The first in a series of posts I can write when the sample sizes don’t give me anything else interesting to write.
Increasingly some of the Advanced Metrics are becoming a lot more popular and are getting cited in more mainstream sources (IE not baseball nerd sites like Fangraphs). So I figured it’s probably worth explaining to those of you who read this and have some questions about what the deuce it is that I’m talking about.
What Does it Mean for a Stat to be Normalized
In this case, it means that it’s accounting for things that can cause someone numbers to look better than they actually are. In practical terms it works like this; there are places in baseball where it is just a lot easier to hit a home run due to either tiny outfields (Yankee Stadium) or because of altitude (Coors Field in Colorado). Conversely, there are some places like Marlins Park and and Petco Park that suppress offense because of deep outfields and low elevation. This means that a hitter who plays in Yankees park multiple times a year might have more impressive offensive stats than a player in Miami. In order to handle the unfairness that this can cause in comparing raw performance, stats like OPS+(quality as a hitter) and ERA+ (quality as a pitcher) adjusts for this issue by taking into account where players played. This means that 100 is average for that year with those factors and anything above or below is how much better or worse than average a player was in that given season.
Here’s an example from two outfielders in 2017:
Howie Kendrick (Philadelphia/Washington): .315 Batting Average, 9 Home Runs
Gerardo Parra (Colorado): .309 Batting Average, 10 Home Runs
Based on comparing these numbers in a vacuum, you might be forgiven for thinking that they were pretty similar players. The difference however is that Parra played the entire year with the Rockies and mostly hit in notoriously hitter friendly parks while Kendrick split the season between Washington and Philadelphia and generally played in less offense friendly environments(Kendrick also got on base more, but that’s a different post). When we calculate the OPS+ for both pitchers it comes out to
Howie Kendrick: 119 OPS+
Gerardo Parra: 94 OPS+
This translates to “Howie Kendrick was 19 percent better than an average hitter and Parra was 6 percent worse than an average hitter”. Effectively, even though the two had similar numbers, Kendrick was 25% better when batting than Para.
Jhoulys Chacin (Padres) 13-10, 3.89 ERA
Jordan Montgomery (Yankees) : 9-7, 3,88 ERA
On the surface these two pitchers are identical once you discount the wins stat. However, in this case Chacin pitched in the very pitcher friendly Petco Park, while Montgomery was a regular in Yankee stadium, who’s very narrow RF wall can destroy pitching stats. Using ERA+, you instead get that
Jhoulys Chacin: 107 ERA+
Jordan Montgomery: 116 ERA+
While they’re pretty comparable as far as results, Montgomery being 9 percent better would seem to make him the clear winner and the one you would probably pick between the two.
Too Long, Didn’t Read, What’s the message?
If a stat has a + after it (Like ERA+ or OPS+) that means that 100 is average, and any other number is how much better or worse they are than average, this matters.