An Autopsy of MLB Umpire Ron Kulpa’s Bad Night in Arlington

Umpire Ron Kulpa, whose work was at the cener of a major dustup with Houston Astros brass Wednesday night. (Photo by Tom Szczerbowski/Getty Images) *** Local Caption *** Ron Kulpa
Umpire Ron Kulpa, whose work was at the cener of a major dustup with Houston Astros brass Wednesday night. (Photo by Tom Szczerbowski/Getty Images) *** Local Caption *** Ron Kulpa /
facebooktwitterreddit

Taking apart MLB umpire Ron Kulpa’s umpiring during Wednesday’s controversy-filled 4-0 Astros loss to the Texas Rangers; what role did the umpire play in that outcome?

The MLB umpire at the center of Wednesday’s dustup with the Houston Astros did have a bad game, although — ironically –  the controversial second inning was among his best.

The Astros had plenty to complain about in the ball-strike work of plate umpire Ron Kulpa during Wednesday night’s 4-0 Astros loss to the Texas Rangers.

By the standards of major league umpires, Kulpa had an off night. It’s also true that an uncommon majority of Kulpa’s errors hurt the Astros.

Having acknowledged that, on the specific second-inning beef leading to the ejections of Astros coach Alex Cintron and manager A.J. Hinch, Kulpa was right.

More from Call to the Pen

The issue erupted in that inning when MLB umpire Ron Kulpa ejected Cintron, and later Hinch, for what they believed were bad calls leading to Carlos Correa’s strikeout.

A significant part of Hinch’s objection, and to the subsequent outrage expressed by Astros fans, lies less in the calls themselves and more in what appeared to be Kulpa’s imperial attitude regarding those calls. Hinch later confirmed that Kulpa had told him, “I’ll do what I want,” and many suggested he appeared to be smirking while doing so.

The intimation was that the Astros did not get a fair shake.

Setting aside questions about the propriety of what Kulpa told Hinch and his attitude in saying it, the raw data surrounding the ball-strike umpiring of the game suggests that neither the Astros nor Kulpa left the field with much to be proud of.

Start with the specifics of the calls that raised the Astros’ ire in the first place.

MLB’s PitchTrax data shows that among five strikes Kulpa called on Astros batters in that second inning, two were borderline but Kulpa got all five correct, including the first-pitch strike call to Correa that triggered the Astros outburst.

In fact, Cintron and Hinch probably were incensed as much by a pair of first-inning Kulpa miscalls. One was a first-pitch strike Kulpa called on Jose Altuve on an inside pitch, the other a 3-1 strike on Michael Brantley on an outside pitch. Both hitters eventually struck out.

We can use several metrics to assess Kulpa’s work behind the plate. Generally speaking, and with individual variations, umpires get about 93 percent of ball/strike calls correct. On Wednesday, Kulpa was called upon to make 145 such calls, and based on PitchTrax he miscalled 17 of them. That’s an 88 percent accuracy rate.

So he had a bad game.

But that is a raw figure and it requires refinement since the vast majority of ball/strike calls are so rudimentary that anybody with a fundamental understanding of strike zone could get virtually all of them correct. Most pitches are either basically right down the middle, in the dirt, conspicuously high or somewhere well off the plate. So when an umpire is rated at 93% (or, in Kulpa’s case, 88%), it needs to be noted that the vast majority of the ball-strike questions amounted to the umpiring version of “how much is 2 plus 2?”

For major league umpires, the challenging calls are those where the pitcher succeeds in coming close to the corners of the strike zone. There were 41 pitches in Wednesday’s game thrown within a ball’s width of the zone’s perimeter; that is, they were tough pitches to call. Most umpires only call about two-thirds to three-quarters of these correctly. MLB Umpire Ron Kulpa did even worse than that; he got 24 of them right, just 59 percent.

So he had a bad night.

And his bad night measurably hurt the Astros in a way that the conspiratorial among us would find indictable. As a general proposition, an umpire’s ball-strike decisions basically break evenly against both teams, but that was decidedly not the case Wednesday. Of Kulpa’s 17 bad calls, 13 favored the Rangers and only four favored the Astros.

We can use Weighted Win Probability Added – a measurement of the influence of any at-bat outcome on the game’s outcome – to at least estimate the impact of Kulpa’s wrong decisions on the Rangers’ victory. T

The answer: as much as 16 percent.

Given the fact that Texas starter Mike Minor and two relievers shut out Houston in a game the Rangers won by four runs, it would be wrong to ascribe to Kulpa too large an impact on the Houston defeat. Beyond that, there is no guarantee that had Kulpa made the correct calls on, for example, those first inning strikes-that-were-actually balls to Altuve or Brantley, the Astro hitters would have done something more productive than fanned.

Still, at least on a large scale, there is plenty of evidence that MLB umpire decisions can be impactful. Since only a handful of Kulpa’s miscalls came in ball-strike counts that tend to have large influences on an at-bats outcome, his actual influence on the game might have been significantly below 16%.

He showed a particular predisposition to ring up first pitch strikes whether they were on the plate or not. In a way, that’s good for Kulpa, since a first-pitch call influences the outcome of an at-bat to only a relatively small measure.  As a general rule, the average major leaguer’s batting average on a first pitch is about .345. At 1-0 it falls to  .344, but at 0-1 it only drops to about .324.

Next. Jays offense is simply offensive at start of year. dark

Compare that with the change that results when a pitch in a 1-1 count is wrongly called. If the count moves to 2-1, a batter tends to hit .344. At  1-2, his average falls to .166. But those calls are rare. Kulpa had only 13 1-1 calls to make Wednesday, and according to PitchTrax, he called all 13 correctly.