In the last post, I mentioned how visiting NPB teams were winning more often when they bunted in the first inning with no outs and a runner on first base. Someone suggested that perhaps scoring the first run when on the road was bigger than it is at home, but prior to 2011 — when teams were able to choose very lively balls, it was the home team that benefited by bunting in the bottom of the first in scoreless games.

Starting with play-by-play data from 2003 to 2016, I noted what the first batter did in the first inning what the following hitter did, how many runs were scored, each starting pitcher’s runs allowed per nine innings that season and whether the team won or lost.

From 2003 to 2010, visiting NPB teams posted a .458 win percentage in games when they attempted a bunt in the first inning after the leadoff man reached first base via a walk, a hit batsman, a single, an error, a fielder’s choice or an uncaught swinging third strike. When faced with those situations and the No. 2 hitter’s plate appearance did not end in an attempted bunt, the visitors posted a .504 winning percentage.

From 2011, visitors bunting in the first inning had .502 winning percentages, those not bunting in the top of the first with the No. 2 hitter had a .459 figure.

For home teams it was the reverse. Before 2011, they won more often when bunting. Since 2011, they are bunting more often and costing themselves wins.

Few aspects of Japanese baseball are as reviled by outsiders as much as the routine first-inning sacrifice bunt by a low-average, slap-hitting, small middle infielder. Boring because it’s predictable, and because teams score fewer runs when sacrificing seemingly indefensible.

At times, Japan seems like the land that logic forgot, but the arguments against the first-inning sacrifice may be making some headway. In 2013, 49 percent of first-inning plate appearances by No. 2 hitters after the leadoff man reached first ended in an attempted bunt. Since then, they appear to be in decline. In 2016, that figure was down to 29 percent.

Some managers appear to be listening to the argument that scoring fewer runs is a bad thing.

We know sacrifice attempts decrease run scoring. It follows that teams costing themselves runs at the start of a game when it is not clear how many runs will be needed are shooting themselves in the foot. Following that rational, if one matches actual wins and losses with games in which these first-inning sacrifices occur, one should be able to measure the cost of bunts in terms of wins.

So by bunting less, Japanese teams are ostensibly getting smarter, but are they winning more games?

The answer, if you are a visiting team, is no.

Using play-by-play data since 2003, one can track what No. 2 hitters do after the leadoff man reaches first.

From 2003 until 2010, when juiced balls disappeared after the season, visitors scored 0.76 runs per inning after 736 sacrifice attempts. Those teams had a .456 winning percentage.

When not bunting, visitors in that era averaged 0.92 runs in 1,107 innings with a .504 winning percentage.

Since then however, the tables have turned. Visitors from 2011 to 2016 averaged 0.68 runs in the 669 first innings they sacrificed in. They posted a .502 winning percentage. In the 732 innings without a sacrifice attempt, visitors averaged 0.81 runs and posted a .459 winning percentage.

When I raised this possibility a couple of years ago, at least one reader suggested the possibility of quality leakage, because teams tend to sacrifice more with their better starting pitchers on the mound.

Since 2011, the visiting starters when their teams sacrificed after the leadoff man reached first allowed had an average season runs allowed per nine figure of 3.92. The opposing starters in those games averaged 4.00 runs per nine.

In games without sacrifices, the visiting starters averaged 4.00 R/9, the home starters 4.12. It’s a small difference. Indeed, visiting managers are slightly more inclined to sacrifice in the first inning when their best pitchers are on the mound, but those pitchers don’t appear to benefit from the bunt anymore than their less-heralded colleagues.

In the tables below, I have included the average of the season R/9s of the visiting teams and opposing starters. The column labeled “Expected” is the expected winning percentage if teams scored and allowed runs at the same rates as the starters of those games.

Here is the next table:

Japan’s most bunt-happy manager is Hideki Kuriyama of the Nippon Ham Fighters. This past season, he had 28 situations with a runner on first and no outs in the first inning at home. His guys attempted a bunt 10 times, and his team managed an impressive .600 winning percentage. In the other 18 games, however, the Fighters were .875. On the road, the Fighters followed the NPB norm, a .600 win percentage with the bunt, a .556 win percent without it.

Teams are bunting less in the first inning, but what they should be doing is bunting less at home, and more on the road — where it appears to make a difference.

Bunts are not always just free outs. Who would have thunk it?