Analytics

Monday, October 26, 2020

Post #4856 Rant of the Day: The 2017 Astros Sign-Stealing Kerfuffle--Again

 I was enjoying last night's World Series' game 5, when one of the idiot announcers started ranting about the Astros' cheating during their 2017 championship run. It was in terms of discussing starting pitcher Clayton Kershaw's World Series record; the Dodgers have won 3 of the last 4 NL pennants, losing the 2017 and 2018 Series. The announcer was speculating how much better Kershaw's record would have been if the Dodger had played in a fair Series with the Astros

The speculation is that the Astros stole catcher signs to the opposing pitchers using a camera planted in the outfield stands and would audibly signal to the batter which pitch is coming (the pattern was alleged trashcan banging to signal a curveball; I guess they figured the other team couldn't steal their trashcan bangs. #sarcasm). Each pitcher has a limited number of pitches, e.g., fastball, curve, slider, changeup, and/or others. Part of the duel between the  pitcher and batter is the uncertainty of pitch selection. For example, the availability of a changeup could affect a batter's timing against a  fastball specialist. 

Stealing signs is part of the history of the game. For instance, a runner on second might be able to see the catcher's signs, see the pitcher's grip on the ball, etc., and somehow tip off the batter. The general opinion is that this type of "natural" intervention is limited in scope and not the same as the unwarranted use of technology.

As a baseball fan during my high school and college years in Texas, I followed the Astros and in fact to the present, I've only attended a small number of Astros' home games in person as a paying fan, mostly when Nolan Ryan was their ace. (I'm really a Twins' fan, a story I've mentioned multiple times in the blog, which put me in an odd place when the Astros eliminated the Twins in this year's first round.) To those of us who have followed the Astros for years, while the Astros were part of the National League, the 2017 World Series win was the first in the history of the franchise, and in essence the effort to dismiss a hard-won, 7-game series on speculation over the nature, extent and/or effectiveness of alleged cheating is outrageous, and I wrote multiple tweets and posts on this topic months back.

First, let me point out roughly half of the games including the seventh game of the World Series in question were on the road, without this alleged home field advantage. The Astros had a slightly better road record than home record. I don't see any reason to conclude the alleged sign stealing improved the Astros' performance or that the home performance was inconsistent with the road performance. In fact, the Astros have been in the ALCS for 4 consecutive years, twice winning the AL pennant without any evidence of said misconduct extending beyond the first year. Houston simply has a talented roster.

Second, I read an LA Times piece that mentioned that rumors of Astros' sign-stealing had reached the Dodgers before the World Series. Now the very first thing I would do if I suspected the opposition knew my signs would be to randomly rotate/change my signs across, even within innings. The alleged cheating is only as good as the valid interpretation of the catcher's signs.  And unreliable tips from the dugout could adversely affect batting performance. 

Third, there are a lot of nuances to pitches. It's not just the type of pitch, but the speed, location, rotation, etc. I mentioned in a prior post it's one thing to know the pitcher is coming with a 100 mph fastball; it's quite another thing to hit it. Some batters have known issues with curveballs, may have issues with certain parts of the strike zone. Opposition research knows about it; the pitcher may have his own history of encounters with a batter, what works and doesn't. And pitchers make mistakes, e.g., a hanging curveball. A similar thing goes on with scouting or observing pitchers. For example, a pitcher may be having trouble getting his curveball in the strike zone today. His go-to pitch may be his fastball. Similarly, there may be a go-to pitch on a full count. 

There is no requirement for a catcher to call a game. There is no requirement to use signs. 

Finally, at least some data scientists, like Robert Arthur, have parsed the trashcan data compiled by Tony Adams who analyzed every 58 home game on video of the 2017 season by pitch and trashcan bangs. Up to a third of the alleged pitch calls were wrong, especially with runners on base, and related batting performance was significantly below average, whereas for correct calls aggregate batting average was only a few points higher without any increase of on-base percentage. Arthur concludes any benefit from purported stolen signs was negligible at best, certainly not worth the risks of getting sanctioned from the baseball commissioner who had warned clubs of consequences.