Components And The War On ERA

The biggest problem some people have with the new wave of “advanced” statistics in baseball seems to be a detachment from the realities on the field. Guy gets a hit with a runner in scoring position, guy gets an RBI. Simple, right?

Over time and after some heady thinking by some very clever people, it became clear that a lot of our tried and true measures didn’t provide enough context. Who created the run, the guy who got the hit with two men on base or the two guys that worked their way on base in the first place?

Pitchers on good offensive teams receive rich contracts after posting gaudy win totals while talented players on weaker teams fall victim to not only their team’s offensive failings but defensive too. What more can a man do?

Enter Fielding Independent Pitching or FIP. Based on a few simple relationships between that which a pitcher can control and that which is out of his control, FIP attempts to level the playing field and value pitcher’s for their skills and skills alone.

More advanced pitching metrics like xFIP, SIERA, and tERA incorporate batted ball data to their formulas, complicating matters but providing additional insight into what makes a good pitcher. Information on batted balls is still slightly hazy as not all line drives are created equal and the amount of control a pitcher has on batted balls is something of a sore spot among the deepest thinkers.

For regular old dummies like us, we can keep it pretty simple. Strikeouts are key, walks are bad, home runs are worse. Ground balls are good because they tend not to turn into home runs, no matter how small you build your ballpark.

If you’re skeptical, I understand. Attempting to shove all the weird and wild varieties of pitcher into one little box seems unfair and judging a pitcher without using, you know, his actual results feels like cheating. Again: not all runs are created equal.

Consider the below graphic pioneered by Rich Lederer at Baseball Analysts, which plots the strikeout differential (strikeouts minus walks, divided by total batters faced) against the per batter faced-home run of all qualifying starting pitchers in 2010. Why strikeout differential? Because a pitcher who strikes out 12 and walks 3 is a lot more valuable than a pitcher who strikes out 4 and walks 1 over 9 innings, that’s why. Strikeouts are the great equalizer, no assistance required.

You can see the graph divided into four sections, each section representing an echelon of pitching talent. Some important names and/or outliers are highlighted, to help provide some valuable context to the image. If you look in the bottom right corner, you see the best of the best. The cream of the crop. Pitchers who limit walks, rack up strikeouts, and keep from allowing too many home runs. There is an excellent chance the names of both Cy Young award winners are in that valuable quadrant (Yes, both CC Sabathia and Felix Hernandez are mixed in there too.)

The upper right hand section features many talented pitchers who struggled to keep the ball in the park in 2010. Here is where we look to our batted ball data for more context. Whatever the reason behind the elevated home run numbers, those on the furthest right side of this quadrant are pitchers you can absolutely work with at the top of your rotation.

As soon as we go to the left of the average strikeout to walk differential, we run into problems. Unsurprisingly, many of the names we see here are names of bad or awful pitchers. Incapable of getting any outs themselves, these pitchers are at the mercy of their defense. Many in the lower left corner — like Aaron Cook and Mike Pelfrey — are ground ball machines who make it easy on themselves. A good, heavy sinker is a decent alternative for those who lack the big out pitch for racking up strikeouts.

As for the upper left, I caution you to stay away. Scott Kazmir and Ryan Rowland-Smith (the Hypen) are examples of how one should not conduct their business. Too many walks for guys like Kazmir and the Ghost of Rich Harden, too many home runs and not enough strikeouts for RRS. Harden and Kazmir are excellent examples of what happens when the strikeouts end: the happiness dies right alongside.

There is one name of significance omitted from this graph, mostly for aesthetic purposes. One name who, in a 12 start sample, put up numbers that were quite literally off the charts. Stephen Strasburg racked up a strikeout differential of 27.4%, more than 8 percentage points above the next highest player. His home run allowed rate was also well below Cliff Lee, Roy Halladay, and Tim Lincecum. In other words, he’s really good. It doesn’t take any mathematic gymnastics to prove it.

Comments (2)

  1. Great example today on usefulness of FIP from Aaron Gleeman at HardBallTalk:

    http://hardballtalk.nbcsports.com/2010/11/12/look-past-era-and-chad-qualls-can-be-a-free-agent-bargain/

Leave a Reply

Your email address will not be published. Required fields are marked *