Take it easy with your colleagues

Pitchers in the National League can take a breath every two or three innings when their colleagues come to bat. You are not guaranteed to get an out every time the opposing pitcher steps up to the plate, but life is certainly easier against the ninth spot hitter in the National League than in the American League.

Here I won’t talk about exceptions like Carlos Zambrano or Micah Owings. (If you want to look at good hitting pitchers of the past, point to this article written by David Gassko a couple of seasons ago).

Here I would like to look at how—and if, indeed—pitchers (those on the mound!) take a break when one of their colleagues grabs a bat.

Pitch selection.

Let’s start by comparing pitch selection against pitchers versus pitch selection against “normal” hitters.

Table 1a: Pitch selection against same-handed batters
(NP=National League non-pitcher; P=pitcher) 

           NP   P
  Change-up  7%  5%
  Curve    10%  9%
  Fastball 60% 71%
  Slider   23% 15%

Table 1b: Pitch selection against opposite-handed batters
(NP=National League non-pitcher; P=pitcher) 
     
           NP   P
  Change-up 17% 11%
  Curve    11% 10%
  Fastball 61% 72%
  Slider   11%  6%

Pitchers favor the fastball when throwing to their peers, going less to the slider and the change-up. The percentage of curve balls is roughly the same.

Following I’ll show you a similar table, this time comparing pitch selection against the No. 9 hitters in the American League versus pitch selection against hitters first through eighth. I will do this again in the remainder of the article, to check if any of the emerging differences are due to the pitcher being at the plate or they just denote a typical behavior against a back-of-the-lineup hitter.

Table 2a: Pitch selection against same-handed batters
(1-8=American League batters one through eight; 9=American League ninth batters) 

           1-8  9
  Change-up 16% 15%
  Curve    11% 10%
  Fastball 63% 65%
  Slider   10%  9%
Table 2b: Pitch selection against same-handed batters
(1-8=American League batters one through eight; 9=American League ninth batters) 

           1-8  9
  Change-up  6%  5%
  Curve    11% 11%
  Fastball 62% 65%
  Slider   21% 20%

In this case we see a much smaller difference.

Selection by count.

Let’s now see if pitchers use the fastball more frequently in all the counts against their colleagues. Actually, before having a look at the following tables, we may even think that hurlers find themselves more often in fastball counts against a really weak opponent, thus the difference in the pitch selection might just be an artifact.

Table 3a: Pitch selection by count against same-handed batters
(National League non-pitchers; percent by row)
    
      Change-up Curve Fastball Slider
  0-0        5    10       63     22
  0-1        7    13       54     26
  0-2        6    14       50     30
  1-0        7     7       66     20
  1-1        8    11       57     24
  1-2        8    16       48     28
  2-0        6     3       76     15
  2-1        8     7       66     20
  2-2        9    13       53     25
  3-0        3     0       93      4
  3-1        5     2       81     13
  3-2        7     7       64     22

Table 3b: Pitch selection by count against same-handed batters
(National League pitchers; percent by row)
  
      Change-up Curve Fastball Slider
  0-0        5     4       81     10
  0-1        5    12       61     22
  0-2        5    25       43     28
  1-0        5     1       89      5
  1-1        4     6       76     13
  1-2        5    25       42     28
  2-0        5     0       94      2
  2-1        3     1       94      2
  2-2        4     8       73     15
  3-0        7     0       93      0
  3-1        3     0       95      2
  3-2        2     1       93      4

Table 3c: Pitch selection by count against opposite-handed batters
(National League non-pitchers; percent by row)
    
      Change-up Curve Fastball Slider
  0-0       14    12       66      8
  0-1       19    14       53     13
  0-2       14    16       55     15
  1-0       20     6       66      8
  1-1       22    12       53     13
  1-2       17    17       50     16
  2-0       14     3       78      6
  2-1       19     6       65     10
  2-2       18    14       52     16
  3-0        5     0       93      2
  3-1       11     3       81      6
  3-2       14     7       67     12

Table 3d:- Pitch selection by count against opposite-handed batters
(National League pitchers; percent by row)

      Change-up Curve Fastball Slider
  0-0        9     5       82      5
  0-1       16    14       63      6
  0-2       14    27       48     10
  1-0        7     1       90      2
  1-1       15     8       69      8
  1-2       15    26       46     13
  2-0        2     0       96      2
  2-1        6     0       93      1
  2-2       10    11       70      9
  3-0        2     0       98      0
  3-1        1     0       99      0
  3-2        3     1       95      2

The fastball is thrown more frequently in the majority of the counts when pitchers are at the plate, and becomes the only choice when the batter has gone ahead. It also is used four out of five times to open the at-bat against a pitcher, compared to three out of five against the other batters.

