Tag Archives: runs per inning

Baseball, Paid Content, Research

Japan’s bunt paradox part 2

Leave a comment
Kenta Imamiya of the Hawks bunts with no outs in the top of the first inning against the Buffaloes.

In a previous rant and observation about Japan’s ubiquitous first-inning sacrifice bunts, I noticed that teams in Nippon Professional Baseball that bunt in the first innings of scoreless games gain no advantage in how often they put at least one run on the board AND score fewer overall runs, BUT win games more often.

Those results, based on the first innings of the 2,592 regular season games played between 2012 and 2014, looked suspicious, so I increased the study to include the games played from 2007 and 2011.  Of the eight years in the study, in only three of them did visitors win more often when trying to bunt the leadoff man to second in the first inning. The three years were 2007, 2013 and 2014–three of the lower-scoring seasons in the study.

NPB introduced a uniform, less-lively ball in 2011. Since then, scoring has decreased sharply. With that decrease, the cost of the first-inning sacrifice has decreased. Since the switch, visiting teams can expect to score .79 runs per inning when the leadoff man is not sacrificed to second. That is a decrease of .11 runs per inning in the same situations before 2011, while the number of runs expected per inning after a sacrifice has remained nearly constant (dropping from .69 to .68.

The strangest thing about bunting in the first inning–and almost half the time the leadoff man is on first in NPB a successful sacrifice follows–is that the chance of scoring one or more runs in the first inning after the leadoff man reaches first is NOT effected by a sacrifice. The NPB data show a slight advantage to sacrificing after the 5th through 8th hitters are on first base with no outs but no appreciable difference in the first inning with the team’s best hitters coming to the plate.

With current low levels of offense, bunting the leadoff man to second base in the top of the first is costing Japanese teams a 10th of a run per sacrifice — yet despite giving away outs and runs, the visitors employing this strategy are now making out like bandits: winning their games at a .513 clip compared to the .459 winning percentage of visiting clubs that “fail” to sacrifice the leadoff man to second.

One person suggested on Twitter that sacrifice bunts lead to more wins BECAUSE teams sacrifice more often with their best starting pitchers on the mound. A quick look shows there is something to this. From 2007 to 2014, Japanese visiting teams with a big winner on the mound (12 wins or more that season) will sacrifice the leadoff man to second in 54 percent of their opportunities. The percentage with lesser pitchers on the mound is 47 percent.

This bias remained more or less constant from 2007 to 2014, but somehow didn’t help visiting teams before 2011. Before 2011, visitors that sacrificed the leadoff man to second base in the first inning went 204-253 (.446), while teams that did not bunt the runner over went 255-265 (.490). 

[supsystic-tables id=”3″]

[supsystic-tables id=”2″]

Baseball, Paid Content, Research

The paradox of 1st inning bunts

Leave a comment
Hichori Morimoto getting his “wa” on with the obligatory sacrifice bunt.

There may be nothing duller in sports than teams employing tactics routinely in a predictable fashion. In Nippon Professional Baseball, the biggest offender is the nearly automatic sacrifice bunt after the leadoff man reaches first in a tie game. This begins in the first inning and never stops.

Yet, as much as we despair of watching Japan’s bunt pageant, something very strange is going on.

As expected, bunting with a runner on first base increases the expectation of scoring at least one run, but decreases overall scoring. In 2,592 NPB games from 2012 to 2014, the visiting team’s leadoff man were on first base 731 times. The next batter bunted 385 times — 344 of which were credited as sacrifices).

The visitors scored in 168 of those innings for a total of 266 runs. That’s at least a run 43.6 percent of the time and an average of .743 runs per inning. In the 346 times when the next visiting bunter — I mean batter — does not strike out trying to bunt or put a bunt in play, teams scored 299 runs and scored at least one 148 times. Sounds like a good deal doesn’t it. Teams that “fail” to bunt score nearly as often — 42.8 percent to 43.6 percent — while scoring 16 percent more total runs.

Yes, it looks like the visiting teams should retire the bunt if they’re giving away so many runs for so little gain. But that’s not the whole story. The teams that benefited by failing to bunt, also failed to win as often. It doesn’t make sense, but visiting teams scoring fewer than three runs after a bunt, won more often than teams scoring the same number of runs in an inning without a bunt.

winning percentages with: 0 runs: bunt .455; no bunt .378 — 1 run: bunt .538; no bunt .485 — 2 runs: .700; no bunt .526.

To say that Japan adores the sacrifice bunt is no exaggeration, and despite doing much better on the scoreboard without first-inning bunts, visiting teams from 2012 to 2014 did worse in win the win column when not executing the nation’s favorite tactic.