A Guide to Advanced Statistics
There is an information gap when it comes to advanced stats now. Columnists who rant against them seem to only know of the existence of wins above replacement. Meanwhile, many enthusiasts just use them casually without pausing to explain the stat each time. What is a fan who is interested in these stats, but has no knowledge of them to do? I now present to you my guide to the best introductory advanced stats and their uses.
The three stats that every baseball fan will know are batting average, on-base percentage and slugging percentage. These are the stats that make up the ubiquitous triple slash and the last two can be combined into another stat, OPS. OPS is a basic read of how well a player gets on-base and the extra value they provide with extra base hits.
However, there is a flaw with slugging percentage and therefore OPS. In calculating slugging percentage doubles are multiplied by two, triples by three and home runs by four. In other words, a double is worth twice what a single is. While doubles are good, any baseball fan would tell you that it is not worth twice as much as a single.
When I was at Nationals Blogger Day in June one of the people we were able to meet with was their Baseball Operations Analyst Samuel Mondry-Cohen or in other words the Nats’ stat guru. When asked what he thought was the best stat to get a holistic view of a player on offense he answered wOBA and I have to agree. The reason is that wOBA measures the same thing OPS purports to measure, but does so in a logical manner. wOBA stands for Weighted On-Base Average. It adds unintentional walks, hit by pitches, singles, doubles, triples and home runs with weights and divides those numbers by at-bats, walks minus intentional walks, sac flies and hit by pitches. Instead of using the weights that slugging percentage does, wOBA uses linear weights which change year-to-year based on how good or bad offense is league wide. Here’s a handy guide to determine how good or bad an individual player or team’s wOBA is:
An easier stat to understand quickly is wRC+. wRC+ stands for Weighted Runs Created and simply shows how much more or less a player contributes to his team scoring runs, relative to league average and adjusted for park and league. A league average hitter will have a wRC+ of 100 and any point above or below 100 is a percentage point above or below league average. So a player with a wRC+ of 120 created 20% more runs than league average. wRC+ is based off of wOBA, so you know the information you’re getting is the most accurate available. But enough about offense, there are two other main parts of baseball, pitching and defense. Let’s start with pitching.
Most big picture pitching stats have a fundamental problem, they don’t separate what the pitcher does from how the team around him affects it. A good pitcher with a bad offense could end up with more losses than a bad pitcher with a good team around him. ERA suffers from the same problem, an error prone defense could suppress a player’s ERA because of the large number of unearned runs.
To combat this uncertainty we have a statistic called FIP or Fielding Independent Pitching. FIP directly measures what a pitcher can control: strikeouts, walks, hit by pitches and home runs. FIP is then adjusted by a constant so it is on the same scale as ERA to aid in comprehension. A warning though, FIP is not as accurate with smaller samples, so this is definitely a season statistic rather than a single game one.
The main defensive stats, errors and fielding percentage, suffer from the same problem, they don’t take into account how a player got to the ball. Player A could field every ball hit right at him, but only those hit right at him, while player B could field 90% of balls within 10 yards of him and fielding percentage would say player A is better. That seems unfair.
Enter the main advanced defensive stats UZR and DRS. Both UZR, or Ultimate Zone Rating, and DRS, or Defensive Runs Saved, use data provided by Baseball Info Solutions to rate how well a player plays defense in terms of runs.
UZR is calculated by creating run values versus league average for many aspects of fielding including: Outfield Arm Runs (ARM), Double-Play Runs (DPR), Range Runs (RngR) and Error Runs (ErrR). These run values are then combined into one overall defensive score, UZR.
DRS is a bit simpler in comparison, it simply measures how many more or fewer successful defensive plays a defender makes versus league average. For example, if a third baseman makes a play only 27% of third baseman make, then they are credited with .73 of a point. On the other hand, if they mess up a play 90% of third baseman make they would lose .9 of a point. Each play is assigned this value and their respective point values are added up to create DRS.
There are a couple notes about UZR and DRS. First, there are significantly less data points on defense than on offense, so it is suggested to look at three years of either stat for an adequate sample size. Second, Baseball Info Solutions data is collected by human scorers, so it is subject to human error. However, since FIELDF/x data is unavailable to the public, BIS data is the most accurate available.
This should hopefully give you a better idea of what advanced stats are and which to follow and use to evaluate players. If you have any questions you can leave a comment or send me a note on Twitter at @nextyeardc, I’m always ready to talk about advanced statistics.