Tag Archives: Relative age effects

Growing your own

The Pacific League has now won the last seven Japan Series and has a .532 interleague winning percentage. People have attributed the gap to more hard-throwing pitchers in the PL, or to larger ballparks and the DH that helped that league be better at developing pitchers.

But two years ago, former Giants pitcher Scott Mathieson attributed it to the drafting philosophies of the two leagues, that the six Central League teams have shown more inclination to draft “baseball instinct” over physical tools.

Is it the draft?

This study won’t address that question, but it does ask whether one league has had an advantage in the draft and its Siamese twin, player development.

The answer is yes, and no one will be surprised to find that the PL has had a clear edge in this area.

To answer this and another question for another study, I created a database with the draft ranking of every signed player in Nippon Professional Baseball’s annual autumn amateur draft. For each player there is also a corresponding measure of career value, using Bill James’ win shares as I have adopted them to fit NPB.

Since we’re more interested in the CL’s current troubles, I looked at the drafts since 2000 and broke them down into two, ten-year periods. Interestingly enough, the top four players drafted since 1999 all were by CL teams: Shinnosuke Abe, Takashi Toritani, Hayato Sakamoto and Norichika Aoki.

The win share totals include those from the major leagues.

Players drafted from 1999 to 2008

LeagueWin SharesEdge
Pacific League9,714+21%
Central League8,041

The results of the second group are of course much smaller since many of these players have had little or no chance to impact the score, so it’s probably too early to think the PL’s edge is shrinking.

Players drafted from 2009 to 2018

LeagueWin SharesEdge
Pacific League4,317+10%
Central League3,914

The good and the bad

Here are the breakdowns of the return on amateur talent by team draft. Just a note, the three players still active from the 1998 draft, Kosuke Fukudome, Kyuji Fujikawa and Daisuke Matsuzaka, are not included in the study.

The first PL table includes the Kintetsu Buffaloes in the Rakuten Eagles, since the Eagles are the team that took Kintetsu’s place in the league.

In both the 1999 to 2008 group and the 2009 to 2018 group, the CL placed just two teams in the top six. Another interesting point is that while the Yomiuri Giants have invested heavily in developmental players, essentially the team’s return came from two guys who are both retired, reliever Tetsuya Yamaguchi and center fielder Tetsuya Matsumoto, who both won CL Rookie of the Year Awards over a decade ago.

Central League drafts 1999 to 2008

TeamTotalRegularDevelopmentalNPB rank
Giants1,6431,5191244
Swallows1,4631,46306
Carp1,4121,41207
Tigers1,2641,26319
BayStars1,2471,247010
Dragons1,0831,083012

Pacific League drafts 1999 to 2008

TeamTotalRegularDevelopmentalNPB rank
Fighters1,9501,95001
Lions1,9471,94702
Hawks1,8111,80473
Eagles1,5791,579535
Marines1,3111,226858
Orix1,1161,116011

Central League drafts 2009 to 2018

TeamTotalRegularDevelopmentalNPB rank
BayStars828792362
Carp73173015
Giants70169747
Dragons59959549
Swallows545543211
Tigers5105001012

Pacific League drafts 2009 to 2018

TeamTotalRegularDevelopmentalNPB rank
Lions89389121
Marines77677243
Fighters73973904
Hawks7115401716
Buffaloes63662888
Eagles568562610

The months matter

A recent discussion in the “Hey Bill” feature in billjamesonline discussed why some players do better than others and brought up the topic of relative age effects. I did a study about 10 years ago about the effects of NPB players’ birth months that was published in the Daily Yomiuri, which means it’s disappeared from the web. The upshot of that study was that players born from April 2 to June 30 are over-represented in the NPB amateur draft and, on average, have less valuable careers than player born from July 1 to April 1–the cutoff date for school admissions.

Children born on April 1 will enter school in Japan a year before a child born the following day.

