Home > Sports > WAR Mini-Series Part 1: Defining “Replacement”

WAR Mini-Series Part 1: Defining “Replacement”

In baseball, the term “WAR” has become one of the new buzz words, with many journalists and analysts frequently referring to the term and its “value” to back up arguments they’re trying to make, especially when evaluating players or seasons. Because of this, I’ve decided to embark on a mini-series about the topic, attempting to define it as well as find flaws, inadequacies or reasons to hold off from making WAR gospel. (All numbers quoted will be as of August 8th, 2012.)

Let’s start with the basics, “WAR” stands for Wins Above Replacement. It’s an analysis to determine how much better a player performs than a “replacement” player, putting it in terms of wins added to a team. Baseball-Reference.com refers to a replacement player as a “AAA or AAAA player,” likely meaning minor league veteran. Sound a little subjective? Yeah, we thought so, too.

The term replacement player, in regards to WAR, brings a couple questions to mind. First off, how many wins does a replacement player create? Zero? One? Five? What about a team filled with replacement players? What would their record be? One would assume a team full of replacements would obviously be the worst team in the league, right? After all, any player with a positive WAR would be better than any player on that team. By that same logic, any team that had a WAR above zero should win that many more games than the replacements (for the sake of ease — and fun — we’ll call them the Keanu’s). However, that seems to be far from the case.

Using Baseball-Reference.com’s WAR calculations (FanGraphs runs their own version, as well), I compiled the team totals for Batter’s WAR, Pitcher’s WAR and Combined WAR. Not surprisingly, the Houston Astros were the lowest ranked team in WAR with a total score of just 3.9 (For those of you keeping score at home, Andrew McCutchen currently has a WAR of 6.1 by himself). Is we use the Astros as our pacesetter for Keanu’s, we’ll come to the conclusion that at this point in the season, the expansion team of Keanu Reeves would have roughly 32 wins and be on pace for a 46-116 record. This would be the second-worse season in MLB history for a 162 game season (The 2003 Detroit Tigers hold this dubious honor with a 43-119 mark. For the record, their team WAR was -0.9).

Using a 32-80 record through 112 games (Houston’s Keanu level), we calculate a winning percentage of 28.7%. Using this, we can convert every team’s Keanu level through 112 games. We can now see how close WAR comes to quantifying actual wins. By simply adding the total team WAR to the calculated Keanu level, we can figure out what their wins should be if WAR was a truly analytical tool, essentially breaking down and dividing each win among the players. If this were to happen, each team would be roughly within a win or two of their record, with the sum of the 30 teams being close to zero (end of the season marks should be zeroes across the borad). However, what we find is quite different. Teams ranged from being one win less than WAR establishes (no surprise the Yankees find themselves in this category) to WAR shortchanging teams by up to 12 wins (the Pirates). The table below shows the results.

Team W L Bat WAR Pit WAR Total WAR Keanu Wins Keanu Level Difference
HOU 36 76 2.9 1 3.9 32.1 0.287 36 0
CHC 43 66 7.5 2.2 9.7 31.2401786 0.287 41 2
MIA 51 61 2.9 7.8 10.7 32.1 0.287 43 8
COL 40 69 3.4 8.5 11.9 31.2401786 0.287 43 -3
CLE 51 60 13.9 -0.7 13.2 31.8133929 0.287 45 6
SDP 49 64 9.5 3.8 13.3 32.3866071 0.287 46 3
NYM 54 58 11.1 3.9 15 32.1 0.287 47 7
PHI 50 61 10 5.2 15.2 31.8133929 0.287 47 3
MIN 49 62 15.5 0.1 15.6 31.8133929 0.287 47 2
KCR 47 63 8.6 7.7 16.3 31.5267857 0.287 48 -1
MIL 51 59 12.6 4.8 17.4 31.5267857 0.287 49 2
SEA 51 62 10.7 7 17.7 32.3866071 0.287 50 1
BAL 60 51 4.2 13.7 17.9 31.8133929 0.287 50 10
LAD 60 52 6.8 11.9 18.7 32.1 0.287 51 9
PIT 63 47 12.1 7 19.1 31.5267857 0.287 51 12
TOR 53 58 16.7 3.4 20.1 31.8133929 0.287 52 1
BOS 55 57 16.4 4 20.4 32.1 0.287 53 3
TBR 59 52 10.3 11.7 22 31.8133929 0.287 54 5
DET 60 52 10.5 12.2 22.7 32.1 0.287 55 5
ARI 56 55 13.9 9.1 23 31.8133929 0.287 55 1
SFG 61 51 16.4 6.8 23.2 32.1 0.287 55 6
ATL 64 47 19.6 5.2 24.8 31.8133929 0.287 57 7
OAK 60 51 9.4 15.5 24.9 31.8133929 0.287 57 3
CHW 60 50 9.9 16.9 26.8 31.5267857 0.287 58 2
LAA 59 53 24.3 3.6 27.9 32.1 0.287 60 -1
CIN 66 45 10.1 17.9 28 31.8133929 0.287 60 6
TEX 65 45 14.2 14.4 28.6 31.5267857 0.287 60 5
WSN 68 43 12.2 16.8 29 31.8133929 0.287 61 7
STL 61 51 22.6 7.2 29.8 32.1 0.287 62 -1
NYY 65 46 17.4 16.4 33.8 31.8133929 0.287 66 -1

It’s easy to discover that WAR seems to undervalue teams that play higher than the sum of their parts (see: synergy). Pittsburgh and Baltimore lead the group of overachievers with 12 and 10 win discrepancies, respectively. An interesting find, however, is that the Miami Marlins apparently have overachieved by 8 wins so far this year. How does a team that’s 10 games under .500 in a season that included the debut of a new stadium (and monstrosity of a sculpture) and high hopes for a division championship get the title of overachiever? Probably due to the fact that the team Miami actually built never really had much of a chance to win. Their “big” winter acquisitions were vastly overrated and Ozzie Guillen may have finally lost his mojo, but that’s a different column for a different day.

Now lets backtrack a few steps. What should the definition of “replacement level” be over a full season? Using the current win percentages of each team, we extrapolated the current wins and the current WAR over a 162 game season, allowing us to calculate the amount of replacement wins (Wins – WAR). The average comes out to 52.5 wins, almost exactly what Houston is projected to win (52). This means that with a 25 man roster and the occasional September call-ups, each player on a replacement team is worth approximately two wins.  However, if you want a real-life answer, look no further than your Houston Astros, Major League Baseball’s one and only quadruple-A team.

Advertisements
  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: