The 1931 New York Yankees had NO Passed Balls during the season, none.  They are the only team that did not; every other team had at least two. 

            And, on the other end of the scale, we have:

            All of the teams which gave up a lot of Passed Balls had a Knuckleball Pitcher in the starting rotation except the the 1965 White Sox, who had no knuckleballer in the starting rotation but two knuckleballers in the bullpen who threw a total of 300 innings, and maybe the 2006 Padres; I don’t recognize any of those guys as knuckleball pitchers.  The 2006 Padres were a good team; they won their division, and the Red Sox traded with them a lot, so I know most of those players pretty well.  But a couple of those pitchers. . . I don’t know. 

            Anyway, I wrote many years ago that the distinction between Wild Pitches and Passed Balls was an improper distinction, since it is not exactly a record of what happened, but a record of a third party’s judgment about whose fault it was.   It is, in a sense, like this:  suppose that when a Home Run is hit, the official scorer made a judgment as to whether a Home Run was hit because it was a bad pitch, a fat pitch that caught too much of the plate, or because the batter just hit a good pitch.   If it was a fat pitch, the pitcher would be charged with a Home Run Allowed, but there would be no entry for the batter.   If it was a good pitch that was hit anyway, the batter would be credited with a Home Run, but there would be no entry on the pitcher’s record.   This is the same thing; less dramatic, but it’s making a record of the official scorer’s judgment, rather than a record of the fact.  The party judged to be innocent is entirely left out of the record book. 

            I believe that I was the first person ever to make that argument.  Now, that is what millions of people have come to believe.  Looking back on it, I think that I over-sold the argument, and that there is more useful information in the "Passed Ball" category than I believed at the time, four decades ago.   I think that I over-sold the argument, but I don’t think that I was wrong.  I believe that if I had my choice between having two categories distinguished by the official scorer’s judgment (Wild Pitches/Passed Balls) or just one category which was recorded both against the pitcher and the catcher (Passed Pitches), I think we would be better off with just the one category. 

            In a previous article I wrote, about a team that was 6.1 standard deviations worse than the period norm:

6.1 standard deviations below the norm is by far the worst performance in any area within this series.   It’s weird data, kind of fake data.  

            This turns out to be untrue; I spoke prematurely.  The 1987 Texas Rangers were 8.1 Standard Deviations from the norm. 

            The 8.1 Standard Deviations number is essentially proof that the category is not described by a Bell-Shaped Curve.  If it was a normal distribution, the odds against one team out of 2,550 being 8 standard deviations from the norm would be astronomical—billions to one.  As I said about the Balks, it is weird data, kind of fake data. 

            We now have three reasons to look cockeyed at the Passed Ball data—first, that we know that it represents an arbitrary distinction between like events, second, that we know (from the knuckleballers) that the pitcher is a heavy controller of the event, and third, that we know the data itself is weird. 

            From 1900 to 1909 there were 397 Passed Balls per 100,000 plate appearances, with a standard deviation of .00163.            In 1900 there were limited and primitive Shin Guards and Face Masks for catchers, if there were any; my understanding of the history of Shin Guards and Face Masks is shaky.  Anyway, the number of Passed Balls dropped dramatically from 1900 to 1920 due to the development of catcher’s protective equipment.   Passed Balls have been fairly steady in number since 1920, and the ratio of the average to the Standard Deviation has been relatively constant:

            The teams which had the lowest Passed Ball Rates had an average won-lost record of 81-75, while the teams with the most Passed Balls averaged 75-81:

            But, as I said earlier (about Balks), 16 Passed Balls cannot reasonably lead to six additional losses, so most of this effect has to be incidental.  In other words, it isn’t that Passed Balls lead to losses as much as it is that good teams tend to be good at everything, including not letting balls get by the catcher.

            Mackey Sasser was a catcher with the Mets, 30 years ago.  He wasn’t a bad hitter; he hit .285, .291 and .307 in the first three seasons when he had 100 plate appearances, but he had 0.2 career WAR because, frankly, he was the worst catcher anybody ever saw.   In 1990 he caught 583 innings, which is the equivalent of 65 games, and in those 65 games he committed 14 errors and allowed 91 stolen bases.   I always kind of liked him; I always thought that there was a useful player in there somewhere, if you could find the place to play him.  Anyway, thanks for reading.