Golden Glove voting

Here’s how I voted for this year’s Golden Glove Awards

Central League
Pitcher: Takumi Akiyama, Tigers
I would have prefered Nomi, but he’s not on the ballot. Shoichi Ino and Kazuto Taguchi also appear to be deserving winners.
Catcher: Ryutaro Umeno of the Tigers.
More playing time would have made Umeno a no-doubt selection this season over Seiji Kobayashi and Yasutaka Tobashira…
1st Base: Jose Lopez, BayStars.
2nd Base: Tetsuto Yamada, Swallows.
Ryusuke Kikuchi was well off his game this year. He still makes lots of fantastic plays, but he was well off his norms, while Yamada – has become increasingly dependable.
3rd Base: Toshiro Miyazaki, BayStars. Tomohiro Abe of the Carp probably wins this if he spent more time at 3B.
SS: Hayato Sakamoto, Giants.
Sakamoto took a stride forward from last year, when he was about even with Kosuke Tanaka of the Carp.
OF: Masayuki Kuwahara, BayStars
OF: Takayuki Kajitani, BayStars
OF: Yoshihiro Maru, Carp.
Yoshitomo Tsutsugo had a terrific year and should get some votes, although Yang Dai-kang was probably the CL’s best outfielder in terms of quality.

Golden Glove
Pacific League

Pitcher: Hideaki Wakui, Marines.
He NEVER hurts himself with his fielding.
Catcher: Tatsuhiro Tamura, Marines
This is really a tough choice. There are indications Kai was a little better defensively, but Tamura played a LOT more, so he deserves credit for that.
1st Base: Sho Nakata, Fighters.
2nd Base: Daichi Suzuki, Marines
3rd Base: Nobuhiro Matsuda, Hawks
Zelous Wheeler had a great season and should get some votes.
SS: Kenta Imamiya, Hawks.
Takuya Nakashima missed playing time, while Ryoichi Adachi was fit and playing well. This could be the most interesting race of all next year if Sosuke Genda cuts down on his errors, since he could be right there among those three.
OF: Haruki Nishikawa, Fighters
OF: Yuki Yanagita, Hawks
OF: Shogo Akiyama, Lions
An off year for Shogo Akiyama, and Hiroaki Shimauchi of the Eagles could easily have gotten a vote instead of him, while Akira Nakamura had an excellente season for the Hawks as well.

Why Sarfate and not Kikuchi?

So why does Yusei Kikuchi play second fiddle to Dennis Sarfate in win shares? The essential answer is context.

First of all, it is very hard for relievers to rank so high unless they are extremely dominant and pitch a fair number of innings and get lots of saves – indicating many high leverage innings and that moves Sarfate into the conversation.

Still, the win shares system recogizes that Kikuchi was better at one level – the estimated contribution his raw numbers made to his club’s success. So why does Sarfate end up on top despite that.

The answer is wins. Not being credited with wins as the pitcher of record, but team wins.

On one level, this is a normal part of the system: Teams that win more games have more credit for wins to be shared by their players. But in the Pacific League in 2017, the Hawks won four games more than their run production and prevention would predict, and the Lions five games fewer.

Because the system is anchored on wins, you can’t get around the fact that in the big picture, the Hawks’ players’ numbers were therefore more valuable than the Lions’ players – who needed to score and prevent more runs to produce the same number of wins.

The system rewards individual performance on claim points. Pitchers get points for preventing runs in your innings beyond that which your fielders are credited with saving the team per inning.

You get points for striking out more batters and walking fewer, and for giving up fewer home runs. Because Kikuchi pitched nearly three times as many innings, he was able to save many more runs, but Sarfate was extremely effective and had a high leverage bonus because of his 54 saves. Still, Kikuchi gets 87.8 claim points – more than Sarfate’s 75.5.

Kikuchi’s claims give him 21 percent share of the Lions’ pitchers’ win shares. That is larger than Sarfate’s 17 percent of the Hawks total. But because the Hawks’ players’ numbers were more valuable, a Hawks pitcher saving 20 runs in 100 innings (adjusted for context and team defense) created more wins than a Lions pitcher who did exactly the same.