Change-ups and sliders are used less frequently in every count.

Finally, pitchers try to strike out their colleagues with the curveball: With a count of either 0-2 or 1-2, you have a chance of one in four to see the Uncle Charlie delivered to a National League pitcher, while the other batters in the senior circuit see the No. 2 around 15 percent of the time in those counts.

Comparing hitters in the ninth slot in the AL against those first to eighth in the order, we don’t see differences of the same magnitude.

Table 4a: Pitch selection by count against same-handed batters
(American League hitters one through eight; percent by row)
    
      Change-up Curve Fastball Slider
  0-0        4    10       66     20
  0-1        6    15       56     23
  0-2        5    18       49     27
  1-0        6     8       68     19
  1-1        8    13       58     22
  1-2        7    18       48     26
  2-0        4     3       83     10
  2-1        6     7       68     19
  2-2        8    14       54     24
  3-0        1     1       96      3
  3-1        3     2       87      8
  3-2        7     6       69     18

Table 4b: Pitch selection by count against same-handed batters
(American League ninth hitters; percent by row)
 
      Change-up Curve Fastball Slider
  0-0        5     8       70     17
  0-1        6    16       56     23
  0-2        5    16       49     30
  1-0        5     4       75     16
  1-1        7    13       58     21
  1-2        7    21       48     25
  2-0        3     1       87      9
  2-1        4     4       77     14
  2-2        8    14       53     25
  3-0        1     1       95      3
  3-1        4     2       91      4
  3-2        4     4       78     14

Table 4c: Pitch selection by count against opposite-handed batters
(American League hitters one through eight; percent by row)
    
      Change-up Curve Fastball Slider
  0-0       14    11       69      6
  0-1       19    14       56     11
  0-2       12    18       55     15
  1-0       19     6       67      7
  1-1       21    12       55     12
  1-2       15    17       50     18
  2-0       12     2       81      5
  2-1       17     6       68     10
  2-2       17    13       54     16
  3-0        4     0       94      1
  3-1        8     2       84      6
  3-2       12     6       69     13

Table 4d - Pitch selection by count against opposite-handed batters
(American League ninth hitters; percent by row)
    
      Changeup Curve Fastball Slider
  0-0       13     9       73      6
  0-1       18    16       56      9
  0-2       11    19       54     16
  1-0       16     4       73      7
  1-1       22    12       55     11
  1-2       15    18       49     18
  2-0        8     1       88      3
  2-1       15     4       74      7
  2-2       17    13       55     15
  3-0        4     0       95      1
  3-1        7     1       90      3
  3-2       12     5       73     10

Location.

Pitchers are obviously less afraid to let a ball—expecially a fastball—go right down Broadway when they face the opposing starter. They also expect their weakest opponents to go fishing on breaking balls low and outside: in fact, higher percentages of curveballs and sliders are delivered out of the zone in that spot when facing pitchers.

Table 5a - Pitch location by pitch against National League non-pitchers (percent by column)
                 
                        Change-up Curve Fastball Slider
  High Inside Ball             2     3        6      2
  High Inside Strike           1     1        2      1
  High Middle Ball             2     2        6      2
  High Middle Strike           2     3        6      3
  High Outside Ball            8     7       10      4
  High Outside Strike          2     2        3      2
  Low Inside Ball              6    10        4      8
  Low Inside Strike            2     2        2      2
  Low Middle Ball             12    10        4      6
  Low Middle Strike            7     7        5      6
  Low Outside Ball            19    17        8     24
  Low Outside Strike           4     4        3      4
  Middle Inside Ball           2     2        5      3
  Middle Inside Strike         3     4        5      4
  Middle Middle Strike        11    13       15     13
  Middle Outside Ball         12     7       10      9
  Middle Outside Strike        7     6        8      7


