Saturday, October 10, 2020

Ball! Chapman's 0-2 Pitch to Brosseau in NYY-TB ALDS

Before Tampa Bay eliminated New York from the 2020 postseason, HP Umpire Marvin Hudson had a key 0-2 ball call during Rays batter Mike Brosseau's 8th inning at-bat vs Yankees pitcher Aroldis Chapman, but was it correct? We ran the numbers and checked in with Robo-Ump for the verdict.

With the score tied 1-1 in Game 5 of the American League Division Series, Brosseau stayed alive by taking Chapman's 0-2 offering somewhere in the vicinity of the outside corner: but exactly where was the pitch located relative to the outer edge of home plate and was Hudson's ball call proper?

After all, Brosseau eventually hit the eventual game-winning home run later in the at-bat, giving extra importance to the crucial 0-2 ball call. Yankees fans online demanded an explanation for this game-changing "blown call" (or was it?).

Setting the Framework
: As we've explained (many times) in the past, the pertinent variable for horizontal ball/strike calculation is px. To calculate the boundary of px that delineates ball from strike, we perform a bit of upfront math.

Because px is measured from the center of home plate (negative values to the left [from the catcher/umpire's perspective], positive values to the right; for simplicity's sake, we are only concerned with the absolute value of px [all values are treated as positive numbers]), we divide the 17 inch-wide home plate in half to get 8.5". Converted to feet (8.5 x 12), this is 0.708 feet to either side of home plate.

We also know that any part of the ball traversing any part of the strike zone makes the pitch a strike, so we must add the radius of a baseball to .708 to get .831 feet as our px boundary (.708 ft half-plate + .123 ft ball radius = .831 ft zone edge).

Thus, with the horizontal boundary set at 0.831 feet, all we must do is consider the 0-2 pitch's px value: if px is greater than 0.831.

Chapman-Brosseau's 0-2 px value was 0.85.

Now, the math becomes really easy: Because 0.85 is greater than 0.83 (sig digs), we know that the expected call for this pitch under ML Public is a ball (by just 0.23 inches): all three systems thus agree, Hudson's ball call was the correct call.

For better or worse, if Twitter is any judge, a pre-computerized strike zone prognostication from Baseball Prospectus may have come true. The prediction? "The primary issue is that using machine measurements to call balls and strikes will simply shift disagreements with the call from the umpire to the machine, or to the machine's operators."

And for the record, Hudson's final plate score numbers indicated a skew of +3 New York.

Video as follows:

Friday, October 9, 2020

Umpire Plate Scores - Comparing Scoring Methods

When we introduced UEFL f/x 3.0, we spoke of the plate score tool's transparency, accuracy, timeliness, and thoroughness for ball/strike umpire analysis. One question we received asks why numbers by another ump scorecard source doesn't match ours.

To begin, there exist roughly two main methods for strike zone analysis: one is to let MLB's Baseball Savant make all determinations as to whether a given pitch is located within or outside of the strike zone and the other is to parse the data to calculate and assign a probable ball/strike status to each pitch.

Boston University/UmpScores
The BU version, made famous from Mark T. Williams' umpire study, employs the former method, which is most convenient and less processor-intensive for a tremendously large set of data points. As we found in 2019, this Baseball Savant scraping method relies on a variable roughly known as "zone," which does exactly what it looks like: It assigns a single-digit number (1-9) for a pitch located within the strike zone and a double-digit number (11-14) for a pitch located outside the zone. Much like a MIDI file sans percussion from the 1990s, track 10 isn't used.

Because the Zone variable is derived from other variables to calculate what number to assign as a zone statistic, this approach essentially doesn't work with measured data and instead relies on a processed number pursuant to Savant terms and manipulation. Not great for the scientific method, but a lot easier and simpler to calculate whether an umpire's call of ball/strike matches with Savant's zone designation of 11-14 (ball) or 1-9 (strike). Thus, a ball call on a pitch with zone 14 would be deemed Correct and a ball call on a pitch with zone 9 would be Incorrect. Simple.

UmpScorecards
UEFL f/x and UmpireScorecards appear to share the second of data collection methods through analysis of raw variables px, pz, sz_bot, and sz_top.

The difference between UEFL f/x's tri-score report and Scorecards or Auditor's single, simplified report is primarily how calculations are made after collecting raw data.

Whereas UEFL f/x runs through a process to analyze each pitch three different ways for the three different scores, accounting for a certain measure of error and physical properties of a baseball (e.g., as we've stated since the beginning, Auditor artificially inflates umpire "missed call" statistics by failing to acknowledge that a baseball has a radius), the other systems only return a number for what we call ML Public, or a zero-error system.

Scorecards, meanwhile, took to creating graphics to represent what it calls "Worst Missed Calls in terms of change in run expectancy" (yes, there really is a significant font size difference, and that's part of the issue: it's designed as a tool to measure changes in run expectancy first and considers umpire accuracy determination somewhat of an afterthought). This proved problematic when, for Angel Hernandez's October 7, 2020 game, Scorecards claimed the worst missed call of the game (see above), labeled Pitch 1, was a ball call that...well...looks like a correct call, drawing several critical replies, such as "1 is a ball all day long. But we are counting that as 'inaccurate'?"

This is what a zero-error (ML Public) system does: creates zero allowance on the edges. Yet perhaps, thanks to these graphics, it might be easier to comprehend why exactly ML Public scores tend to trend lower than UEFL f/x (other than the margin of error issue): because computers, despite acting entirely logically, seemingly act decidedly inhuman in robotically applying criteria.

The other issue pertains to Scorecards admitting to not knowing where to draw the line as to the edge of the strike zone, and admitting to not knowing how Savant graphs its strike zones (e.g., how a full game's plot isn't normalized, dating back to the Brooks Baseball days of strike zone graphics). This, too, might just be problematic for large-scale analysis using very precise numbers where baseball is colloquially a game of inches, though in the 21st century, it has become a game of mere fractions.

This also explains the difference between Scorecards' numbers and our ML Public numbers: the values used for plate edges as well as the values used for the radius of a baseball might be different, which would explain different plate scores using conceivably similar zero-error methods. ML Public, for instance, just happens to calculate radius from circumference pursuant to Official Baseball Rule 3.01 (Auditor's favorite rule to ignore), and plate width from a combination of OBR 2.02 and Definition of Terms [Strike Zone].

Apparently, the pressure might be getting to Scorecards as evidenced by Friday morning's tweet, "I KNOW BALLS 1 AND 2 ARE VERY CLOSE. Please do not yell at me."

Must not be an umpire.

Video as follows:

Thursday, October 8, 2020

Dale Scott Guest Op-Ed - Of Umpire Labeling, CB & Angel

While watching MLB's 2020 postseason, 32-year Major League umpire and 16-yr Crew Chief Dale Scott noticed a troubling trend toward labeling, perception, and prejudgment, writing the following Op/Ed article exclusive to CloseCallSports following Angel Hernandez and CB Bucknor's Division Series plate games:

There comes a point where certain umpires trigger automatic "terrible" comments, regardless of reality or common sense.

CB and Angel are two of the names where this automatic critique takes place.

Umpires get labeled, good or bad, and once you get one it’s damn near impossible to get rid of it, that label is baked in the cake.

Example:  If an umpire is labeled as "a really solid ball/strike umpire," then has a sub par game, many of the comments are similar to, "he’s a good umpire, was just a little off," or "doesn’t usually miss that many but he controlled the game," or "more misses than usual but had a ton of close pitches." 

Angel Hernandez and CB Bucknor.
In other words, he’s given the benefit of the doubt and in most cases that positive benefit follows him for years if not his entire career, even if there is a slow decline as seasons go by.

Conversely, when for whatever reason, an umpire gets a bad label like "he has trouble behind the plate, misses a lot of pitches," (this label can, but not always, be a result of just one below average “big” game). Then if he has a great ball/strike game, you hear comments similar to, "amazing, not his usual bad job," or "anyone can have a lucky day," or "not bad but he still sucks."

In other words he’s not given the benefit of the doubt. In fact many times the focus isn’t on the good, but goes directly to any misses with "that was a terrible pitch, he’s horrible," or "no surprise, he’s always terrible." The negative label, fair or unfair, sticks and can follow him until he retires (and beyond).

CB Bucknor 'Fan Art' as UmpShow.
If you start a season with the staff on a level playing field, the negative labeled umpire starts a couple notches below and has to perform above average over several seasons (or several post seasons) just to get even. But one average or sub par game and you’re buried with an even deeper hole to crawl out of because of the baked in bias.

So enough of the automatic bashing of Angel or CB. Judge them without the negative label anchor before they walk on the field.

Editor's Note: See Related Post "Booby Bumbaca" (10/7/20).

We’re taught on day one that every game is a new day. As we’re putting on our uniform, we don’t drag former situations or issues with managers, coaches or players on with it.

I wish commentators, media, fans, and most importantly, the umpire brotherhood would use the same standard.

Author Dale Scott officiated in the American League from 1985 through 1999, and on the unified Major League Baseball staff from 2000-2017. During his 32-year AL and MLB career, Dale worked the 2013 Wild Card Game, 12 Division Series, six League Championship Series, and three World Series. He resides in Oregon.

Wednesday, October 7, 2020

Discussion of 2020 AL and NL Division Series

Join us for discussion of the 2020 American and National League Division Series: As is postseason tradition, we will post UEFL f/x plate scores for all ALDS and NLDS umpires, both preliminary statistics immediately following games and final statistics based on computer adjustment (post-game processing) the day after.

Performance plate scores are listed following the completion of each contest according to UEFL f/x (StatCast data and application of UEFL Rules 6-2-b-a [horizontal bound, "Kulpa Rule"] and 6-2-b-b [vertical strike zone, "Miller Rule"]) for called strikes and balls. Foul balls, swinging strikes, balls put in play, automatic balls, pitchouts, and hit-by-pitches are excluded from the analysis. Click here to learn how UEFL f/x 3.0 plate scores work.


- 10/5 HOU@OAK Gm 1: Tripp Gibson. 102/102 Balls + 49/55 Strikes = 151/157 = 96.2%. + 0 Neutral.
Final (Post-game processed) for T Gibson: 102/102 + 48/55 = 150/157 = 95.5%. Effect: -1 / -0.7%.
- 10/5 NYY@TB Gm 1: David Rackley. 123/127 Balls + 47/49 Strikes = 170/176 = 96.6%. + 0 Neutral.
Final (Post-game processed) for Rackley: 124/127 + 47/49 = 171/176 = 97.2%. Effect: +1 / +0.6%.

- 10/6 MIA@ATL Gm 1: Andy Fletcher: 73/78 Balls + 37/42 Strikes = 110/120 = 91.7%. +2 ATL.
Final for Fletcher: 74/78 + 37/42 = 111/120 = 92.5%. Effect: +1 / +0.8 / +1 Skew. +3 ATL.
- 10/6 HOU@OAK Gm 2: Ed Hickox. 90/95 Balls + 47/49 Strikes = 137/144 = 95.1%. +7 OAK.
Final for Hickox: 91/95 + 47/49 = 138/144 = 95.8%. Effect: +1 / +0.7% / -1 Skew. +6 OAK.
- 10/6 NYY@TB Gm 2: CB Bucknor. 111/115 Balls + 48/56 Strikes = 159/171 = 93.0%. +4 TB.
Final for Bucknor: 112/115 + 48/56 = 160/171 = 93.6%. Effect: +1 / +0.6% / -1 Skew. +3 TB.
- 10/6 SD@LAD Gm 1: Lance Barrett. 137/144 Balls + 52/54 Strikes = 189/198 = 95.5%. +1 SD.
Final for L Barrett: 139/144 + 52/54 = 191/198 = 96.5%. Effect: +2 / +1.0% / +0 Skew. +1 LAD.

- 10/7 MIA@ATL Gm 2: Dan Bellino: 93/95 Balls + 35/37 Strikes = 128/132 = 97.0%. +2 MIA.
Final for Bellino: 93/95 + 33/37 = 126/132 = 95.5%. Effect: -2 / -1.5% / +0 Skew. +2 MIA.
- 10/7 OAK@HOU Gm 3: Jerry Meals. 105/106 Balls + 45/49 Strikes = 150/155 = 96.8%. +3 HOU.
Final for Meals: 105/106 + 44/49 = 149/155 = 96.1%. Effect: -1 / -0.7% / -1 Skew. +2 HOU.
- 10/7 TB@NYY Gm 3: Mark Carlson. 103/107 Balls + 51/52 Strikes = 154/159 = 96.9%. +3 NYY.
Final for Carlson: 106/107 + 50/52 = 156/159 = 98.1%. Effect: +2 / +1.1% / 0 Skew. +3 NYY.
- 10/7 SD@LAD Gm 2: Angel Hernandez. 91/92 Balls + 44/47 Strikes = 135/139 = 97.1%. +2 LAD.
Final for Hernandez: 92/92 + 44/47 = 136/139 = 97.8%. Effect: +1 / +0.7% / +1 Skew. +3 LAD.

- 10/8 ATL@MIA Gm 3: Mark Wegner. 117/120 Balls + 29/32 Strikes = 146/152 = 96.1%. +0 Neutral.
Final for Wegner: 115/120 + 29/32 = 144/152 = 94.7%. Effect: -2 / -1.4% / +2 Skew. +2 ATL.
- 10/8 OAK@HOU Gm 4: Laz Diaz. 126/127 Balls + 46/55 Strikes = 172/182 = 94.5%. +2 OAK.
Final for Diaz: 124/127 + 45/55 = 169/182 = 92.9%. Effect: -3 / -1.6% / -1 Skew. +1 OAK.
- 10/8 TB@NYY Gm 4: Todd Tichenor. 107/108 Balls + 42/44 Strikes = 149/152 = 98.0%. +1 TB.
Final for Tichenor: 108/108 + 42/44 = 150/152 = 98.7%. Effect: +1 / +0.7% / +1 Skew. +2 TB.
- 10/8 LAD@SD Gm 3: Bill Miller. 126/128 Balls + 53/55 Strikes = 179/183 = 97.8%. +0 Neutral.
Final for Miller: 126/128 + 54/55 = 180/183 = 98.4%. Effect: +1 / +0.6% / +1 Skew. +1 SD.

- 10/9 NYY@TB Gm 5: Marvin Hudson. 105/107 Balls + 45/49 Strikes = 150/156 = 96.2%. +2 NYY.
Final for Hudson: 106/107 + 43/49 = 149/156 = 95.5%. Effect: -1 / -0.7% / +1 Skew. +3 NYY.

Note: The highest plate score during the 2019 Division Series was James Hoye's 99.3% (NLDS Gm 4).
The highest overall plate score during the 2019 Postseason was James Hoye's 99.3% (NLDS Gm 4).

Booby Bumbaca - USA Today Criticizes Bucknor While Simultaneously Showing Graphics Exonerating Him

USA Today
's Chris Bumbaca executed a most incredible self-own in the very first sentence of an article attempting to criticize umpire CB Bucknor's Yankees-Rays strike zone, writing, "blaming the umpires is a lazy excuse, especially in the playoffs."

Facts and figures cited by Bumbaca himself further cemented the article author's relationship with fatuity, as, while attempting to malign Bucknor for missing pitches, Bumbaca used graphics that accomplished two unintended goals.

First, the article included images from MLB-owned Baseball Savant that clearly indicated Bucknor made correct calls on several pitches Bumbaca alleged the umpire had missed.

And, second, these graphics seemed to expose Bumbaca's rapport with the rules: specifically Definition of Terms' "A STRIKE is a legal pitch when so called by the umpire, which..."is not struck at, if any part of the ball passes through any part of the strike zone."

Or, alternatively, the unusual critique unintentionally belied Bumbaca's own quality of visual acuity.

Either way, here's a Commentary Critique and fact check: Print Edition.

Claim: Pitch 5 is off the plate.
Claim 1
: "Pitch 5 is off the plate" (see graphic)
Factual Analysis: Although UEFL f/x found this pitch was located over the edge of home plate and thus categorized it as QOC = Correct (affirmed after processing), we know the general population prefers graphics, and indeed, Bumbaca's article referenced a Savant Gamefeed graphic of the at-bat. Of all pitches during the at-bat to critique, Bumbaca chose Pitch 5 of a first-inning NYY Deivi Garcia vs TB Randy Arozarena at-bat, which the graphic clearly indicates was indeed touching the strike zone.

We're left with two options: Either the author doesn't know the rule referenced above or the author failed to see the visualization referenced in his article and cited as the basis for the claim. Either way, the claim lacks merit. As for the lower pitches ruled "ball"—recall the post-game processing effect on sz_bot/sz_top and remember the accompanying bot/top strike zone boundaries are inaccurate for this reason. Curious that MLB hasn't yet addressed this issue after at least a decade of having been made aware of this problem.
Fact Check Result 1:  False .

Claim: Pitch No. 3 is a ball.
Claim 2
: "Pitch No. 3 is a ball" (see graphic)
Factual Analysis: It would be helpful if the Savant Gamefeed graphic indicated "called strike" versus swings and misses. Nonetheless, Pitch 3 (sidebar: choose a style-guide format and stick with it. Pitch 5 vs Pitch No. 3?) during Tampa Bay pitcher Tyler Glasnow's swinging strikeout of New York shortstop Gleyber Torres was located off the edge of home plate.
Fact Check Result 2 True .

Claim: Pitch No. 3 is off the plate.
Claim 3
: "Pitch No. 3 is off the plate" (see graphic)
Factual Analysis: Once again, if a journalist wants to use a graphic, it would really be helpful to use a plot or chart that actually supports the claim being made. Though less overt than Claim 1, above, this claim relating to Nick Anderson's strikeout of DJ LeMahieu similarly features a graphic indicating the pitch in question actually touched the strike zone. UEFL f/x had it as a correct strike call.
Fact Check Result 3 False .

There are others, but it's simpler to just list fact check results rather than waste more time: True False True True.

Bucknor's ALDS score on 10/6/20.
The bottom line is really the Bumbaca's lede: "blaming the umpires is a lazy excuse, especially in the playoffs." Even lazier? Not making the connection between graphics being used to denigrate an umpire's performance and the actual facts of the graphics and/or rulebook.

Nonetheless, if Bucknor's 93.6% game Tuesday (higher than one, lower than two others on Tuesday) is to be considered subpar by Bumbaca's standards, what of an article whose own accuracy percentage, pursuant to fact checking, is just 57% (four true claims vs three false ones)? And how is it that certain umpires—who routinely score higher than some and lower than others—routinely absorb criticism while the umpires they score higher than evade such blown-call hit piece critique?

A pending lawsuit alludes to racism.
Expect more science-averse "analysis" from the general public for Angel Hernandez's plate Wednesday night in the Padres-Dodgers series. Does Jamaican or Cuban heritage have anything to do with it; does Hernandez's famous lawsuit against MLB alleging racially-motivated actually portend to a larger systemic problem in the sports world?

I'm sure Twitter will remain civil and decidedly not prejudicial, not even for a second.

Video as follows:

Tuesday, October 6, 2020

MLB Ejection P3 - Lance Barrett (2; Jayce Tingler)

HP Umpire Lance Barrett ejected Padres Manager Jayce Tingler (balls/strikes; QOCY) in the bottom of the 6th inning of the #Padres-#Dodgers game. With one out and none on, Dodgers batter Chris Taylor took a 3-1 fastball from Padres pitcher Garrett Richards for a called fourth ball, after which Mookie Betts took a 0-0 fastball and 1-0 slider from Richard for called first and second balls before hitting a double; Tingler was ejected during a mound visit/pitching substitution. Replays indicate that of the 10 callable pitches thrown by Richards in the bottom of the 6th inning prior to Tingler's ejection, all 10 were properly officiated (10/10 = 100% accuracy), the call was correct.* At the time of the ejection, the game was tied, 1-1. The Dodgers ultimately won the contest, 5-1, taking a 1-0 lead in the NLDS.

