Better late than never graphs

As we pulled together this year’s Hardball Times Annual, I left something out on purpose. In every previous Annual, we had included league graphs, the ones I created on Baseball Graphs for all previous years. We had decided to pull back on the book’s stats and graphs, so they were left on the drawing room floor in 2012.

But I’ve always posted them on the Hardball Times site—I used to update them once a week—and I should have at least posted them here. Now, over two months later, I’m rectifying that mistake.

Today, we’ve got three basic team graphs for the American League, illustrating the season that was. I know this is old news, but it’s good to look back now and then, right? Tomorrow, I’ll post the National.

First, each team’s relative standing in runs scored and allowed.

image

There was no .600 team in the AL in 2012. The Rays and Yankees came closest, based on their run differentials, but the Rays lagged five actual wins behind their run differential. The Orioles were the surprise, finishing eleven games over their run differential record. If you believe that run differential is a truer indication of team production, then you might also notice the Indians, whose sub-.400 run differential record was masked by a won/loss record that was favored by the baseball gods by five games. I doubt that Indians management was fooled.

No team besides the Orioles and Indians had a positive variance of more than two games.

In terms of runs scored per game, the Rangers and Yankees were the class of the league (special footnote: these stats aren’t adjusted for park) while Seattle was at the bottom of the pack (left on the graph) by a goodly margin. The next graph picks apart each team’s offensive strengths into getting on base (also known as not making outs) and hitting for power:

image

Yah, the Rangers and Yankees are in the upper right and the Mariners in the lower left. When you lead the league in offense, chances are good you did both things well—and vice versa. The interesting stories are those in the middle. Two groups stand out: those who could have slugged more (Twins, Indians, Royals) and those who might have gotten on base more (White Sox, Rays, Orioles, A’s—the A’s lacked OBP. Oof.). Parks are making an impact here too, such as those in Chicago (slugger’s park) and Kansas City (not).

Our last graph picks apart the pitching and fielding of each team.

image

I used simple DER (Defense Efficiency Ratio—the percent of batted balls fielded for outs) for the left-hand axis of the graph, in an attempt to make a clear measure of fielding. The THT Annual has a better measure, courtesy of Carson Cistulli, in which the DER is measured for the ballpark’s impact. (There it is again. Ballpark.) But by most measures, the Angels belong near the top of this graph, as their team fielding was excellent, and the Royals belong near the bottom despite their cavernous home grounds.

Perhaps the Tigers stand out the most, by virtue of their superb pitching (second-best, as measured by strikeouts, walks and home runs allowed) but substandard fielding (second-worst DER, even adjusted for ballpark). In the end, Detroit had what it takes to be the American League champs. Graph or not.


Dave Studeman was called a "national treasure" by Rob Neyer. Seriously. Follow his sporadic tweets @dastudes.
3 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Detroit Michael
11 years ago

Thanks.  I always learn something new from looking at these graphs.

BirdWatcher
11 years ago

Isn’t DER basically the equivalent of 1-BABIP? Eyeballing the AL DER numbers, it looks like the league averaged about .315 for 2012, yet AL BABIP in 2012 was .296. Let’s take a specific example. The chart shows a DER of around .66 for KC, so that would give a pitchers’ BABIP of around .340 for 2012, but the actual number was .315. What am I missing here ?

studes
11 years ago

Good question. DER and BABIP are calculated differently. Our formula for DER is given here:

http://www.hardballtimes.com/main/statpages/glossary/#der

It’s (TBF-H-K-BB-HBP-E)/(TBF-HR-K-BB-HBP)

One issue here is that I don’t use errors on batted balls only, cause I’m using total team errors (which is admittedly a problem.  About half of all errors are typically “reaching on base” errors, though I’ve found that the difference doesn’t really vary much between teams.)

I don’t know how Fangraphs (if that is the site you’re quoting) calculates BABIP, but our definition is:

(H-HR)/(AB-K-HR+SF)

You can find our definition here:

http://www.hardballtimes.com/main/statpages/glossary/#babip

According to my calculations, the ML average DER was around .683 and the ML average BABIP was around .297 (our definitions). BABIP gives fielders more credit than DER does.

One of the differences is the inclusion of sacrifice bunts in DER. The other difference is that DER includes batters reaching base on errors as an unfielded ball. BABIP leaves errors out of the equation altogether.

That’s assuming that your source defines BABIP the same way we do.