremely well in these rankings last year, ranking fourth in the league for roster core quality. I didn’t comment on it at the time, as Cleveland had just enjoyed a Cinderella season and its team WARP obviously reflected that. But here we are for a second year in a row, and Cleveland not only maintained its top-quartile status, but moved up. It did so by coaxing strong performances from additional core assets for a second year in a row. Whereas last year featured superb production from Jason Kipnis and Carlos Santana, this year the club added outstanding performances from Michael Brantley and Corey Kluber (both 6+ WARP). In a fairly weak division, this emerging roster core has the needle trending up.

Worrisome News in Cincinnati
The best parts of Cincinnati’s rotation are approaching free agency, and the Reds appear to have extended the wrong starter (Homer Bailey). In the lineup, Joey Votto is under long-term control, but didn't stay healthy. Jay Bruce, despite playing 137 games, had less BWARP than two members of the starting rotation.

Two years ago, Cincinnati was the envy of the league, with a potent combination of pitching, defense, and lineup power; now, as they tumble to nearly last in these ratings, there is a sense that Cincinnati’s window is closing fast. Both Votto and Bruce should regress positively next year, but Cincinnati’s core seems to be withering before our eyes.

Top Individual Performances

We’ll conclude by listing the individuals who offered the highest core value to their teams this year. Here are the top 25:

NAME

TEAM

AGE

BWARP

PWARP

Control Years

Core Wins

Mike Trout

ANA

22

9.41

5

29

Clayton Kershaw

LAN

26

0.8

6.11

5

18

Andrew McCutchen

PIT

27

7.08

5

18

Yasiel Puig

LAN

23

5.79

5

17

Josh Donaldson

OAK

28

6.71

5

16

Anthony Rizzo

CHN

24

5.63

5

16

Anthony Rendon

WAS

24

5.54

5

16

Kyle Seager

SEA

26

7.29

4

15

Michael Brantley

CLE

27

6.03

5

15

Buster Posey

SFN

27

5.93

5

15

Corey Kluber

CLE

28

6.03

5

15

Jose Abreu

CHA

27

5.54

5

14

Giancarlo Stanton

MIA

24

8.18

3

14

Jose Altuve

HOU

24

4.78

5

13

Robinson Cano

SEA

31

6.15

5

13

Chris Sale

CHA

25

0.01

4.72

5

13

Nolan Arenado

COL

23

4.33

5

13

Paul Goldschmidt

ARI

26

4.81

5

12

Jose Quintana

CHA

25

0.1

4.52

5

12

Miguel Cabrera

DET

31

5.53

5

12

Jonathan Lucroy

MIL

28

5.99

4

12

Freddie Freeman

ATL

24

4.1

5

12

Jean Segura

MIL

24

3.84

5

11

Brian Dozier

MIN

27

4.29

5

11

That Mike Trout guy is the worst: He always needs to be on top of all the lists.

You’ll notice a few trends here: the most valuable core players are under control for four or more years, either through free agency contracts or the collective bargaining agreement. They also tend to generate 4 or more WARP, and usually, but not always, are on the right side of 27. Naturally, as you consider the value of these players in future seasons, you need to take account of their Control and Age Indexes accordingly.

Conclusion
Team Core Value is just one way of profiling a team’s roster. But knowing the quality of a team's core allows for more informed use of future projections and suggests the best focus for the team’s farm system, as well as the ideal uses of a team’s payroll flexibility. Certain teams have the luxury of relying largely on their current rosters, whereas others have some (comparatively risky) shopping to do.

Jonathan Judge is a product liability lawyer. You can follow him on Twitter (where he talks about baseball, not product liability) at @bachlaw.

***

Appendix

Here is the list of core players by team:

Team

Total Core Players

Core Players

Dodgers

9

Puig, Kershaw, Kemp, Gonzalez, Ryu, Van Slyke,
Greinke, Gordon, Turner

Indians

5

Brantley, Kluber, Gomes, Santana, Carrasco

Cardinals

5

Carpenter, Wainwright, Peralta, Martinez, Lynn

Pirates

6

McCutchen, Harrison, Marte, Mercer, Cole, Walker

Mets

5

Lagares, deGrom, Tejada, Duda, d'Arnaud

Mariners

5

Seager, Cano, F. Hernandez, Miller, Taylor

White Sox

4

Abreu, Quintana, Sale, Eaton

Astros

4

Altuve, McHugh, Carter, Springer

Braves

6

Freeman, Upton, Wood, Hayward, Gattis, Simmons

Angels

4

Trout, Calhoun, Richards, Pujols

Tigers

5

Cabrera, J.D. Martinez, Verlander, Sanchez, Kinsler

Rays

4

Longoria, Kiermaier, Odorizzi, Archer

Giants

4

Posey, Bumgarner, Pence, Crawford

Rockies

4

Arenado, Tulowitski, Dickerson, Matzek

Yankees

4

Ellsbury, Gardner, Betances, Tanaka

Cubs

3

Rizzo, Castro, Arrieta

Marlins

4

Stanton, Yelich, Ozuna, Hechavarria

Twins

3

Dozier, Hughes, D. Santana

Athletics

3

Donaldson, Gray, Norris

Brewers

5

Lucroy, Segura, Gomez, K. Davis, R. Braun

Orioles

3

Jones, Machado, Gausman

Nationals

4

Rendon, Werth, Harper, Span

Royals

2

Ventura, Cain

Rangers

2

Darvish, Beltre

Diamondbacks

2

Goldschmidt, Inciarte

Phillies

2

Hamels, Revere

Blue Jays

1

Stroman

Reds

3

Mesoraco, Hamilton, Frazier

Red Sox

2

Pedroia, Betts

Padres

1

Grandal



[1] Jeff Zimmerman found that peak ages seem to be declining. Jeff’s most recent calculations, which he graciously shared, suggest that hitters begin declining (by wRC+) at 25, and pitchers (by FIP-) at about 26. I re-ran my numbers with those peak ages and found they did not meaningfully change the results.

[2] Last year, we used total core players (pitching plus position) instead of breaking position players out separately. After some discussion, we think keeping pitchers and position players separate makes more sense. Our references to last year’s rankings have been updated to take that into account.

You need to be logged in to comment. Login or Subscribe
greenengineer
11/12
Perhaps I missed it? What is the difference between the first table called "Here are the team rankings from 2014:" and the second table called "Here are the team rankings from 2014:"?
RossBukouricz7
11/12
There really isn't. I wonder if he meant to either post the projected 2014 or actual 2013 for one of the graphs?
bachlaw
11/12
Yup, the second table shouldn't be there. We'll get it fixed and let you know when it's updated.
lyricalkiller
11/12
Formatting error, ignore the second table, there is no second table, there never was a second table, speak not of a second table.
chapmantime
11/12
Way too soon to pronounce Bailey as the wrong pitcher to extend, with Cueto and Latos having equally problematic health histories and considering Bailey's stuff and ability.
Dawgswood
11/12
Why no Teheran under Braves?
RossBukouricz7
11/12
His WARP was only 1.3 for 2014, so even a good Age and Control index weren't enough to get him to 5 core wins.
dcj207
11/12
Using only most recent season WARP makes this highly leveraged to recency. Machado tweaks his other knee, and falls from second (23.1 after 2013 season) to outside the top 25 on the control index? Also, I know this metric isn't trying to be a tradeability index, but other than age adjustment factor any thought to including cost of control? Altuve and Cano may represent similar years worth of controlled WARP, but at vastly different costs.
bachlaw
11/13
It is definitely leveraged to recency. Machado is still under control, but he didn't play much this year and didn't play well when he did. So, not a core player showing in 2014. I've considered cost of control, but since the costs of control are essentially sunk either way, I don't think it adds much. The issue is that outside of the factors I'm considering, there really isn't a "mean" response to other conditions that can be used to further rate rosters, in my opinion.
davezahniser
11/12
Most surprising statement in this article? That the Cardinals were in a tougher division than the Red Sox.
ColKiner
11/12
Shouldn't there be more to the core of the Royals? (Gordon, Perez, Holland) all contributed more than Cain or Ventura to that team this year.
bachlaw
11/13
Nope. As I discussed earlier this year, the Royals, like the Orioles, were entering a "win now" situation. Of course, that is what they both did, so there you go.
crperry13
11/12
I struggle to understand how Dallas Keuchel only has 1.1 PWARP. Particularly considering how, by rival measures, he was the 10th-most valuable pitcher in the American League in 2014.
froston
11/13
I think the approach here is interesting and I think it illustrates what it tries to illustrate fairly well. I also understand that a lot of this is pretty arbitrary - the indexing factors for age and control; the cutoff for how much control matters; the cutoff for a "core player". However, I disagree strongly with not including salary information in any way. The price of a team's core is extremely important. It dictates how much funds can be appropriated to non-core players in any given season. I think there is probably a sensible way to evaluate team cores while including crucial salary information.
bachlaw
11/13
Thanks. Much of it is arbitrary, in the sense that I've decided what has value. Again, much of it is based on the Rays up until last year and what they did. For the reasons stated above, I'm not a fan of weighting salary in a league of guaranteed contracts. I'm also suspicious that expressed concerns about salary are really ones about ownership priorities and profit-taking. Teams with a desperate desire to win now often seem to find extra money; those that find it advantageous to plead poverty are happy to do so. I'm not saying there aren't some serious resource-disparities, but if owners don't respond to extra money in the same average way, salary flexibility doesn't have a lot of inferential value. Thanks for reading and commenting.