This is Lance Barrett (16)'s second ejection of 2020.
Lance Barrett now has 8 points in the UEFL Standings (3 Prev + 3 MLB + 2 Correct Call = 7).
Crew Chief Bill Miller now has 8 points in Crew Division (7 Previous + 1 Correct Call = 8).
*UEFL Rule 6-5-d-2 governs pitching change/mound visit ejections. 10 callable pitches occurred during the fractional half-inning preceding ejection (bot 6). All 10 were officiated properly. Thus, QOC is Correct.

This is the 84th ejection report of 2020 and third of the 2020 MLB postseason.
This is the 36th Manager ejection of 2020 and first of the 2020 MLB postseason.
This is San Diego's 3rd ejection of 2020, T-2nd in the NL West (LAD 4; ARI, SD 3; COL 2; SF 1).
This is Jayce Tingler's 2nd ejection of 2020, 1st since Aug 5 (Mark Ripperger; QOC = Y [Balls/Strikes]).
This is Lance Barrett's 2nd ejection of 2020, 1st since Aug 23 (Joe Maddon; QOC = N [Check Swing]).

Wrap: San Diego Padres vs. Los Angeles Dodgers, 10/6/20 | Video as follows:

Monday, October 5, 2020

2020 AL and NL Division Series Umpire Roster

MLB announced its American and National League Division Series umpire roster as Major League Baseball's 2020 postseason continues. The crews of six umpires, supported by Replay Officials who do not join on-field umpires for their best-of-five ALDS and NLDS.

Crew Chiefs are indicated in bold text and by the -cc suffix, regular season crew chiefs with an asterisk (*), interim crew chiefs with a degree (°), and first Division Series assignments by the tag (^1st Division Series^). The following crews are presented in their Game 1 rotation, such that umpires assigned to the plate in Game 1 move to right field for Game 2, left field for Game 3, third base for Game 4, and second base for Game 5.

Umpires assigned to HP, 1B, and 2B, thus, are guaranteed to work home plate during the series, with umpires starting at 3B and LF scheduled for decisive Games 4 and 5, respectively, and if necessary. Due to baseball's unique 2020 postseason format, MLB has assigned an unprecedented number of umpires to the postseason. The Division Series is the second round of four.

AL Division Series A (Tampa Bay Rays vs New York Yankees) Umpires:
HP: David Rackley ^1st Division Series^
1B: CB Bucknor [Game 2 Plate]
2B: Mark Carlson° -cc [Game 3 Plate]
3B: Todd Tichenor [Game 4 Plate]
LF: Marvin Hudson° [Game 5 Plate]
RF: Mike Estabrook ^1st Division Series^

AL Division Series B (Oakland Athletics vs Houston Astros) Umpires:
HP: Tripp Gibson
1B: Ed Hickox [Game 2 Plate]
2B: Jerry Meals -cc [Game 3 Plate]
3B: Laz Diaz° [Game 4 Plate]
LF: Ron Kulpa° [Game 5 Plate]
RF: Adrian Johnson

NL Division Series A (Los Angeles Dodgers vs San Diego Padres) Umpires:
HP: Lance Barrett ^1st Division Series^
1B: Angel Hernandez° [Game 2 Plate]
2B: Bill Miller -cc [Game 3 Plate]
3B: Doug Eddings [Game 4 Plate]
LF: Alfonso Marquez* [Game 5 Plate]
RF: Quinn Wolcott

NL Division Series B (Atlanta Braves vs Miami Marlins) Umpires:
HP: Andy Fletcher
1B: Dan Bellino [Game 2 Plate]
2B: Mark Wegner -cc [Game 3 Plate]
3B: Bill Welke° [Game 4 Plate]
LF: Chris Guccione [Game 5 Plate]
RF: Stu Scheurwater ^1st Division Series^

ALDS and NLDS Replay Officials: Adam Hamari, Brian Knight, DJ Reyburn, Carlos Torres.

Pursuant to UEFL Rule 4-3-c, umpires selected to appear in the Division Series receive two bonus point for this appearance, while postseason crew chiefs receive an additional bonus point, but replay officials who do not appear on the field (e.g., Wild Card & Division Series replay personnel) do not receive points for this role.