Strikes, Trackman, the Giants, & the 9

Last week “The Curious Case of Giant Strikes” detailed how the Yomiuri Giants were so good at getting called strikes in 0-0 and 1-0 counts from 2009 to 2019 that they could make the rest of NPB look like amateurs.

The Giants aren’t the only team whose strike rate on pitches taken by hitters was unusual over a period of time in some counts, but they were the most extreme. The data is confusing, but thanks to a reader’s suggestion that I look at individual players, umpires and home-road splits, I discovered some more interesting stuff.

My first study was focused on the period of Yomiuri’s extreme superiority in 0-0 counts between 2009 and 2019, but between looking at players and umpires, and trying different things with the program I used to model thousands of randomly generated seasons, three things eventually jumped out at me:

  1. The real break occurred at the start of the 2018 season.
  2. Yomiuri’s most unlikely advantage was not in 0-0 counts as originally expected, but in 1-0 counts, while its pre-2018 dominance in 2-1, and 1-1 counts was also more impressive than its results on first-pitch called strikes.
  3. Although not a lot of pitchers worked many innings for both Yomiuri and other clubs during the 2009-2022 period in which I have data for, nine did. And as a group, these nine had an easier time getting called 1-0 strikes when they played for the Giants than for other teams.

Here’s looking at you, ump

In 2018, 11 teams, excluding the infamously penurious Hiroshima Carp, began sharing data from the Trackman pitch tracking systems they had installed in their main parks, so that Nippon Professional Baseball could use it for “umpire evaluation and improvement.”

The data variance is so subtle within the scope of a single season that I doubt there was ever an “Aha” moment within NPB that caused umpires superiors to say, “stop giving easy 1-0 called strikes to Giants pitchers at Tokyo Dome,” but things did change.

Changing the turning point from 2009 to 2018 brought the Giants’ pre-Trackman dominance in all six counts in the study (0-0, 1-0, 2-0, 0-1, 1-1, 2,1) into focus.

The table below shows the Giants average from 2009 to 2018 of annual “Z scores”–the number of standard deviations above or below the league mean the team was–in each count, as well as the chance a team that is automatically assigned to the league’s best called-strike chance would achieve similar results in that count over 1,000 nine-season spans.

Yomiuri Giants compared to the CL in called strikes

CountsPre 2018 From 2018Pct chance
The 2nd and 3rd columns represent the Yomiuri Giants’ average of annual Z scores for these six counts, for the years 2009-2017, and 2018-2022. The final column indicates the chance that the best team at getting called strikes every year between 2009 and 2017 would do as well as the Giants did in that count during that period.

The chances above were calculated based on models of 1,000 nine-season stretches in which the Giants were automatically assigned that season’s highest possible chance of getting called strikes.

So while there was a 78 percent chance that the best team would match the Giants’ success in called-strikes in 1-0 counts, the chances that that team would match or better the Giants across the board is less than 1 percent.

Upon re-examination of the data before and after that point, it began to look like a subtle adjustment took place across Japan in the degree to which some teams’ pitchers suddenly got much better or worse at getting called strikes.

Since 2018 when umpires were ostensibly being graded on their strike calls, the Giants’ sizeable advantage in getting 1-1 and 1-0 counts was reduced to normal proportions. In fact, it is impossible that the team that is the best in the league at getting called strikes each year, such as the one the model generates, would be as bad as the Giants have been since 2018.

This is evidence of nothing but does suggest that the Giants were probably not the best team at getting called strikes every year from 2009 to 2017, making their called-strike rate even harder to explain from the standpoint of the team’s talent.

It’s not just the umpires

Ball and strike calls are volatile from season to season. It is normal for one team to be so much better than the league in a count one year, but below average the next because there are so many variables: individual pitchers’ abilities and approaches, umpires, and the quality of the hitters faced. But over a period of years, some teams have been consistently good, bad or average.

After pitch tracking data became available, some teams’ ability to get called strikes also changed quite a bit for better or worse, and some of that change might be attributable to the umpires, just as some of it might be attributable to better catcher framing and better pitch selection by teams making use of tracking data.

The team that has benefitted the most from called strikes since 2018 has been the Nippon Ham Fighters, who went from typically being well below the PL average before 2018 to being well above after. The Fighters, are, coincidentally, one of Japan’s more analytically driven teams.

Coincidentally, the team that said, “We’re good with the analytics we have (and won’t spend a penny to install Trackman),” the Hiroshima Carp, may very well have fallen behind the rest of the league in terms of employing that data to analyze their own pitching and catching approaches and have gone from slightly below average in the CL through 2017 to NPB’s worst at getting called strikes since 2018.

The nine

Here are the nine pitchers who split their time from 2009 to 2022 between the Giants and other teams, the fewest pitches that were taken in 1-0 counts–used to weight the averages, and their percentages of 1-0 called strikes with the Giants, and their other clubs.

PitcherMin pitchesGiantsOthers
Shun Yamaguchi444.444.386
Hirokazu Sawamura25.411.400
D.J. Houlton270.493.478
Kan Otake445.512.445
Toshiya Sugiuchi473.505.381
Yohei Kagiya109.370.371
Kazuto Taguchi111.499.541
Yuya Kubo88.351.443
Yuhei Takanashi82.415.459
Weighted Average.462.433

As a group, their weighted average called-strike rate in 1-0 counts improved from 43.3 percent with other teams to 46.2 percent during their time with the Giants.

This is not a huge amount but it does represent a 6.7 percent improvement when wearing a Giants uniform.

Again, it is symptomatic of the Giants’ numbers, although not evidence in itself that the Giants were getting outside help. It could have been, for example, the effect of having Shinnosuke Abe getting more called strikes than the rest of the league’s catchers.

Leave a Reply

Your email address will not be published.