Much has been made about the “core” of a baseball team. This is the set of star players around which a general manager adds pieces and parts in hopes of building a champion. But what makes a core? Older players? Pitchers? Good “clubhouse guys”? This post of Dave Roberts’ Dive will utilize an incredible August 12th article in the “community research” section of FanGraphs.com to analyze the Red Sox “core” over the past few years. The article, written by Jonathan Judge, can be found here, and is titled “Does Your Team Have a Winning Core? Profiling Sustainable Roster Construction”.
Judge uses three elements; fWAR (wins above replacement, as defined by FanGraphs), a control index (the remaining years of player control the team has, maxing out at 5), and age index (peak baseball age — in this case, 27 — divided by true age). Judge defines “Core Wins” as the product of each factor (i.e. fWAR*[Control Index]*[Age Index]) and further defines a “Core Player” as one with more than 5 “Core Wins” in a season. In his analysis, he notes that the Rays are the “gold standard” of a roster made up of “core players” whereas the New York Mets lie on the opposite end of the spectrum. Let’s use Judge’s formula to take a look at the core make-up of the Boston Red Sox over the past few of years, starting in 2010, and see how this profile might translate to future success.
2010, 89-73, 3rd place AL East, missed playoffs
I’ll start by walking through a calculation of “Core Wins” for a single player to get the gist of what it entails. Consider the example of Jon Lester; the Red Sox had five more years of team control over Lester in 2010, so his control index stood at 2.5. In his age 26 season (1.04 age index), he pitched to a superb 5.4 WAR. Multiplying the factors (1.04*2.5*5.4), we extrapolate that Lester accumulated 14.0 “Core Wins” in 2010, making him a “Core Player” for Boston by the author’s definition. There were a few other Sox players (four, to be precise) who made the cut:
Some notes: Jed Lowrie was nearly a Core Player (he had 4.9 Core Wins) with a solid 1.9 WAR. The Red Sox player with the most WAR was Adrian Beltre at 6.6. But the formula obviously rewards teams for have younger, controllable players and Beltre was a 31-year old who would became a free agent after the season. Thus, he accumulated just 2.9 Core Wins. The Red Sox accumulated 71.9 Core Wins on aggregate during the 2010 season. Keep that number in mind when comparing to future seasons.
2011, 90-72, 3rd place AL East, missed playoffs
Ah, yes. The epic collapse that was the motivation for a complete change of scenery by John Henry and Larry Lucchino. The Sox signed Carl Crawford and traded for Adrian Gonzalez in the offseason, which translated to tremendous success through August. But the wheels fell off in September and the rest is history. How did the Sox fare by the “Core Wins” and “Core Player” analysis?
Notes: The Red Sox had only four players in their “core”. One reason for this was the attrition of Jon Lester, Clay Buchholz, and John Lackey. Jacoby Ellsbury was one of the most exciting players in baseball during the season, and his power surge led to a 9.1 WAR and a second-place finish in the AL MVP voting. Carl Crawford, who signed a 7-year, $142 million contract in the offseason, was one of baseball’s biggest busts ever at a -0.1 WAR. The benefit of Dustin Pedroia’s contract extension is clear after he spent two consecutive years in the “core”. The Red Sox managed 85.2 “Core Wins” on aggregate and locked themselves in the long-term, with two-thirds of their roster “controllable” for 3 years or more (as opposed to 50% from the previous season).
2012, 69-93, 5th place AL East, missed playoffs
The beginning of the season marked the arrival of the Bobby Valentine circus in Beantown. And what a circus it was. Conflicts with players such as Kevin Youkilis, Dustin Pedroia, Adrian Gonzalez, and David Ortiz put Valentine in the doghouse before he could establish any credibility. Gonzalez, Beckett, Crawford, and Punto were mercifully traded to the Dodgers in the “gift-that-keeps-on-giving” for GM Ben Cherington. Let’s take a look at how the Sox performed using the model of our “Core Wins” formula:
Notes: Now, only three players comprise the Sox “core” (and one was traded in August). Jon Lester fell out of the “core” again even though he posted similar (slightly lower) numbers to those in 2010, which is a result of the formula rewarding players who produce at a young age. That is to say, if a player’s performance does not improve regularly in the seasons before his age-27 season, it may cause the team to reevaluate if he truly is in that team’s “core”. The Daniel Bard experiment was truly a disaster, causing him to go from a “fringe core” player to next-to-nothing in value. Carl Crawford was simply awful again, and never entered the Red Sox “core”. Ellsbury did not perform due to injury, and (in what was likely the only bright spot of the season) Doubront entered the “core” as a productive young starter. Also, Will Middlebrooks was not included in the calculation because he was not under contract at the beginning of the 2012 season. He would have emerged with 5.6 “Core Wins”. In any case, the Sox aggregated 43.7 “Core Wins” according to the formula. A disaster indeed.
