Saturday, March 21, 2015

Fun Facts About Called Strikes

(Edited on 03/22/2015)
When reading the very good but complex recent article about pitch framing at Baseball Prospectus, I found myself wanting to step back and look more closely at the core of the measurement.  Before all the adjustments and the statistical modelling, it all starts with counting how many pitches are called for strikes as opposed to balls. 

So, I went to the retrosheet database for 2014 and analyzed all pitches that were called by umpires.  Any pitch that resulted in a swing was excluded.  There were 371,964 pitches called in Major League Baseball last year -  248,298 balls and 123,666 (or 33.2%) strikes.  The pitches are broken down by different splits such as home/away and ball/strike count in Table 1 below.  Here are some of the highlights:
  • The pitching team was more like to get a called strike at home (33.5%) than on the road (33.0%).  Could this be a home field advantage?  We would need to look at it more closely, but it's certainly possible. 
  • Not surprisingly, the most likely count in which to get a called strike was three balls and no strikes where pitchers know that most batters will be hesitant to swing (64.5%) were called strikes).  Conversely, only 8.3% of calls on no ball, two strike counts were strikes.
  • There was a higher percentage of strikes called in the first inning (34.7%) than any other inning.  The percentage remained high for the first three innings, perhaps because starting pitchers are sharpest early in the game and the first time through the order.  The proportion dropped below average in the middle innings, but rose to 34.2% in the ninth inning when closers came into the game.
  • Called strikes were more common with right-handed batters at the plate (33.5%) than left-handed batters (32.9%).  
  •  Pitchers/catchers got more strike calls when their team was ahead (34.1%) versus behind (32.1%).  Of course, we need to consider that they may have had leads because they were throwing strikes rather than the other way around.   
  •  More strikes were called when there were fewer outs - 34.8% with no outs, 32.8% with one out and 32.1 with two outs.
  • Pitches were called strikes more frequently with the bases empty (35.3) than any other base runner situation. Strikes were only called 25.6% of the time with a runner on third.

Table 1 -Called Strike Splits -MLB Totals, 2014


Split
% Called Strikes

(N=371,964)


Total
33.2


Site

Home
33.5
Away
33.0


Count

0-0
45.8
0-1
21.6
0-2
8.3
1-0
41.9
1-1
25.0
1-2
10.6
2-0
46.6
2-1
29.2
2-2
14.7
3-0
64.5
3-1
38.7
3-2
19.4


Inning

1
34.7
2
34.4
3
34.5
4
32.3
5
33.0
6
31.7
7
32.4
8
32.3
9
34.2
10+
31.8


Pitcher Hand

LHP
33.2
RHP
33.3


Batter Hand

LHB
32.9
RHB
33.5


Score

Behind
32.1
Tied
33.6
Ahead
34.1


Outs

0
34.8
1
32.8
2
32.1


Baserunners

Bases empty
35.3
1st
32.3
1st, 2nd
30.6
1st, 3rd
27.8
2nd
29.1
3rd
25.6
2nd, 3rd
25.9
1st, 2nd, 3rd
29.8

Data source: Retrosheet.org


3 comments:

  1. It would be interesting to see if some teams get the benefit of more called strikes when pitching, or fewer when hitting, also whether good teams catch more of a break than bad teams, although this could be a chicken/egg thing. Thanks as always for interesting data that nobody else is offering.

    ReplyDelete
  2. There definitely seem to be pitcher, batter and umpire effects and the most recent advanced measure at BP adjusts for those things. It is kind of circular though and it's hard to isolate the catcher from other factors.

    ReplyDelete
  3. Berdj Joseph RassamMarch 27, 2015

    It's interesting to see such high percentages of strikes called when there have been no strikes called so far (0-0, 1-0, 2-0, and 3-0 counts).

    ReplyDelete

Twitter

Blog Archive

Subscribe

My Sabermetrics Book

My Sabermetrics Book
One of Baseball America's top ten books of 2010

Other Sabermetrics Books

Stat Counter