The Hawks’ individual performances were not all that much better, but in terms of wins, they were noticeably more valuable. Because the Hawks pitching staff produced many more wins, Sarfate’s contribution to the Hawks was a smidgeon more valuable than Kikuchi’s contribution to the Lions.

This connundrum pops up when the star of one team that wins more games than its runs scored and allowed suggest is compared to the star of a team that wins fewer games than it ought to. The Hawks won four games more than expected, the Lions five fewer. But you have to give the credit for that to the players, meaning, the Hawks’ players’ stats need to carry slightly more weight than the Lions.

That’s the rationale.

Is it accurate? It has its failings here and there, and it is not hard to believe that somehow Kikuchi must earn more credit, but in the end, everything depends on wins. At least this system doesn’t give players credit for winning games that their team didn’t – as WAR would.

If the Lions’ wins had more accurately reflected their runs and runs allowed, then the system would have seen Shogo Akiyama as the PL’s most valuable player – instead of Yuki Yanagita.

It also explains the presence of so many BayStars players and the absence of Tigers. The BayStars were hyper efficient, while the Tigers were not.

My 2017 NPB Awards Ballot

Here is my postseason award voting for 2017:

I have four rules:
1) Everything is about THIS season. It doesn’t matter what a guy did last year, that’s a different staory.
2) If two players are really close and one won a championship, go with the guy on the league championship.
3) No weight is given to age or potential unless a player is overwhelmingly superior but whose season value is low only because he missed playing time. This applies generally to rookies, but also applied to Shohei Otani last year.
4) The Best Nine Awards go to the most valuable player at each position. Whoever gets my MVP vote is automatically going to win a Best Nine Award. In the case of Otani last year, I gave Rule 3 precedence since he was easily the most productive pitcher and DH, but had slightly less total season value at each position than another player.

Here are my votes and a brief explanation of how I derive them.

Postseason Award Voting
Central League
MVP
1. Kosuke Tanaka, Carp 田中 広輔 (広島)
2. Yoshihiro Maru, Carp 丸 佳浩 (広島)
3. Tomoyuki Sugano, Giants 菅野 智之 (巨人)

Rookie of the Year
Yota Kyoda, Dragons 京田 陽太 (中日)

Best Nine
P – Tomoyuki Sugano, Giants 菅野 智之 (巨人)
C – Tsubasa Aizawa, Carp 會澤 翼 (広島)
1B – Jose Lopez, BayStars ロペス (DeNA)
2B – Ryosuke Kikuchi, Carp 菊池 涼介 (広島)
3B – Toshiro Miyazaki, BayStars 宮﨑 敏郎(DeNA)
SS – Kosuke Tanaka, Carp 田中 広輔 (広島)
OF – Yoshihiro Maru, Carp 丸 佳浩 (広島)
OF – Seiya Suzuki, Carp 鈴木 誠也 (広島)
OF – Yoshitomo Tsutsugo, BayStars 筒香 嘉智(DeNA)

Postseason Award Voting
Pacific League
MVP
1. Yuki Yanagita, Hawks 柳田 悠岐 (ソフトバンク)
2. Shogo Akiyama, Lions 秋山 翔吾 (西武)
3. Kenta Imamiya, Hawks 今宮 健太 (ソフトバンク)

Rookie of the Year
Sosuke Genda, Lions 源田 壮亮 (西武)

Best Nine
P – Dennis Sarfate, Hawks サファテ (ソフトバンク)
C – Takuya Kai, Hawks 甲斐 拓也 (ソフトバンク)
1B – Hotaka Yamakawa, Lions 山川 穂高 (西武)
2B – Hideto Asamura, Lions 浅村 栄斗 (西武)
3B – Nobuhiro Matsuda, Hawks 松田 宣浩 (ソフトバンク)
SS – Kenta Imamiya, Hawks 今宮 健太 (ソフトバンク)
OF – Yuki Yanagita, Hawks 柳田 悠岐 (ソフトバンク)
OF – Shogo Akiyama, Lions 秋山 翔吾 (西武)
OF – Akira Nakamura, Hawks 中村 晃 (ソフトバンク)
DH – Alfredo Despaigne, Hawks デスパイネ (ソフトバンク)

The rationale behind my award votes
I base my votes on Bill James’ win share system, which – like all of us – has flaws, but also does one thing I like that WAR doesn’t: It only gives credit for actual games won. Players don’t accumulate wins by putting up numbers against a scale but by putting up numbers within the context of games won by a team within a league.

