Three years ago, I recall watching the World Series, a match-up between the Boston Red Sox and the Colorado Rockies, and discussing with a fellow Red Sox fan what he felt was poor play by shortstop Julio Lugo.
I remarked that despite his low batting average and on-base percentage, Lugo was a decent hitter because he had driven in 70 runs from the nine-spot in the lineup. Little did I know that I was dead wrong in my defense of Julio Lugo.
It turns out that Lugo was only worth 0.4 wins above replacement (WAR) that year; equaling $1.7 million out of the $8.3 million he actually was paid. I would not have known this, had I not learned sabermetrics since.
Influential baseball writer Bill James defined sabermetrics as “the search for objective knowledge about baseball.” In this objective study of the game, everything must be proven by numbers; there is no room for subjective criteria such as clubhouse presence and veteran leadership, just throw those facets of the game out the window. Here is an overview of some sabermetric evaluations.
Game State / Run Expectancy / Run Value: Baseball can be described as a collection of states like how many runners are on base and how many outs there are. For example, every half inning begins with the game -state of no outs and no runners on base.
Run expectancy is the average number of runs a team is expected to score in an inning, depending on game-states. At the game state of no outs and no runners, a team has a run expectancy of 0.555. Compare this to the game-state of no outs and bases loaded, a team has a run expectancy of 2.417.
By comparing the changes of run expectancy due to different events, we can calculate the run value of an event. Of course, the run values are based on averaged game-states.
A home run is worth 1.397 runs, while a single is worth 0.475 runs. Run values are assigned linear weights, with each event having a different weight.
Hitting: Most baseball fan are most exposed to the three Triple Crown statistics- batting average, home runs and RBIs. While there is nothing wrong with measuring home runs, there are holes to peck at for average and RBIs.
The RBI is not a good measure of hitting ability, as it relies too much on other players. Spot in the batting order will greatly vary RBIs, as the players in front must get on base in order for RBIs to be generated.
Batting average is overly rudimentary, as it only measures the percentage of hits a player receives, while eschewing walks and extra-base hits. The only baseball players that need a 0.300+ batting average to be competent hitters are slap hitters who refuse to draw walks.
On-base percentage, slugging percentage, and wOBA are good complements for average, as they help account for the holes in batting average.
wOBA, weighted on-base average, is the linear weights divided by plate appearances.
