Friday 21 June 2013

New Stat

On Tuesday night, Josh Hamilton grounded into 3 double plays during a 3-2 loss to the Mariners.

In honour of that, I think its time to talk about a new stat - OPPA, or Outs Per Plate Appearance.

Even good hitters have bad nights, and bad hitters probably have more bad nights, but it takes a special confluence of opportunity and execution to create more outs than plate appearances.

To continue my apparent obsession with GIDPs, and how funny I apparently find it to be when hitters ground into multiple GIDPs, this post is all about the most-GIDP games ever. Unfortunately, triple plays are not as easily accessed, so if a batter created triple play outs I am going to ignore that for this post.

The methodology here is pretty simple:
1. Find all the batters who have GIDPed 3 or more times in a game
2. Count the number of outs they produced
3. Divide those outs by plate appearances.
4. Notice other interesting things.

That search returns 109 batters that have accumulated 3 or more GIDPs in a single game, with the first coming in 1934. So maybe every time I say "of all time" I don't really mean it.

Surprisingly, only one batter ever has GIDPed 4 times in a game. That batter is likely going into the hall of fame... as a manager. Need more clues? He goes by the name of J Torre. No wait, that's too obvious. Joe T.



On July 21, 1975, Joe Torre went 0-4 with 4 GIDPs, for an astounding 8 outs in 4 plate appearances, and an OPPA of 2.00. Yikes - that's as high as you can get, in this "study". The Mets lost 6-2 to the Astros in 9 innings. Torre created an obnoxious 29.6% of his team's outs.
Bot 1: Line Out / Single / Torre GIDP
Bot 3: Groudout / Single / Single / Torre GIDP
Bot 6: Single / Torre GIDP / Groundout
Bot 8: Single / Single / Torre GIDP / Walk / Flyout

Oops.

Next, there are eight illustrious batters who GIDPed 3 times in 3 plate appearances, for an equal but less impactful OPPA of 2.00:
Joe Adcock - July 20, 1955. Milwaukee loses 4-3 to Pittsburgh
Ted Martinez - September 25, 1974. Mets lose 6-3 to the Phillies
Jose Morales - May 17, 1980 (entered game as PH). Twins lose 14-11 to the Brewers
Dave Conception - June 13, 1986. Reds lose 3-2 to the Braves
Julio Franco - April 9, 1989. Ranges win 3-2 over the Blue Jays
Pete Incaviglia - August 17, 1989. Rangers lose 6-1 to the White Sox.
Jeff Baker - June 2, 2007. Rockies win 4-1 over the Reds.
Kevin Kouzmanoff - August 5, 2009. Padres lose 6-2 to the Braves.

There are some good players and some relative nobodies on this list, but I would bet that there are no or few speedsters who are getting GIDPed 3 times a game.

Josh Hamilton, this week, did not reach base in his 5 plate appearances, for a OPPA of 1.60. That ties him for 53rd all-time with 12 other batters (for players with at least 3 GIDPs in a game). Not bad.

Are you noticing a trend with these GIDP masters? I am! Most of these teams are losing! Not surprising when so many rally-killing outs are made by a single player.

But it looks like it's more complicated than that. OPPA at the high end of the scale tends to bunch to common numbers. Let's take a look at the Win percentage for some common OPPA numbers:

8 outs in 4PA or 6 outs in 3PA - OPPA = 2.00 (9 games): 2-7, 0.222
7 outs in 4PA - OPPA = 1.75 (43 games): 13-30, 0.302
8 outs in 5PA or equivalent - OPPA = 1.60 (13 games): 7-6, 0.538
6 outs in 4PA or equivalent - OPPA = 1.50 (16 games): 9-7, 0.562
7 outs in 5PA or equivalent - OPPA = 1.40 (16 games): 7-9, 0.438
Between OPPA = 1.20 and OPPA = 1.39 (9 games): 7-2, 0.778

What does this mean? Well, someone who could use statistics might start using p values and t tests, but I don't think I will bother with that.

I am reluctant to say that when a team's batter GIDPs 3 times and has an OPPA between 1.20 and 1.39, they are almost certain to win that game. That seems crazy. And I don't quite believe this trend all the way to the left side:


However, it actually does make some sense to me that when a team has a batter with at least 3GIDPs and an OPPA of at least 1.75, that hurts their chances of winning that game. That is 6-8 outs from a single batter in 3 or 4 plate appearances - not cool.

Other interesting things:
Only one player has >= 3 GIDPs and an OPPA of exactly 1.00, and that occured during a (I didn't know these existed) TIE GAME. This had to be some sort of strange perfect storm, right?

On Monday July 4, 1960 at Wrigley Field, in the second game of a doubleheader, the Giants and Cubs tied 1-1 in 14 innings. Wrigley Field had no lights back then, and both teams had games the next day, so...? Frank Thomas went 1-5 with 2BB and 3GIDPs for 7 outs in 7PA.

Then I looked further. The Cubs played 156 games that year, in a 154 game schedule, because they had ANOTHER TIE GAME. Again in the second game of a doubleheader, they tied the Cardinals 4-4 in 10 innings. Get some lights! Ernie Banks led the league with 156 games played for a 60-94 team. Weird.

Okay. Last interesting thing. More like a crazy batting line, so there must be something interesting about this game that was played like 30 years ago.
On May 8, 1984, HOF SS Robin Yount went 3-10 with a walk and 3 GIDPs in a 7-6 loss to the White Sox. That's weird, right? Who gets 11 PA in a game?

It is accompanied in the Baseball-Reference database with the following note:
Game was suspended in the top of the 18th with the score 3-3 and was completed May 9, 1984 at Comiskey Park I.

This game ended up taking twenty-five innings. Which is to say, they played 7+ innings after the game was suspended in the 18th.

BOTH teams scored 3 runs in the 21st inning to keep the score tied and the game going. How obnoxious. Finally, in the 25th, Harold Baines hit a walkoff home run with one out to end the game.


That game and box score just leads to more blog ideas. Oh boy...



No comments:

Post a Comment