The Daily Northwestern

Baseball: Advanced stats and what they tell us about Northwestern’s season

Alex Putterman, Sports Editor

Hang on for a minute...we're trying to find some more stories you might like.


Email This Story






Traditional baseball statistics like batting average and ERA have survived more than a century for a reason: They’re intuitive and instructive in evaluating players.

But in recent years, statisticians have improved upon those stats, creating “sabermetric” measures that provide more complete pictures of player performance.

Using basic numbers offered by the NU Sports and Big Ten websites, The Daily dusted off its Microsoft Excel skills and calculated advanced stats for the 2014 Northwestern baseball season.

Here are brief descriptions of the stats calculated, along with what they can tell us about the performances of individual Wildcats players this season.

We’ll begin with offensive numbers (all team ranks minimum 75 at-bats, unless otherwise noted):

OPS and OPS+

Most fans and analysts don’t quite consider OPS an advanced statistic, but it’s not listed on the official team or conference websites. So we’ll include it here.

OPS simply adds up on-base percentage and slugging percentage to create a number fairly descriptive of a batter’s production. While batting average counts a single the same as a home run and entirely ignores walks, OPS weighs those actions.

OPS+ presents OPS relative to league average (it also usually adjusts for park effects, but that is impractical to calculate here). An OPS at league average means an OPS+ of 100, an OPS 10 percent higher than average is a 110 OPS+, 10 percent lower than average is a 90 OPS+ and on and on.

Senior designated hitter Jack Livingston led NU in OPS (.879) and OPS+ (124), largely thanks to 11 extra-base hits in only 77 at-bats. Among players who saw everyday playing time, junior second baseman Scott Heelan paced the Cats with an .806 OPS and 114 OPS+.

Freshman Matt Hopfner comfortably led the team in batting average. But 59 of the outfielder’s 66 hits were singles, so he came in only third on the team in OPS and OPS+.

wOBA and wRC+

wOBA (weighted on-base average) is a more advanced version of OPS, weighing walks, hit by pitches, singles, doubles, triples and home runs according to their respective values to a lineup.

The weights fluctuate every year based on league production, and such data is obviously not available for the Big Ten. We’ll (somewhat arbitrarily) use the weights for Major League Baseball in 2010, as presented by FanGraphs.com.

wRC+ is to wOBA as OPS+ is to OPS. It puts wOBA against league average, with 100 representing the mean.

Again, by these formulas we see Livingston as the team’s best hitter (.388 wOBA, 122 wRC+) with Heelan second (.370, 117). Hopfner falls to fourth, narrowly behind junior utilityman Reid Hunter.

Freshman outfielder Joe Hoscheit looks good by measures that account for extra-base power, thanks to 10 doubles and a team-high four home runs, posting a 102 OPS+ and 105 wRC+ despite only a .258 batting average.

Junior outfielder Walker Moses also does well in advanced stats because he often walks and was hit by five pitches in 2014. Because wRC+ counts walks more heavily than OPS+, Moses’ wRC+ is 100, while his OPS+ is only 93.

On to the pitchers (all team ranks minimum 10 innings pitched unless otherwise noted):

ERA+

Just like OPS+, ERA+ puts individual ERAs against league average.

It was a rough year for NU pitching, and the numbers reflect that.

Other than Hoscheit, who posted a 438 ERA+ in only 10 1/3 innings, and sophomore Jake Stolley, whose 144 ERA+ came in 17 innings, only one Cats pitcher was even above league average in 2014.

That lone hurler was senior Dan Tyson, whose 3.47 ERA and 110 ERA+ were team bests among pitchers who threw at least 20 innings.

WHIP

WHIP (walks plus hits per innings pitched) measures how many base runners a pitcher allows. This year, every NU pitcher but Hoscheit averaged more than 1.25 per inning, and no one who threw more than 20 innings bested the Big Ten average of 1.40.

WHIP is especially useful in evaluating relievers, whose ERAs can be distorted by inherited runners. The Cats had bullpen struggles all season, and WHIP illustrates why: Even trusted closer senior Jack Quigley  allowed 1.61 base runners per inning.

Senior Nick Friar was NU’s best starting pitcher by WHIP, finishing at 1.48, while Tyson, despite his solid ERA, came in last on the starting staff at 1.67.

FIP

FIP (fielding independent pitching) relies on the controversial but research-backed assumption that over the long run, pitchers have no control over the result of a ball in play, only strikeouts, walks, hit by pitches and home runs.

While poor defenses (by missing tough plays that could have been fielded but are difficult enough not to count as errors) can hurt a pitcher’s ERA, they don’t affect his FIP, which estimates what a pitcher’s ERA would be with average defense behind him.

Almost every NU pitcher posted an FIP better than his ERA, suggesting the Cats’ defense was lacking. (This is backed up by traditional stats: NU led the Big Ten in errors.) Four of the team’s five starters had FIPs better than their ERAs in 2014, including sophomore Matt Portland, whose FIP was a full run better than his ERA. His 3.71 FIP was best among Cats pitchers who threw at least 20 innings, with Friar next at 3.96.

BABIP

Besides being the most fun acronym in the sabermetric vernacular, BABIP (batting average on balls in play) is valuable in assessing the degree to which a pitcher benefits from good luck.

As explained above, the results once a ball is put in play tend to even out over time, so if a pitcher allows an especially high BABIP, he’s probably unlucky — and vice versa.

It makes sense, then, that NU’s three worst pitchers this season by ERA — Livingston, senior Ethan Bramschreiber and freshman Joe Schindler — all allowed high averages on balls in play. This means all three were likely victimized by poor batted-ball luck: dribblers sneaking through the infield, line drives finding gaps instead of gloves, infield hits, etc.

In small sample sizes, Hopfner and Hoscheit both allowed low BABIPs, which helps explain the success they (especially Hoscheit) had on the mound late in the season.

BABIP for batters is more prone to variation, but numbers far from league average (.318 in the Big Ten this season) can raise eyebrows. Expect some regression next season from single-hitting Hopfner (.393 BABIP) and perhaps a hike in batting average from Hoscheit (.271 BABIP).

Email: asputt@u.northwestern.edu
Twitter: @AlexPutt02

Comments