Long-term readers will remember that, a few years back, we used to have a feature where we looked at the worst umpire calls of the week. This came thanks to a Twitter account which used Pitch f/x data to determine how location matched up against an actual strike zone. Sadly, the account in question went silent. But thanks to a Shelley Duncan retweet, the @UmpScorecards account was brought to my attention, just in time for Opening Week. I think it will provide an adequate replacement. So you understand what’s going on, let’s break down their card for the D-backs first game of the year, and talk about what it shows.
[I wanted to embed the Tweets, but that feature here crops a significant amount off the bottom off the images. I’m directly embedding the images so you can see the whole card. I would, however, strongly encourage you all to give @UmpScorecards a follow on the Twitter for this invaluable work]
- Overall accuracy is based on the percentage of calls the umpire got right, based on the official strike zone.
- Overall favor. Every situation has an expected run value, based on the number of outs, base runner situation and ball/strike count. So each mistake can be assigned a cost, based on the difference between before and after expected run values. A blown call on the first pitch of the inning will be less important than one on a full count with two outs and the bases loaded. Tally them up over the course of a game, and you can see how each team benefited or was hurt by the missed call. They are from the perspective of the offense, so Barret’s calls cost San Diego 0.70 runs, but Arizona even more, at -0.87. A negative number indicates an umpire has a generous strike zone, calling actual balls as strikes, and hurting hitters.
- It’s often said that consistency is equally important for an umpire. It’s fair enough to have a large or small zone, providing balls are always balls and strikes, strikes. Overall consistency is an attempt to measure this. It starts with the umpire’s actual strike zone that day, defined as the smallest polygon encompassing all strikes. Everything inside that should be a strike; any pitches called balls are not consistent with his zone, and are counted here.
- Missed calls. This shows the standard strike zone in black, and the umpire’s actual strike zone in red. Note: these are shown from the catcher’s perspective, so are flipped from the TV view. You can see above this confirms Ted Barrett was feeling more generous. In green are shown the most extreme balls which should have been strikes, and in red are the strikes which should (based on the regulation strike zone, note) have been balls. On the right, you get more specific information about the three worst calls (which can be in either direction).
- Zone accuracy. This breaks down the pitches into “true” balls and strikes i.e. as decided by the regulation zone, and shows how accurate an umpire is at calling one or the other. The green chart is for balls, and red for strikes.
So, looking at the above, we can see that Ted Barrett had a bit of a stinker on Opening Day. That’s something with which anyone who watched the game will tend to agree. He had a large strike zone, but it wasn’t particularly consistent. He was at least fairly equitable, not favoring either side in his calls, but had about equal difficulty with getting the balls and strikes correct.
Now, let’s look at the rest of the games since Opening Day.
Angel Hernandez in “not worst umpire” shocker! Okay, his accuracy was almost as crap as Barrett’s. But if you could figure out the amorphous blob which was Angel’s strike zone, he called it very consistently, just one ball called inside it. Also, he did a lot better at getting the balls right than the strikes, with a woeful thirteen true strikes out of sixty missed by Hernandez. A lawsuit against @UmpScorecards is probably incoming. :)
Best performance of the San Diego series goes to Stu Scheurwater, though even his 92% accuracy rating was below the MLB average of 94%. The same goes for his consistency figure, though that came in only marginally under average (95% vs 96%). His zone seemed quite broad, adding a couple of inches on either side, and he also seemed fond of the low strike. But it was the edges that helped out the D-backs bigly, the biggest “mistake” a called strike three which ended the fourth inning when San Diego should have had the bases loaded.
Jose Navas was even more Diamondback-friendly, with a pair of big calls that went Arizona’ way, when Taylor Widener was working to his opposing number in the second with the bases loaded. As shown in the video below, both strike one and strike three were outside (remember, the missed calls chart is flipped from the TV), but were good enough to let Widener escape the jam. All told, his calls cost the Padres 0.8 runs and helped the D-backs by 0.31, making Navas the biggest difference maker of the opening week.
Finally, in Coors Field last night, Bill Welke showed how it should be done. Despite a temperature in the thirties at first pitch, and a game that went 13 innings, he became the first home-plate umpire to deliver an above-average performance for accuracy, rating 95% there. He still had his moments - though, perhaps surprisingly, not in extra innings, when he might have been forgiven for widening the strike zone! It was generally quite large, but he called it fairly, with the smallest overall favor so far, just 0.06 runs towards the visitors.
We’ll keep this up as a weekly feature going through the season, and see who are the best and worst umpires the D-backs face over the year. So far, they have benefited by 1.54 runs from umpire error. We’ll see how that changes!