Cleveland Sports Insiders

Ernie Camacho's favorite blog…

Cleveland’s ERA-FIP Gap: The Defense IS That Bad

1 Comment

The ‘Thesis, Antithesis, Synthesis’ process is regarded as a central tenet of Western Dialectic; namely, one begins with a thesis, an original assertion, which is refuted by an an antithesis, a rebuttal of the thesis, before finally a synthesis is reached, a combination of the thesis and antithesis, perceived to be the truth of the matter.

In terms of evaluating pitcher effectiveness, the thesis in question is what has been termed the ‘Old School’ of baseball thought – the idea that pitchers can will the ball to be hit harder or softer, that pitchers have control over how hard a ball is hit once it reaches the bat. The antithesis, then, is Voros McCracken’s idea of Fielding-Independent Pitching, the idea that the Three True Outcomes – strikeouts, walks, and home runs, those outcomes considered most within a pitcher’s control – are the only things within a pitcher’s control.

The synthesis of these two, then, has been the advent of derived, or regressed, FIP-like equations, such as SIERA or xFIP. It’s been proven to be the case that pitchers do have control over Ground Ball and Fly Ball rates – Justin Masterson’s sinker, for instance, induces ground-ball contact at an elite rate, but it does not limit the strength of contact that is made with the sinker. Hence, the incorporation of the two pitcher-dependent Batted Ball outcomes – GB/FB – has led to the advent of xFIP, the synthesis between old-school and new-school.

Yet there exists an opinion that the Indians’ pitching staff is not so skillful as their xFIP lets on. In the article ‘The Indians’ Paradoxical Pitching Staff‘ published nearly a week ago, Tony Blengino detailed why he believed the Indians’ pitching staff was better than its below-average ERA but worse than its xFIP. While Blengino has compiled thus far an extremely lucid, intelligent body of work, this particular assertion is very probably incorrect.

Blengino notes, correctly, that the Indians’ pitching staff has been excellent at those things within its control and has experienced poor luck with regard to its sequencing (a low 68.0% left-on-base rate for its pitchers – highly likely to regress to league average, improving run prevention), and he also rightly notes that the Indians’ defense has been poor, particularly its infield, ranking 29th in the majors at 25 runs below average. Yet the assertion that the pitching staff is in part to blame for the gap both neglects that the pitching staff has allowed weaker-than-average contact and also tremendously understates Cleveland’s defensive deficiencies.

Blengino asserts that FIP/xFIP fail to capture all of a pitcher’s effectiveness: in sum, that the Cleveland pitching staff has a deficiency in strength of contact allowed. If the Indians’ pitching staff did have substantial strength of contact issues, it should be reflected by a higher-than-average rate of difficult defensive plays. Fortunately, there exists a way to prove or disprove whether allowed contact results in low-probability or high-probability plays: FanGraphs has, within the past several months, added Inside Edge Fielding statistics. Inside Edge Fielding breaks down defensive play types as follows:

  • 0%: Impossible Plays
  • 1-10%: Remote Plays
  • 10-40%: Unlikely Plays
  • 40-60%: Even Plays
  • 60-90%: Likely Plays
  • 90-100%: Routine Plays

The descriptor attached to each play type indicates the difficulty with which one might make that play – a team that had an above-average ratio of routine plays would be expected to have a lower BABIP against, whereas a team with a disproportionate number of remote plays would expect to have that reflected in their BABIP. Pitching staffs with a large number of Remote or Impossible plays, for instance, would appear to have substantial ‘strength of contact’ issues. The following chart explores difficulty tiers as a percentage of overall plays and compares that rate to the rest of the league. As noted, it would make sense that stronger contact would result in higher rates of low-probability plays.

Induction Rate Cleveland Pitchers League Average Cleveland Rank
Impossible 9.62% 9.74% 18th
Remote 5.84% 6.84% 23rd
Unlikely 3.44% 4.46% 26th
Even 2.75% 3.37% 25th
Likely 7.22% 7.78% 20th
Routine 71.13% 67.82% 6th

Source: FanGraphs, Inside Edge Fielding

Contrary to the narrative presented, the pitching staff has allowed a higher rate of routine plays than league average, and a rate of unlikely plays substantially below league average. If ‘strength of contact’ were a pitcher skill, and if the end result of weak contact were high-percentage plays, the chart above decisively indicates that the Indians are one of the better teams in the league at inducing weak contact, directly subverting what one assumes to be the central premise of the argument.

One is left to find explanations for this dissonance – the dissonance between an excellent pitching staff and a worse-than-average rate of runs allowed. While there exists some appeal to place a portion of blame on the pitchers, the entire dissonance can be explained by the Indians having the 29th-ranked (or, as the following charts suggests, the far-and-away worst) defense in the majors. While the distribution between the aforementioned difficulty tiers is a pitcher-batter interaction, the rate at which each of these plays are converted are solely a fielding action. What follows are the rate at which the Cleveland defense converts each play type, followed by the league-average conversion rate, and the rank of the Cleveland defense in that category. (Note: Impossible Plays are removed, given that they – by definition – are never converted).

Conversion Rate Cleveland Conversion League Average Cleveland Rank
Remote 0.00% 4.68% T-21st-30th
Unlikely 0.00% 25.22% 30th
Even 12.50% 56.20% 30th
Likely 76.20% 81.80% 25th
Routine 94.70% 97.53% 30th

Source: FanGraphs, Inside Edge Fielding

In baseball and in life, the overwhelming majority of situations call for qualified, nuanced responses. Things are rarely so cut-and-dry that absolute answers are merited, and frequently, the existence of an apparently absolute answer prompts one to dig deeper. When it comes to the gulf in ERA and xFIP, this author did precisely that, looking for some indication that the pitchers were in some way responsible or did something that might lead to a gulf in xFIP. As evidenced by the above chart, however, reality has no time for nuance – it suggests that the Cleveland Indians’ defense is not merely bad, but the worst in the league by a margin that isn’t particularly close.

Extreme situations like this is why statistics like FIP and xFIP exist and are so critical to player and team evaluation: so that extreme defensive outcomes don’t tarnish an otherwise effective pitching staff. Output statistics are a function of their inputs. Given extreme inputs – in this case, defense so extreme as to be the worst in the league by a large margin – the output statistics will also be extreme – the second-largest ERA-FIP gap in the league behind only the very sad Diamondbacks.

Attempting to find nuanced syntheses is an admirable pursuit. In the case of the Indians’ FIP-ERA gap, however, no further synthesis is necessary. Although it is the marginal cases that frequently challenge theories, this particular fringe case seems to only prove the point: namely, garbage defense in, garbage ERA out.


Author: The Zach Walters Appreciation Guild


One thought on “Cleveland’s ERA-FIP Gap: The Defense IS That Bad

  1. Pingback: Rebuilding Salazar: Never a Groundball Pitcher | Cleveland Sports Insiders

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s