October 4, 2016

Flu-Like Symptoms

Five Things You (Probably) Didn't Know About the 2016 Season

by Rob Mains

A lot of things happened this year that you know about. There were a record number of strikeouts. There were almost a record number of home runs. Mike Trout was once again the best player in the American League, but (SPOILER ALERT) won’t win the MVP. None of this is news to you.

So, I decided to look up some features of 2016 that might’ve eluded you. After all, the season went by quickly, and you were probably busy for a lot of it. Here’s what went on without drawing headlines.

The Hit-By-Pitch Scourge Lives On

If you ask a lot of older fans about hit batters, they’ll say one of two things:

1. They’re down because pitchers won’t pitch inside.
2. They’re down because umpires won’t let pitchers pitch inside.

This is one of the longest-lived, demonstrably false tropes about contemporary baseball. Here are the 15 seasons with the highest rate of hit batters since the American League was formed in 1901:

PA/HBP	Year
98.9	2001
101.4	2003
101.9	2004
102.5	1901
103.5	2006
103.7	2005
106.9	2002
107.5	2007
111.2	1911
111.3	2014
111.7	1905
111.8	2016
112.2	2008
114.61	1903
114.62	2015

Of the 15 seasons with the highest rate of hit batters, four are from 1901-1911 and 11 are from the 21st century. Batters were hit more frequently in 2001-2007 than in any year since 1901. This season’s hit-batter rate was the 12th-highest in MLB history. Here, let me show you graphically:

See? After the tiniest respite in 2015, the hit-batter rate moved up again in 2016. I fear that with contemporary pitch velocity, the sport may be heading toward another Ray Chapman moment. I’m serious.

Wild Pitches Everywhere

There were more wild pitches in 2016—1,808—than in any year in history. And that’s not just at the gross level; there were more wild pitches per game than in any year in history.

And if you associate wild pitches with a lack of pitcher control, you’re wrong. Here’s a graph of walk rates—a decent proxy for control—and wild pitches per walk.

We’re seeing record wild pitches despite stingy walk rates.

As I’ve argued, I think the high levels of both hit batters and wild pitches is directly related to the rise in strikeouts, since more strikeouts mean more pitchers’ counts, and more pitchers’ counts mean more pitches at the periphery of the strike zone in an attempt to get batters to chase, and more pitches at the periphery of the zone mean more hit batters and wild pitches if the pitcher misses by a few inches. Of course, the fact that these pitches are coming in faster than ever doesn’t help.

Starters’ Outings are Shorter than Ever, and There Are More Relievers Than Ever ...

Another refrain often heard is that modern starters lack the conditioning/toughness/pride to go nine innings, and instead they keep pitching fewer and fewer innings. As the chart below illustrates, that’s not really true. Starters’ outings did, in fact, pretty steadily decline, from 6.3-6.5 innings in the late 1970s to about 6.0 innings following the 1994 strike. But the decline stopped there. Innings pitched per start were in a narrow range of 5.79-6.06 for 21 years, from 1995-2015. This year, though, they tumbled to an all-time low, as the average start lasted just under five-and-two-thirds innings--5.65 to be exact.

With starters going fewer innings, managers had two choices: Use more relievers, or have relievers pitch longer. Reliever innings per appearance were up this year, but only negligibly, from 1.01 innings in 2014 and 2015 to 1.04 innings this year. Instead, managers went to the pen more frequently.

There were 3.15 relievers per game this season. That means that for every seven games, there were, on average, between the two teams playing, 44 relief pitchers. If you’re a connoisseur of managers trudging out to the mound, tapping one arm or the other, or seeing a different pitcher at the start of each of a game’s last three innings, contemporary baseball is for you.

… But It Didn’t Work

In April, Russell Carleton showed that late-inning comebacks were down in 2015 even though comeback opportunities, thanks to a lower scoring environment, were up. This year, with starting pitchers turning over the ball earlier to a growing succession of flame-throwing relievers, late-inning comebacks--one of the most exciting events in baseball--were under even more pressure ...

Wait. Scratch that. Comebacks have come back.

Here’s a graph of the percentage of games that teams won after trailing after six, seven, and eight innings.

As you can see, all three percentages moved up in 2016. Teams trailing after six innings won 13.6 percent of games, and those trailing after seven won 8.9 percent. Those are both the highest levels since 2013 and more or less in line with recent averages. Teams trailing going into the ninth won 4.2 percent of games, still low by recent standards but ahead of last year’s 3.5 percent.

And in case you were wondering:

Trends intact.

Oh My God! They Killed Bunting!

For years, we’ve known that bunts are often a poor strategy, because teams in general can be expected to score more runs with runners on x bases with y outs than with runners on x+1 bases and y+1 outs. Consider a runner on first with no outs. In 2016, teams scored 0.87 runs in that situation. After a successful bunt--with one out and a runner on second--the run expectancy drops to 0.67. The math’s different when a pitcher’s at the plate, or in late innings when a team’s playing for one run to win or tie, but overall, sacrifices are not a run-maximizing strategy. Nonetheless, they persist. Batters bunt, the home team cheers, the bunter gets high-fives in the dugout, and the announcers praise the small-ball execution.

Teams, though, have finally gotten the memo. This graph shows plate appearances per sacrifice hit. As with the hit-batters graph, since more plate appearances per bunt means there are fewer bunts, I’ve reversed the scale. And to make an apples-to-apples comparison (make that a pitchers-batting-to-pitchers-batting comparison) I’ve included only National League non-interleague games, to remove the effect of both the DH and American League pitcher batting:

To give you a sense of how dramatic the fallout has been, from 1955-2012 (58 years) there was a bunt, on average, every 83.9 plate appearances. The maximum was 95.6, the minimum 69.2. There wasn’t a lot of variance. Then, in 2013, bunts dropped to one every 96.3 plate appearances, the least frequent in history, though by just a bit. They dropped off a table after that: Once every 99.4 plate appearances in 2014, once every 119.9 in 2015, once every 134.1 in 2016. The average game in 2016 had 38 plate appearances per team, which means that for every 21 games National League games played last year, the two teams combined for 12 sacrifices. Just five years earlier, there would’ve been 19. Bunts aren’t extinct, but they’re endangered.

There are undoubtedly some under-the-radar trends I’ve missed. Come to think of it, that’s sort of tautological. Post them in the comments or shoot me an email if you think this article needs a Part 2.

Rob Mains is an author of Baseball Prospectus. Follow @Cran_Boy
Click here to see Rob's other articles. You can contact Rob by clicking here