This is something that has been bouncing around in my mind for a while; I've been wondering how much of an effect a catcher has on a pitcher's performance. Everyone knows about a catcher's ERA (CERA) and how ineffective of a metric it can be mainly because one guy will consistently catch a team's ace, while the other is stuck catching the #5 guy in the rotation. This is also known as the Mathis-Napoli Effect. I suppose it is possible to judge a catcher's skill by comparing apples-to-apples whenever two catchers catch the same pitcher, but then we get into the strength of opposition that will factor into the numbers.
Recently, there has been some great information coming from the pitch-framing skills using pitch/FX data and called strikes, but I was more interested in the game calling effect that a catcher has on his pitcher's performance. My thinking being pitch sequence and the use of a pitcher's strength, that is using the best pitch more often, or at better times, would have just as large of an effect on the game outcome that framing has. Or maybe even more.
At one time in my life, I used to catch. Nothing significant, just amateur leagues around Orange County. In 1989, a league formed in Florida comprised of former professional players. Former Angel and Brewer relief pitcher Tom Murphy was heading to Florida to play in this new league, but needed to get back into shape, and decided to join our team to get in some innings. It was a pretty big thrill for me to be able to catch a former big-league pitcher, but I was intimidated by my game-calling skills, or rather, my lack of skills. What if I consistently called for pitches he didn't want to throw? However, I was surprised during the few games he pitched for us, he never shook me off. Not once. I'm sure it wasn't because I was calling a great game, but because he probably didn't care what he threw, just as long as he was throwing.
When I started thinking about his study, those days behind the plate caused me to wonder, "Do major league pitchers throw what they want, or do they throw what the catcher wants?" We've all seen a pitcher shake-off a catcher, but how often does that happen and how often does he just throw what's called? Was Murphy in the habit of just looking in for the sign and throwing the called pitch? If this was the case for most pitchers, the catcher's game-calling skills would have a huge impact on the game outcome.
It would be next to impossible to compile the number of time a pitcher shakes off his catcher, but using pitch/FX data, we can count the types of pitches thrown and then separate those pitches by the catcher calling them. The first pitcher I was interested in looking at was C.J. Wilson. To me, Wilson seemed like he was either throwing a great game, or throwing a large number of borderline pitches resulting in inconsistent outcomes. Here are the numbers for Wilson, separated by catcher for this season:
Fourseam |
Sinker |
Change |
Slider |
Curve |
Cutter |
|
Iannetta |
44.0% |
14.6% |
5.6% |
10.2% |
15.0% |
10.6% |
Conger |
36.4% |
11.4% |
5.9% |
23.1% |
11.8% |
11.4% |
And here's his game outcomes:
ERA |
H/IP |
BB/IP |
SO/IP |
BF/IP |
P/IP |
S/IP |
|
Iannetta |
3.92 |
1.01 |
0.46 |
0.80 |
4.44 |
17.86 |
10.74 |
Conger |
3.28 |
0.92 |
0.36 |
0.99 |
4.26 |
17.13 |
11.04 |
BF/IP = Batters faced per innings pitched
P/IP = Pitches per inning
S/IP = Strikes per inning
As you can see, Wilson throws more sliders and less fastballs when Conger is behind the plate. The big question is, is that because Conger calls for more sliders or does Wilson like throwing more sliders when Conger is catching? Whichever the reason, the outcome shows a better performance with Conger, but remember, this data doesn't take opponent data into consideration, however, I think those numbers tend to balance out over the course of a season.
Next was Jered Weaver:
Fourseam |
Sinker |
Change |
Slider |
Curve |
Cutter |
|
Iannetta |
28.0% |
30.4% |
15.0% |
13.8% |
12.8% |
0.0% |
Conger |
29.4% |
26.1% |
14.3% |
12.4% |
17.7% |
0.0% |
ERA |
H/IP |
BB/IP |
SO/IP |
BF/IP |
P/IP |
S/IP |
|
Iannetta |
2.88 |
0.80 |
0.30 |
0.87 |
4.07 |
15.92 |
9.85 |
Conger |
2.78 |
0.99 |
0.15 |
0.74 |
4.05 |
16.21 |
10.36 |
Weaver's data is interesting because it shows similar numbers for both catchers. There's a slight difference in the number of curves and sinkers thrown, but not at the same rate as with Wilson. His outcome is relatively the same too, the only somewhat significant difference is the number of walks allowed.
The other pitcher I was interested in looking at was Blanton. Of the pitchers on the Angels staff, Blanton seemed like the most inconsistent of the bunch.
Fourseam |
Sinker |
Change |
Slider |
Curve |
Cutter |
|
Iannetta |
22.8% |
25.8% |
26.4% |
6.7% |
9.8% |
8.5% |
Conger |
16.8% |
28.9% |
32.0% |
4.5% |
9.8% |
8.0% |
Once again, the data shows Iannetta calls for more fastballs and less change-ups than Conger. The outcome difference is pretty significant:
ERA |
H/IP |
BB/IP |
SO/IP |
BF/IP |
P/IP |
S/IP |
|
Iannetta |
5.90 |
1.36 |
0.23 |
0.78 |
4.59 |
16.49 |
10.74 |
Conger |
4.97 |
1.34 |
0.28 |
1.00 |
4.59 |
17.17 |
11.34 |
Blanton's ERA is almost a whole run less with Conger, in spite of throwing slightly more pitches per inning.
So what's all of this tell us? Probably nothing. Without knowing who is truly making the pitch selection, the numbers shown here and the conclusions we try to make using them is completely speculative. However, if I were to make an educated guess, this data does show that a catcher's game-calling skill does have an impact on game outcome. And if I were to guess some more, this data shows Weaver takes more control over his game by throwing what he wants, while Wilson and Blanton seem to throw the catcher's pitch more often. This results in Wilson and Blanton pitching slightly better when Conger is behind the plate.