Happy Labor Day! Regularly Scheduled Articles Will Resume on Tuesday, September 2.
April 10, 2014
Projected Roster Core Strengths for 2014
Most of our writers didn't enter the world sporting an @baseballprospectus.com address; with a few exceptions, they started out somewhere else. In an effort to up your reading pleasure while tipping our caps to some of the most illuminating work being done elsewhere on the internet, we'll be yielding the stage once a week to the best and brightest baseball writers, researchers, and thinkers from outside of the BP umbrella. If you'd like to nominate a guest contributor (including yourself), please drop us a line.
Jonathan Judge has a degree in piano performance but is now a product liability lawyer. He also blogs about the Brewers, and sometimes other teams, at Disciples of Uecker. Follow him on Twitter @bachlaw.
After last season, I rated major league teams by the “core wins” they accumulated: wins delivered by players most likely to constitute a productive core. We defined core contributors primarily by the years of team control remaining, and secondarily by placing each player on a linear curve peaking at a hypothetical peak production age of 27. By weighting each player’s contributions (as measured by WARP) for (1) years of remaining team control and (2) the player’s baseball age, and then totaling the results for each team, we ranked the teams by the strength of their core contributions. The results looked like this:
Although it’s useful to see how a team’s core developed over the course of a season, it’s also interesting to see whether things changed at all by the following Opening Day. A post-season assessment looks at how things turned out, but the first-day assessment tells us about the team’s outlook, particularly when we look at the players who are given starting roles. When these frontline players are projected not only to be successful, but also under extended control, it suggests a team with a stable core. When those players are under short-term control, or flotsam, or both, it’s fair to conclude that team will be looking elsewhere for new recruits in the near future, and not necessarily by choice. We’ll see both extremes in this study.
First, a quick refresher on the method, as adapted to take a forward-looking view of team performance:
To compare teams to one another, I weight equally each team’s rank on three different aspects of core wins: (1) the number of overall core players on a team; (2) the number of core players who are pitchers; and (3) the overall (hopefully positive) differential between the “core” wins total for the team and their underlying wins above replacement. Only positive WARP contributions were considered. The importance of the first two categories is hopefully self-explanatory, and also has corresponded with much of the Rays’ success in recent years; the importance of the core win differential reflects the core strength of the team’s starters as a whole, even if they can’t all be frontline stars.
Applying that method, here is the ranking of each team’s starters by “core” value heading into 2014:
Some of these ranks will come as no surprise; others may raise an eyebrow.
The Dodgers tied for first by virtue of their long-term contracts for top performers; the Cardinals did it the other way, by having many of their top performers under team control courtesy of baseball’s collective bargaining agreement. Were it not for the recent injuries to their young rotation, the Braves may have topped both of them. The core value of the starters for these three teams truly stands apart from the rest of the league.
At the other end of the spectrum, the Twins and Astros rate lowest because (1) their current starters are generally terrible, and (2) they are counting on star-powered reinforcements from their excellent farm systems over the next few years. For those reasons, very few of their current starters are part of these teams’ long-term plans, and it is evident that those clubs really don’t care about it.
Among the surprises, the Brewers’ relatively high core ranking had me double-checking my math. But it makes sense: The Brewers’ primary challenge isn’t a lack of decent projected starters; it’s a lack of depth, particularly in their farm system. Thus, core wins reveals that the Brewers, if they stay healthy, could be somewhat successful in the short term. They’ll need to be, since farm system reinforcements in the near future will be few and far between.
Toward the bottom, the low rankings of the Royals and Padres may disappoint fans of two teams frequently touted as burgeoning contenders. But the Padres have been wracked by injuries (again), and PECOTA thinks very little of their current rotation. Similarly, the Royals have a roster that is exposed by a core wins analysis. Some of their best players, like James Shields, are on their way out the door, and too many of their best remaining players (Alex Gordon, Billy Butler) will be free agents after next year. Unless the Royals intend to call up an entirely new team—they do have a strong farm system—or find $50 million under the couch cushions, this is a problem. Their current roster core is disintegrating.
How do these rankings compare with the results from last year? Here are the teams ranked again, this time by their projected improvement in overall ranking from last season. Again, we are looking only at starters this time, as compared to whole rosters last time, but the results nonetheless are informative:
Changes in a team’s ranking are interesting for a few reasons. First, the way the formula works, players are held to a higher standard as they get older and as their amount of team control diminishes. Players must improve their performances to achieve the same ranking as each previous year, because their Index weights go down and their WARP must go up to compensate. Teams that simply let their existing stars get a year older will tend to go down in the rankings. Second, the formula measures the steps each team took in the offseason, and whether PECOTA sees those moves as actually improving their core, or just as window-dressing.
It’s important to note that for some teams, the absence of movement is a good thing, because they were already in such good shape there was virtually nowhere to go but down. I found that the Cardinals had the best roster core in baseball last year, and being projected to hold that ranking this year yields a difference of “0”—for them, that’s just fine.
The Yankees receive the most positive measured improvement. They’re the antithesis of the Cardinals: Last year, thanks to injuries and advanced age, they were the only roster that managed to have a core wins differential that was negative. (Before I wrote last year’s article, I had not even considered that to be possible.) The Yankees have “addressed” the problem in the way they address most problems: by opening their checkbook and signing players like Masahiro Tanaka and Jacoby Ellsbury to long-term contracts, and extending Brett Gardner. This helped address the Yankees’ control problem, if not really their age problem, but should kick the can down the road for a year or two in inimitable Yankees fashion.
On the opposite end, from a core wins perspective, Cleveland’s offseason was a disaster. Significant contributors from last year, like Scott Kazmir and Ubaldo Jimenez, have departed, and PECOTA believes that the Indians failed to replace them. Likewise, the Indians failed to extend Justin Masterson in spring training, and merely extending Yan Gomes and Jason Kipnis (who was already under long-term control) did not stem the tide. Combined with what PECOTA sees as a substantial regression from a 2013 over-performance, the Indians appear to face an uphill battle over the next few years as they try to wring performance from existing major league assets.
Also interesting are the Red Sox, whose roster core was near the bottom in my ratings at the end of last year and improved by only two slots this year. Last season, the Red Sox benefited from extraordinary production out of several short-term free agent assets, and their current roster continues to count largely on a roster of older players who should generally be in decline. Of course, that doesn’t mean that the Red Sox are going to have a terrible year. What it more likely means is that the Red Sox, who possess a top-notch farm system and enormous financial resources, are content to ride out their current contributors, knowing they’ll have more options than most when the music stops. However, fans should not be surprised if that transition ends up being abrupt than they would prefer.
Although the strength of a team’s core is an ongoing concern for any club, the first relevant milestone during this season will be the trade deadline.
The Orioles, for example, went all-in with their free agent signings this year for a reason: Their core is depleting rapidly as Matt Wieters and Chris Davis approach free agency, and they have little in the pipeline to replace them. If this go-for-broke season does not go as planned for the Orioles, they would be well-advised to sell their departing assets to try to replenish their core. By contrast, a team with a highly-ranked core (such as the Cardinals or Giants) probably has more freedom to trade farm system assets to put themselves over the top, because it is less likely that they will need reinforcements in the immediate future. As the trade deadline approaches, perhaps we’ll review not only how PECOTA’s relevant predictions fared, but also how each club’s core roster strength should inform their deadline decision-making.
 These charts have probably changed somewhat by the time you read this article. Last season’s study included all players for each club because I knew everyone’s actual playing time. Because playing time estimates for reserves are somewhat speculative, and because those players are rarely expected to make substantial contributions, we’re just focusing on starters here.
 Jeff Zimmerman is challenging our assumptions of when a current player actually peaks. His recent research shows that the peak baseball age, for both hitters and pitchers, may well be closer to 25. To allow an apples-to-apples comparison with my 2013 post-season study, which predated Jeff’s article, I’m going to keep the peak age at 27 for the purposes of this piece.