You start with wins, although I have to assign half a win for every tie because they are so common in Japan. I wish ties were worth half a win in the standings, since that would push the teams tying games closer to .500. NPB used to do it that way, but never mind.

Each team gets 3 shares for each win and 1.5 for each tie. So how do you distribute them? Step 1 is to estimate how many runs an average team in its league would score and allow given the parks each team plays in.

Let’s take the SoftBank Hawks. They won 94 games with no ties. That’s 3 * 94 = 282 win shares to be distributed among their players. The Hawks scored 638 runs, while allowing 483. The Hawks playing context is extremely unusual. The parks they played in in 2017 increased season home runs totals by 22 percent, while suppressing runs by 3 percent.

It’s not just Fukuoka Dome, though. It’s all the small and large parks the Hawks play in through the season. But the Hawks’ context makes them the PL team for which home runs are easiest to hit and runs are hardest to score.

The system uses that information to split the Hawks’ 282 win shares as follows: 136.94 for the hitters and 145.06 for the pitchers and fielders. The system then splits the pitchers and fielders based on things like double play and fielding efficiency, strikeouts, walks and so on.

This then gives SoftBank 102.28 win shares to be divided among the pitchers and 42.78 to be divied up among all the fielders. From that point we get into determine the relative claims of each player to those totals. All the hitters on a team are compared to each other and the win shares are distributed accordingly. The pitchers are a little more complicated because they require claim points for the higher leverage situations that middle relievers and closers encounter.

Fielders are even more complicated, and therein is one of brilliant elements of James’ system.

While modern measures (unavailable to the general public in Japan) can calculate an fielders’ efficiency, a good job can be done by estimating defensive quality by the players at each position for a given team by comparing each team’s results at a position to its league rivals with adjustments for the frequency of innings pitched by lefties (which increases ground ball opportunities for third basemen and shortstops) and ground balls.

If you adjust for the number of strikeouts a team gets, its totals for catcher put outs become relevant. The same goes for pitcher put outs, which influence the totals of assists by first basemen.

Each team’s postion totals are compared to the league norms. The positions on a team that exceed league norms will have more of the team’s fielding win shares to divide among the team’s players at that position.

In the case of the Hawks, the position breakdowns for win shares for 2017 are:
Catchers = 9.2 (No. 1 in Japan)
1st Base = 2.0 (5th)
2nd Base = 6.1 (1st)
3rd Base = 5.8 (1st)
Shortstop = 7.7 (5th)
Outfielders = 11.9 (3rd)
The Hawks team ends up looking like this after each player’s total is converted into an integer:
(Hawks players with four win shares or more)
Yuki Yanagita 27 (23 Batting, 0 pitching, 3.7 fielding)
Kenta Imamiya 22 (14.8, 0 , 5.7)
Nobuhiro Matsuda 21 (15.3, 0 , 5.7)
Akira Nakamura 18 (14.2, 0 , 3.9)
Dennis Sarfate 18 (0, 17.5, 0)
Alfredo Despaigne 17 (17, 0, 0)
Nao Higashihama 14 (0, 14.2, 0)
Seiji Uebayashi 14 (11, 0 , 2.7)
Kodai Senga 12 (0, 11.8, 0)
Takuya Kai 11 (5.6, 0, 5)
Rick van den Hurk 10 (10.3)
Seiichi Uchikawa 10 (9.2, 0 , 0.9)
Sho Iwasaki 10 (0, 10, 0)
Kenji Akashi 9 (7.5, 0, 1.7)
Shuta Ishikawa 6 (0, 6.4, 0)
Ryota Igarashi 6 (0, 6, 0)
Hiroki Takayas 5 (2.1, 0, 3.7)
Tsuyoshi Wada 4 (0, 4.3, 0)
Munenori Kawasaki 4 (2.8, 0, 1.3)
Tomoki Takata 4 (2.4, 0, 1.6)
Livan Moinelo 4 (0, 4, 0)
Yuito Mori 4 (0, 4, 0)