Sabermetrics often get a bad rap among baseball fans and the media. The population at large seems to think that sabermetricians are a bunch of nerds who like to obsess about a series of alphabet soup-like stats.
In a conversation with Bob Nightengale of USA Today, Brandon Phillips went on a lengthy rant against on-base percentage. Some of the highlights of this rant include gems like, “I feel like these stats and all of these geeks upstairs, they’re messing up baseball, they’re just changing the game” and my personal favorite, “If we all just took our walks, nobody would be scoring runs. Nobody would be driving anybody in or getting anybody over.”
While I understand Phillips’ sentiment that there are many more ways to measure a player’s value than OBP, he feels that people in front offices, the media and fans use OBP as too heavy a factor when assessing a player’s value. “If you don’t get on base, then you suck,” as Phillips so succinctly put it. No single stat is so valuable that it is able to determine a player’s value and stats are not exclusively the only way to evaluate a player’s value or ability. All kinds of stats (both the old fashioned counting ones and the advanced metrics) as well as scouting and the tried and true “eye test” have a place in evaluating players in baseball in 2015. Advanced analytics or, as I like to call it, holistic player analysis is the only truly effective way to develop players, make personnel decisions, and think about baseball in this current climate.
There is a lot of criticism of the wins above replacement (WAR) stat coming from across the baseball industry. Current players, former players, broadcasters, and writers all have their own objections to what they characterize as an artificial stat. WAR, WARP, VORP or whatever you chose to call it aims to reduce a player’s contributions to his team to one number. In the case of position players, it takes how much you play your defensive position, base running efficiency, and overall production at all three phases of the game and tries to compare a player’s production to a notional replacement player.
Theoretically, a Major League roster with 25 replacement players would have a .294 winning percentage, or a record of 48-114. In theory, 30 teams of replacement level players would win about 1430 games, or 48 multiplied 30. Since there are 2430 Major League games played every year it means that players better than replacement level could win as many as 1000 more games than the replacement level.
WAR also values replacement level. It is important for a player to take his 600 plate appearances because every plate appearance taken by any Major League player is better than one taken by a replacement player.
Another important concept of is that off positional adjustment that breaks down as below according to Fangraphs:
Catcher: +12.5 runs (all are per 162 defensive games)
First Base: -12.5 runs
Second Base: +2.5 runs
Third Base: +2.5 runs
Shortstop: +7.5 runs
Left Field: -7.5 runs
Center Field: +2.5 runs
Right Field: -7.5 runs
Designated Hitter: -17.5 runs
This correctly adjusts for the difficulty of each position to play, with catcher, shortstop and centerfield being the most valuable/difficult and first base and the corner outfield spots ranking as negative. Centerfield, third base, and second base are ranked in the middle. Designated hitters are punished for not actually playing the field.
The concept is simple. You take all of player’s contributions with the bat, the glove, and on the base paths, factor in his positional adjustment and his replacement level. As Dave Cameron puts it “the expected value of a replacement level player is about negative 20 runs per 600 PA. To phrase it a bit differently, if you lost a league average player and replaced him with a freely available guy, you’d lose about two wins.” So the more a Major League average player plays, the more valuable he becomes.
In short, WAR measures Major Leaguers against two imaginary benchmarks, replacement level as well as league average. If a player is a great player but consistently hurt, that is not good for his WAR as he has to be replaced. Conversely an average player who will take 600 PA especially at a defensively valuable position are worth more than can be appreciated.
Ultimately, the goal of advanced metrics is to empirically measure a player’s value against his peers. That is why stats like wRAA, UBR, and UZR are so valuable because they give us context and allow us to compare a player’s performance against a baseline. WAR is just the act of combining all of these stats based on league average to determine how much better than a random player he is.