The Ideal Roster Construction

The Cardinals have won more games than any other franchise this decade. Whether you credit GM John Mozeliak or a variety of other factors–including the Cardinals scouting department and minor league player development–the point is clear: the Cardinals have bypassed the "trust the process" phase teams like the Cubs and Astros, who have cashed in 100-loss seasons for top draft picks like tickets at an arcade. Rather than save up 1000 tickets for that teddy bear at the front counter, the Cubs and Astros gutted their rosters for years in exchange for first round draft picks including Carlos Correa, George Springer, Kyle Schwarber, Javier Baez, and 2016 NL MVP Kris Bryant.

Whether you like the concept of tanking or not, it has certainly led to success on the diamond. The Cubs and Astros, led by revered executives like Theo Epstein and former Cardinals scouting director Jeff Lunhow, respectively, are two teams who have reaped the rewards of tanking and paved a road clubs like the White Sox and Reds are currently following.

That said, Chicago and Houston were "forced" to tank because the cores of those teams in the early-mid 2000s flamed out or moved on. The dangerously heavy concentration on this group of elite players left a black hole on their rosters that could only be recouped through years of top draft picks.

The Mozeliak-era Cardinals are a different story. Mozeliak is often ridiculed for not signing [insert big name free agent] or not handing over a massive prospect haul for [insert marquee player at the trade deadline]. Rather, Mozeliak has built up the Cardinals with a more lateral approach. Headline names like Albert Pujols and Matt Holliday walked while each deadline and offseason the Cardinals land mid-level players to supplement the current roster. The question that was on my mind when I sat down to write the article was "is the Cardinals success stemming from its roster construction an exception to the rule?"

I gathered WAR leaderboard (courtesy of FanGraphs) data from every MLB team since 2012–a study of 150 teams–to see whether teams focused on value at the top of their roster performed better than squads of good-but-not-great players.

Before I jump into my findings, it's time for a quick math lesson. The coefficient of determination (referred to as r^2) is a statistic that measures how good of a fit your regression model is. Look at it this way: an r^2 value of 1 indicates that the independent variable completely explains the dependent variable data while an r^2 value of 0 indicates the exact opposite. Class dismissed.

I looked at each of the 150 teams over the five previous seasons (not counting 2017, of course) and classified the players by their WAR ranking on that team. There were six different classes:

  • Top player (WAR leader for that team)
  • 1-3 (three highest WARs on that team)
  • 4-6
  • 7-9
  • 10-12
  • 13+

The next step was to record the total team WAR for each club and determine what percentage of that total came from each class. From there I compared the 900 data points to their respective team's actual win-loss percentage and pythagorean win-loss percentage (W-L percentage based on the team's run differential). This is where the aforementioned r^2 values come in.

Class r^2 value (actual W-L) r^2 value (pythagorean W-L)
Top Player 0.185 0.165
1-3 0.24 0.227
4-6 0.217 0.199
7-9 0.16 0.136
10-12 0.15 0.15
13+ 0.247 0.23

This shows us that the proportional value of the worst players and top three players most directly influenced that team's success, with the 4th through 6th most valuable players in at a fairly close third. However, r^2 values just tell us how much of a correlation exists, not whether more value at the top of your roster is better or not. This next table made the answer pretty apparent.

Top Player% 1-3% 4-6% 7-9% 10-12% 13+%
St. Louis Cardinals 13.31% 34.85% 23.58% 17.65% 13.30% 10.62%
Playoff Teams 14.86% 35.73% 23.09% 16.88% 11.91% 12.39%
Non-playoff Teams 20.65% 47.84% 26.92% 18.65% 13.31% -6.72%
All MLB Teams 18.72% 43.81% 25.64% 18.06% 12.84% -0.35%
Playoff vs. Non-playoff difference -5.79% -12.11% -3.83% -1.77% -1.40% 19.11%

Indeed, top-heavy teams evidently performed worse than those with more evenly distributed value. Of the 20 teams since 2012 with the most value invested in their top three players, all of them finished with no more than 79 wins. For comparison, the top 20 teams in WAR coming from their 13th best player and below all won at least 86 games.

The Cardinals have never had more than 16.87% (75th of the 150 teams) of their team WAR coming from its top player since Mike Matheny became manager. In fact, the 2016 Cardinals had the lowest top player WAR% and top three WAR% of any team in the study.

Before I leave, I want to slip in what might be my favorite factoid from this entire study. Take a look at the WAR breakdown for the 2013 Astros.

Total WAR Top Player WAR 1-3 WAR 4-6 WAR 7-9 WAR 10-12 WAR 13+ WAR
3 4.4 7.6 3.3 2.1 1.4 -11.4

That's right. Jason Castro was worth more than the entire team.

Thank you for reading. As always you can follow me on Twitter @Tyler_Opinion

Go Cards!