Pitching (and hitting) backwards in 2014

Last March, I listed the starters that pitched backwards the most during the 2013 season. Pitching backwards usually refers to throwing breaking balls or offspeed pitches on fastball counts, but to narrow things down, I specifically looked for pitchers who throw the fewest fastballs to start an at bat. I also looked at the hitters who saw the highest percentage of breaking balls and offspeed pitches on their first pitch.

I defined offspeed and breaking pitches according to the general groupings Brooks Baseball uses. For reference, those groups are below.

Knuckleballs and other pitches not listed here were excluded; otherwise, R.A. Dickey would of course have led the majors.

Let's look at hitters first this year, including only those with at least 300 PA. The five hitters who saw the fewest fastballs on 0-0 counts in 2013 were Josh Hamilton, Pedro Alvarez, Wil Myers, Matt Joyce, and Juan Francisco. And in 2014?

...well, at least Joyce and Francisco switched places.

There is a lot of agreement between the hitters who saw a lot of offspeed stuff last year and those who saw a lot of offspeed stuff two years ago. This makes sense: If the book on, say, Josh Hamilton is that he chases breaking balls, pitchers will be more willing to throw him a breaking ball to try to get ahead of him. However, it's worth noting that, for a number of these hitters (including Hamilton), they actually have a higher wOBA in those backwards plate appearances than in those where they see a fastball first.

The list of backwards pitchers, at least, has more turnover. Like last year, I made two leaderboards. The first is sorted by percentage of non-fastball first pitches.

Even if the top is different, there are still several repeats, including Arroyo, Lohse, and Vogelsong. Justin Verlander is back on this list too. In 2013, his fastball averaged 94 mph and he threw 53 percent first-pitch breaking balls. In 2014, as his fastball velocity fell to around 93, Verlander went to his secondary offerings even more, increasing his backwards first-pitch percentage to 59 percent.

That's what happens to pitchers after ten years in the league: as fastball velocity declines, even the best have to transition from throwers to pitchers. For more evidence, look at the other ten-year veteran on this list, Felix Hernandez. ESPN's Jalen Rose likes to say only two things are undefeated: gravity and Father Time*.

* - Well, actually, Jamie Moyer left the game with a lead over Father Time, but the bullpen blew it and Moyer got a no-decision.

The other way to look at pitching backwards is usage ratio, which R.J. Anderson defined in his article as:

(First-pitch non-fastballs/first pitches)/(Total non-fastballs/total pitches)

From the previous table, Verlander throws a low number of fastballs on the first pitch, but throws basically an identical proportion of fastballs at other points in the counts as well, so his ratio is a hair over 1. On the other hand, Collin McHugh actually throws a lot fewer breaking pitches to start at bats — only 86 percent of what he throws the rest of the time. In McHugh's second year, I wouldn't be surprised to see him throw even more first speed breaking balls to keep hitters from learning his tendencies, especially considering the success he had when he started with his secondary stuff.

This is an eclectic mix of pitchers, and again one with a lot of the same names as last year (Davis, Estrada, and Price, for example). These are pitchers with a clear tendency on the first pitch — ten percent fewer fastballs is a substantial amount! But the same pitchers keep showing up on this list, and judging by the few wOBA numbers we see here, hitters aren't making them pay for it.

Mitchel Lichtman, author of The Book, has often argued that pitchers should randomize the pitches they throw in any given count to be effective. But these tables make it seem as if pitchers with definite tendencies can still make a living in the league, and even be pretty good. And an actual study into this effect (accounting for batter skill, game situation, and batter preferences, among other variables) would probably support Lichtman's hypothesis. Still, even if catchers are calling games suboptimally, the sheer difficulty of hitting a major league pitch probably gives batteries some wiggle room.

. . .

All statistics courtesy of Retrosheet, Brooks Baseball, and Baseball-Reference. PITCHf/x data courtesy of MLB Advanced Media.

Bryan Cole is a featured writer for Beyond the Box Score. He has never faced a breaking ball, but is sure it would end very badly. You can follow him on Twitter at @Doctor_Bryan.