Last month, an article post titled “The Edge,” spelled out the facts that Yomiuri Giants pitchers, had for the past 14 years or so, been far more successful than any other team in getting favorable calls on 0-0 and 1-0 pitches. To be precise, this edge was extremely pronounced from 2009 – when the available pitch-by-pitch data begins – until 2019.

Unfortunately, I didn’t have a tool at that time to reasonably estimate how likely it was for a team to do what Yomiuri pitchers achieved. I have that now and can tell you it’s improbable and unlikely, but not impossible.

But before Giants fans jump up and down in self-righteous vindication that umpires have given their team no help, one needs to reconcile a paradox, which I will explain shortly.

Osamu Ino, a former CL umpire and NPB’s umpiring technical advisor spoke about this data in an unsurprising fashion from someone who believes in umpires’ ability to be extraordinarily objective.

“That’s because the Giants’ pitchers are good don’t you think?” Ino said.

I was skeptical at the time that any team could be that good since I could calculate that an average team’s chances of replicating the Giants’ feat were on the order of one in several million.

While I would willingly concede that during that stretch from 2009 to 2019, the Giants pitching was better than average, I had no way of estimating how that affected a team’s ability to succeed the way the Giants had in 0-0 and 1-0 counts.

Since then, however, I’ve dusted off my programing skills and wrote a program that simulated thousands of random trials per team per season based on each team being assigned a chance of success within an ordinary distribution, and then repeated that for 11-year stretches 1000 times.

Beginning with the assumption that the Giants were far better than the norm, I created a model in which one team had a huge chance of being the league’s best, and barring that, the second best. The first model had our called-strike kings assigned the league’s highest percentage about 73 percent of the time.

After that, a second model assigned the best team an 82 percent chance of being the league’s best, and finally I tried a model that guaranteed that one team would have the best chance of getting called strikes 100 percent of the time.

The results are as follows.

The first column gives the percentage of seasons in which our called-strike kings are assigned the league’s highest percentage of getting called strikes. The next two columns indicate that team’s chances of replicating the Giants’ success in getting called strikes from 2009 to 2019 in 0-0 and 1-0 counts. The final column indicates the chance that league’s best would succeed in both counts in a random 11-year period as well as the Giants did.

For a team to have matched the Giants’ superiority in getting called strikes in 0-0 and 1-0 counts during the period in question, it would have to by far the best in the league at getting called strikes over an 11-year period. It is not beyond reason to assume that from 2009 to 2019, the Giants were the Central League’s best in this regard.

It would therefore be reasonable to assume the Giants had somewhere around a 5.3 percent chance of achieving the called-strike success they actually did prior to 2020. Improbable as this seems, it does not rule out the possibility that Yomiuri’s success in those two counts was essentially due to pitchers’ skill and not any outside assistance from the umpires.

The Giants during this period were also much better than average at getting called strikes in 1-1 and 2-0 counts, but because they were not as extreme this doesn’t change the math a lot or make their overall success seem a virtual impossibility.

But from my point of view, one thing does.

The down side

If the Giants were in fact so good at getting called strikes through their pitching skill that the results they achieved from 2009 to 2019 are remotely possible, then we have to account for their very ordinary results over the same period in 0-1 and 2-1 counts. One might think that if their success was largely a result of skill, then that would carry over to other counts, where they were only better than average.

I estimated that there is just a 1 percent chance that a team best in the league at getting strike calls every year, could simply be above average over a period of 11 year in 0-1 counts. The Giants were the best in the CL over this entire period as a whole, but were not the best year-in and year-out, the way they were in 0-0 and 1-0 counts.

One year it was the Tigers, one year the BayStars, then the Tigers again, then the Giants, then the Dragons – something unlikely if the Yomiuri pitchers’ record in 0-0 and 1-0 counts was a result of their individuals’ skills. The same repeated in 2-0 counts, although the Giants were a little better but still not dominant the way a really, really good strike-throwing team should be.

Now for the boring part

The model I used borrowed from the work of Jim Albert and Jay Bennet in their 2001 book, “Curve Ball: Baseball, Statistics, and the Role of Chance in the Game.”

Using Excel to produce a pool of random percentages within a normal distribution curve, with each number representing the actual ability of a team to get an individual strike call. I then wrote a program that broke that pool into five sub-pools. From these I composed each league at random with one “called strike strength” from each of the best two sixths of the pool, two from the middle third, and two from each of the weakest two sixths.

This made each season’s six-team league a more-or-less normal distribution–the way things normally are when we are talking about groups of athletes.

I then assigned those percentages from best to worst in a fashion that represented a league where one team, ostensibly the Giants, is overwhelmingly dominant and another, representing the Hanshin Tigers, whose pitchers were also quite a bit more successful at getting called strikes than the league’s other four teams.

I then ran 1,000 sets of 11 seasons to see how often teams were able to either match or better what Yomiuri did, recording how often that team was assigned the best chance in the league.

The thing is, we don’t really know how good the Giants were in this period at getting called strikes through their own skill alone, but by assigning a dominant team the best chance of getting called strikes over a period of time, at least, one can get a sense about how reasonable their actual dominance in 0-0 and 1-0 counts was.

Always more questions

In the end, my position on the Giants’ ability to get called strikes much more than the rest of the CL in two of baseball’s most frequent counts, 0-0 and 1-0, has switched from blatantly impossible to highly improbable.

But when one considers that they had no huge advantage over the rest of the league in 2-1 and 1-1 counts, there should be doubts.

I’m also curious why, since 2020, the Giants just look like one of the better teams and not overwhelmingly dominant in the least. I wouldn’t be half surprised if the introduction of pitch-tracking software, ostensibly to evaluate umpires’ performances, has had an effect on called strikes in the CL.