Year of the Pitcher

by fjm235 on Jul 15, 2010 1:51 PM EDT

We've all heard about it. This is supposedly the year the pitchers are taking back the game from the hitters. We've had perfect games (actual and virtual) and no-hitters galore. ERA's are down almost everywhere. But how big a change is it really? Can we quantify it?

The answer is yes. Fangraphs WPA does it for us. If you add up the WPA's for the pitching staffs of all 30 teams you get +56.32. That says collectively the pitchers are nearly 113 games over .500. (Remember this is a zero sum game. So if the pitchers are +56.32 the hitters must be -56.32.) Divide that by 30 and you get +3.75 wins per team on average.

Why does this work? Because the Win Expectancy matrix is based on prior years. If the pitchers (and hitters) were performing this year as they did in the past, they should each be right around 0. +3.75 wins per team is a long way from 0.

To see how big a difference that actually is, it's helpful to count the numbers of teams in various ranges of WPA. First, here are the pitching staffs.

Great (>7 WPA) 4 teams, headed by the Rays at a phenomenal +10.80. All 4 of them are division leaders.

Very Good (4-7) 5 teams.

Good (2-4) 6 teams.

Pretty Good (0-2) 6 teams.

Pretty Bad (-2 to 0) 3 teams.

Bad: (-4 to -2) 1 team.

Very Bad (-7 to -4) 4 teams.

Awful (<-7) 1 team.

There are 21 teams with winning staffs against only 9 losers. 15 are Good or better against only 6 that are Bad or worse. In a normal year this distribution should look more or less normal.

As you'd expect, the distribution of teams from a hitters' perspective is nearly a mirror image. There are no great (>7 WPA) hitting teams and only 2 Very Good ones. Conversely, there are 4 Awful teams and 5 Very Bad ones.

Tweet Comment 6 comments | 0 recs |

Do you like this story?

Comments

is there a link to these distributions on fangraphs or did you organize this yourself? I would like to see which teams fall into each of these categories.

"Baseball is like church. Many attend; few understand." - Leo Durocher

by awix on Jul 15, 2010 8:43 PM EDT reply actions

Reply

Just select Teams, then Win Probability. Select either pitchers or batters, then sort on WPA.

by fjm235 on Jul 18, 2010 11:56 PM EDT up reply actions

I'm not so certain it's true that the Win Exp numbers are based on previous years' results

I believe Fangraphs is using Tango’s Win Expectancy numbers, which are calculated using Markov chains, and can be set to any run environment. I’m pretty sure all the WPA results on Fangraphs are based on either a 4.5 or 5 RPG environment and not adjusted for park or anything.

It may in fact be the year of the pitcher – but the effect you’re seeing is likely just the result of a lower run environment (which might be caused by better pitchers, but might be caused by something else).

by Dan Turkenkopf on Jul 15, 2010 9:10 PM EDT reply actions

You're probably right

I didn’t mean to imply the probabilities are based on any particular set of years, only that they are NOT based on current year data. If they were, you wouldn’t see such a large disparity favoring the pitchers over the batters. Nor did I mean to imply that the only (or even the primary) cause of the Year of the Pitcher is “better pitchers”. Frankly it’s hard for me to believe the quality of pitching throughout MLB could change that much from one year to the next.

by fjm235 on Jul 19, 2010 12:09 AM EDT up reply actions