Table 5b - Pitch location by pitch against National League pitchers (percent by column)
                 
                        ChangeUp Curve FastBall Slider
  High Inside Ball             2     2        4      1
  High Inside Strike           1     1        2      1
  High Middle Ball             2     2        5      2
  High Middle Strike           2     3        6      3
  High Outside Ball            6     5        8      3
  High Outside Strike          1     1        3      1
  Low Inside Ball              5     9        3      6
  Low Inside Strike            2     2        2      1
  Low Middle Ball              9    10        4      6
  Low Middle Strike            8     7        6      7
  Low Outside Ball            19    20        9     29
  Low Outside Strike           5     4        3      5
  Middle Inside Ball           2     2        3      2
  Middle Inside Strike         3     3        5      3
  Middle Middle Strike        13    15       19     14
  Middle Outside Ball         12     7       11      9
  Middle Outside Strike        8     6        9      8

Pitch quality.

I’ve always wondered whether pitchers take something off their offerings against weak opponents. As we can see below, they don’t seem to slow down very much.

A Hardball Times Update
Goodbye for now.
Table 6 - Speed (in mph) by pitch: NP = National League non-pitchers; P = National League pitchers

           NP     P
Fastball 90.8  90.4
Curve    76.4  76.4
Slider   83.6  83.3
Change-up 81.6  82.1

A small decrease in speed occurs for the “hard” pitches (fastball and slider). Though the difference ends up being statistically significant due to the large number of pitches in the PITCHf/x database I’m not inclined to consider it really meaningful – I don’t buy that pitchers can consciously take less than half a mile off of a pitch, but I might be wrong.

The change-up comes a little faster toward batting pitchers, but again I’m not sure I would consider the difference as interesting. Anyway, the combined effect is that the difference in speed between the fastball and the change-up is about one mile per hour smaller when a pitcher is at the plate.

Finally, let’s have a look at movement.

Table 7a - Horizontal movement (in inches) by pitch:
NP = National League non-pitchers; P = National League pitchers

            NP      P
Fastball -6.38  -6.27
Curve     4.88   6.94
Slider    2.05   2.27
Change-up -6.57  -5.77
Table 7b - Vertical movement (in inches) by pitch:
NP = National League non-pitchers; P = National League pitchers

            NP      P
Fastball  8.38   8.67
Curve    -4.75  -4.76
Slider    2.90   0.48
Change-up  5.16   5.84

There are a few things that are striking me in this final couple of tables. The curveballs thrown at the pitchers trying to help themselves in the batter’s box break a lot more on the horizontal plane, while the sliders break a lot less on the vertical plane. It is worth noting that those pitches usually rely for their effectiveness on the break on the other plane.

It looks like the men on the mound use a different kind of curve against their colleagues—more of the sweeping kind. I don’t know whether the reduced ride on the slider means the thrower is going easier; that being the case, shouldn’t we see a similar effect for the fastball?

I must admit I expected to find an answer on whether pitchers refrain a bit when facing their weak-hitting peers, mostly among the lines of the last couple of tables. But while they show some different behaviors when a pitcher is at the plate, I’m not currently able to produce a convincing interpretation of those differences.

Surely, by going a lot less to the slider (as we have seen in the previous paragraphs), they avoid some of the strain to their elbow.

References & Resources
The values you see in the last section are not actual averages. Had I used actual averages we would have been exposed to potential confounding: Let’s suppose that those pitchers whose slider doesn’t ride are the ones who don’t relinquish that pitch against their colleauges. Then we would see numbers similar to those appearing in the last table, but we wouldn’t be able to make a statement that pitchers throw the slider differently against their peers.

To work around this problem, I rescaled all the values in the database; i.e. I artificially gave all the pitchers pitches with the same average behavior.
I’ll make this clearer with an example.

The average major league fastball speed is 90.7 mph. Brad Penny throws his fastball at an average speed of 93.2 mph; thus I rescaled the velocity of every Penny fastball by multiplying it by 90.7/93.2= 0.97.
Horizontal and vertical movement values went through the same process.


Comments are closed.