Overview

I replicated the study using every domestic player signed by an NPB team from the end of the 1965 season through the start of the 1997 season. Omitting four players I don’t have birth dates for, that remaining group of 2,160 players contains two active players, Ichiro Suzuki and Kazuya Fukuura. And whatever they produce in 2019 is not going to affect anything one way or another. The starting point of the study was set by the introduction of NPB’s first draft in 1965.

Breaking down each quarter of a year by birth month — with April 1 counting as March — and draft round. The most populous cell is the 127 signed first-round picks signed who were born from April to June. The second most is the 121 players born in those months taken in the second round. As expected, the 341 “haya umare” or early-born players whose birthdays go from Jan. 1 to April 1, make that quarter the least populous.

The table below gives the career win shares produced by players born in each quarter and the total number in each group, without reference to draft round.

The last thing that needs to be mentioned is the problem of value in the major leagues. Major League win shares are given 20 percent more weight in the calculations. It’s just a guess. They could be 50 percent more valuable for all I know.

Distribution of domestic players by birth-month quarters

Apr-JunJul-SepOct-DecJan-Mar
Avg WS210.0230.5306.8223.0
Number754651414341
Percent of total34.930.119.215.8

The favoritism in the draft show players born in the April-June quarter is exacerbated by an even higher share of those players taken in the first two rounds, and by the performance of those players.


Value rank of birth-month quarter by round

RoundQuarter starting Avg WS Best career
1stJuly68.8Kazuhiro Kiyohara, 1B
1stJanuary60.2Masaki Saito, P
1stOctober59.7Koji Yamamoto, CF
2ndOctober55.6Taira Fujita, SS
3rdOctober54.7Hiromitsu Ochiai, 1B
2nd January48.1Hiromitsu Kadota, DH
1stApril44.6Hideki Matsui, CF
2ndJuly44.5Keishi Suzuki, P
4thOctober39.0Ichiro Suzuki, RF
3rdJanuary38.8Yoshihiko Takahashi, SS

Discussion

Another thing that needs to be mentioned is that the birth-month quarter starting in January is largely populated by pitchers and catchers. In my previous study, I found that more than a quarter of the players drafted as catchers were born between Jan. 1 and April 1.

When I first did this study, a number of people gave me what I’d snarkily call “baseball announcer explanations” for why players born from October to April 1 outperform the players who are chosen more often by pro teams. The most popular one of these was, “Oh, they’re used to overachieving, so they try harder.”

All these guys try hard. I think there are three things going on.

  1. Accessibility
  2. Age bias
  3. Burnout

Accessibility

Players who are born after April 1 are larger and physically more developed than players months younger than they are. This gives them more time to play, more time to stand out and be noticed by coaches, who select them to play so that they can be seen by scouts.

Age bias

Because players born from October to April 1 are less physically developed than the players they are competing against, they are less likely to dominate competitions when scouts are watching.


Burnout

This is something that hadn’t occurred to me until recently. According to people who know a lot about how youth baseball functions in Japan, many of the players who eventually turn pro in Japan are not the best in their age groups when they are young. Amateur sports in Japan are intense, year-round, meat-grinding wars of attrition.

The best players typically become pitchers, and because competition (with the exception of university baseball) is in single-elimination tournaments, those aces throw game after game until their bodies break down. They are then surpassed by those who were a step behind them a year or two earlier.

Many of Japan’s best pitchers were not aces in elementary school or junior high. Masahiro Tanaka was a catcher until high school. Koji Uehara ran track in junior high and was an outfielder until his senior year in high school, when his school’s ace, Yoshinori Tateyama began to break down from injury.

It is not that much of a stretch, then, to see many of those players born from April to June as being at the end of their physical tethers by the time the pros call on them.

I know I’ve talked about this before

If we make a top-25 of players in NPB’s draft era, the best single draft round was the first round of the 1968 draft, Hall of Famers Hisashi Yamada, Koji Yamamoto, a player who has curiously been overlooked for the Hall, Michio Arito, and another who will eventually make it, Koichi Tabuchi.

The second best group are three from the Fab 4, the fourth round of the 1991 draft, Ichiro Suzuki, Tomoaki Kanemoto and Norihiro Nakamura.