2013, 74-53, 1st place AL East, season 78% complete
Right now, on August 20th, the Red Sox have outperformed the wildest expectations of fans (and likely management) for the 2013 season. The payroll was cut by $21 million from 2012 and many regarded the season as a “gap year” as the plethora of young talent the Sox possessed continued to develop. While we have previously set out “Core Player” designation at 5 “Core Wins”, I will pro-rate this to 3.9 to account for that the Sox season is only 78% complete. Let’s take a look at how the squad put together by GM Ben Cherington and managed by John Farrell have fared:
Notes: Clay Buchholz has performed to a “Core Player” level, even though he has not pitched since June 8th due to injury. That speaks to how spectacular he truly was through his first bunch of starts in the 2013 season. Jon Lester has once again found the “core” elusive and Will Middlebrooks, who was thought to be a sure-thing, had his wheels fall off due to some poor plate discipline. The “surprise-of-the-year” award goes to Shane Victorino who has made Cherington’s signing (which was given heat at the beginning) look like a gem.
Interesting notes about the process:
I learned a whole lot by analyzing the composition of different Sox teams in this way. First off, David Ortiz never appears in the Red Sox “core”, even though a poll of Sox fans, writers, and executives would probably say he is a member. We need a model that accounts for this fact (the problem for Ortiz is that he never signed a long-term contract, and never sees his “control index” offset the reality that he’s an older ballplayer). I’ll raise this point when I discuss the limitations of the model.
Also, we learn that Pedroia is the true definition of a “Core Player”. He is a mainstay in the Sox “core” from 2010-2013, the only player not to drop out of the “core” at least once in the four-year span. The 6-year, $40.5 million contract given to him by Theo Epstein looks like an incredible bargain, and solidified his place in the “core” for the last couple of years.
We can also see that Felix Doubront is really valuable, and only has room for improvement. He has improved during his two seasons in Boston, and his productivity coupled with his young age and controllability make him an incredibly valuable piece for the Sox.
Finally, the Sox “Core” is most valuable (according to the formula) in 2010. This is because they had so many young, home-grown prospects under team control. It will be very interesting to re-visit this formula next year once many of the Sox heralded new prospects (Bogaerts, Bradley Jr., Ranaudo, Webster, Barnes, Owens, etc.) finally break into the bigs. If they perform in their first couple of years, their “Core Win” value will sky-rocket.
Limitations of the Model
In my opinon, the “Core Wins” model by Jonathan Judge does not adequately punish teams for signing outlandish contracts (a case like Carl Crawford, John Lackey, or Adrian Gonzalez). A future model that I would like to develop would incorporate these poor signings into the model in order to judge a teams core. One way to do this might be to weight the “Age Index” more heavily, perhaps by squaring it. Since the index is centered at 1, squaring would reward contributions by younger players and discount contributions by older players. Another method might be to incorporate “guaranteed money” into the calculation. It is contracts like these that weigh down a team’s ability to develop their own players, which in turn adds to their “core”.
The formula also needs a tool that rewards players for prolonged production, event though they haven’t signed large long-term contracts. In 2013, Shane Victorino is valued higher than David Ortiz according to the “Core Wins” formula, but many would surely argue that conclusion. Perhaps including a “time spent with same team” factor could give some insight into this type of player’s value.
While the model certainly highlights the contributes of young, team-controlled players, it has no real predictive value on the macro-level (i.e. the number of aggregate “Core Wins” is basically meaningless). It does, however, function to rationalize the contracts given to Pedroia, Lester, and Buchholz which bought out a couple arbitration years, but also ultimately delayed free agency. As the “Core Wins” analysis shows, these extensions increased their value to the Sox.
In the future, I will likely use this method to determine a player’s trade value or candidacy for a contract extension, as I think the application for “Core Wins” is more useful in that realm than in the realm of aggregate team performance.
In other news…go get ’em, Xander!
*Adrian Gonzalez technically signed this contract in the offseason following the 2011 season, as to avoid putting the Red Sox over the luxury tax threshold in the short term. However, I factored five years of control in due to the handshake agreement the Sox made with Gonzalez’ representatives on a contract extension following his trade to Boston.
**Pedroia’s “Control Index” is set to 2.5 due to the eight-year, $110 million contract extension he received at the All-Star break.