Defensive Catchers, Part I

July 24, 2020
  

Catchers

We come, at last, to Fielders, starting with the men behind the Iron Masks.  This is basically a system for fielders, but we had to put in 100 pages of work on pitchers to get here. 

Catchers will be credited with Runs Saved for eight categories of on-field performance:  strikeouts by their pitchers, walks avoided by their pitchers, Wild Pitches and Passed Balls Avoided, Stolen Bases Avoided, Runners Picked Off and Caught Stealing, and Error Avoidance.   That’s eight performance areas, but since "Baserunner Advancement" includes three different things and "Runners Removed" includes two, there will only be five different values in the catcher’s mitt, whereas there were 10 for pitchers.   And the explanations for those five should be much easier, since we have already explained essentially the same concepts in outlining the pitcher values.   This chart, which I showed before, is intended to help you get an overview of the process in your head:

 

Category

P

C

1B

2B

3B

SS

LF

CF

RF

Strikeouts

97%

3%

---

---

---

---

---

---

---

Control

97%

3%

---

---

---

---

---

---

---

HR Avoidance

100%

---

---

---

---

---

---

---

---

Balks

100%

---

---

---

---

---

---

---

---

Wild Pitches

70%

30%

---

---

---

---

---

---

---

Passed Balls

35%

65%

---

---

---

---

---

---

---

Stolen Bases All

40%

60%

---

---

---

---

---

---

---

Outfield Assists

---

---

---

---

---

---

30%

30%

40%

Pitcher Pickoffs

100%

---

---

---

---

---

---

---

---

Catcher Pickoffs

---

100%

---

---

---

---

---

---

---

Runners Caught Stealing

40%

60%

---

---

---

---

---

---

---

Double Plays

16%

---

10%

42%

12%

40%

---

---

---

Error Avoidance

10%

10%

10%

15%

17%

22%

5%

6%

5%

DER

16%

---

10%

13%

10%

14%

11%

15%

11%

 

 

Formula 45:  C-RS-K-C1   (Catcher’s Runs Saved on Strikeouts)

The formula for Catcher’s Runs Saved on Strikeouts is

(a)  The Team’s strikeout total,

(b)  Times the catcher’s putouts (at catcher),

(c)   Divided by the team total of Catcher’s putouts,

(d)  Times .166 (the run-prevention value of a strikeout),

(e)  Times .03. 

 

That is:

C-RS-K-C1 = TmSO * PO(ind-c) / PO(tm-c) * .166 * .03

 

Where PO(ind-c) means the putouts at catcher by an individual catcher, and PO(tm-c) means the team total.   These would be the first leaders among the catchers on these 15 teams.

 

City

Team

Year

Player

Tm K

Poc-tm

Poc-ind

C1

Detroit

Tigers

1968

Bill Freehan

1115

1196

971

4.51

New York

Yankees

2000

Jorge Posada

1040

1076

892

4.29

Toronto

Blue Jays

1992

Pat Borders

954

973

784

3.83

Detroit

Tigers

1984

Lance Parrish

914

956

720

3.43

Philadelphia

Phillies

2008

Carlos Ruiz

1081

1120

623

2.99

 

Formula 46:  Cat-Control (Catcher’s Contribution to Control, or walks avoided)

 

            The TEAM’S walks avoided are as follows:

 

            W=Av = BFP * .145 067 – BB – HBP

 

            That is NOT Formula 46, above; that’s just the start of it.   That part is actually exactly the same as Formula 2, back at the start of this series, except that at that time we were applying it to individual pitchers, and now, because we lack strikeout data for individual catchers, we are applying it to the team.  Anyway, having done that, we pro-rate that to each individual catcher, based on his innings at the position:

 

            Cat-Control = (BFP * .145 067 – BB – HBP) * (C-Inn / Tm-Inn) * .03

 

            Where C-Inn is the Catcher’s Innings, and Tm-Inn is the team’s Innings.           

 

 

Formula 47:  RS-Cat-Control-C2  (Runs Saved by Catcher-Control-2nd Catcher Value)

 

Credit the catcher with .236 Runs Saved for each walk not issued.

            Cat-Control-C2 = W-Av * .236

            These are the top five catchers in the study in terms of Runs Saved by Control:

Player

BFP

BB

HBP

Tm IP

Inn-C

C2

Bill Freehan

6043

486

32

1489.7

1180.3

2.01

Johnny Bench

6191

491

21

1471.0

1071.0

1.99

Lance Parrish

6127

489

30

1474.0

1075.7

1.91

Pat Borders

6108

541

45

1440.7

1160.7

1.71

Smoky Burgess

5805

386

11

1399.7

748.3

1.68

 

 

            And these are the Top 5, combining C1 and C2:

 

 

Year

Player

C1

C2

Total

1968

Bill Freehan

4.51

2.01

6.52

2000

Jorge Posada

4.29

1.64

5.93

1992

Pat Borders

3.83

1.71

5.54

1984

Lance Parrish

3.43

1.91

5.34

1976

Johnny Bench

2.87

1.99

4.86

 

 

            But the numbers here are so small—five to seven runs for the top catchers—that these categories are mostly just a way of acknowledging the catcher’s contribution.   Our ability to actually MEASURE the catcher’s contribution, throughout history, is very limited at this time, although ways could be invented to get closer to it. 

 

 

 

Formula 48: BaSv-Cat (Bases Saved by the Catcher)

           

            Formulas 48, 49 and 50, which outline Runs Saved by the catcher on Base Advancement, are a reprise of Formulas 7, 8 and 9, which outlined very similar calculations for the pitcher.   Formula 7 was this:

 

              A pitcher’s "Budget" for One Base Advancement Events is  .037123 times his Batter’s Faced. 

              OBAE-Budget = .037123 * BFP

 

            Formula 48 is essentially the same, except that we don’t have BFP data for catchers, so we have to modify it as follows:

            OBAE-Budget (c) = .037123 * TmBFP * Innings (catcher) / Innings (team). 

 

In other words, for Johnny Bench in 1976, the Team  BFP was 6,191.  Bench caught 1,071 innings out of a team total of 1,471, so his Budget for One Base Advancements Allowed is .037123 * 6191 * 1071 / 1471, which is 167.83.  Bench will receive credit for bases saved based on how far below that number he is in Base Advancements Allowed. 

 

 

 

 

           

 

 

              Formula 49:  BaSv-Cat  (Bases Saved by the Catcher)

 

 

            This was formula 8, which applied to pitchers:

              BaSv-Pit = [(OBAE-Budget) – Bk - .70 * WP - .40 * OSB - .35 * PB] / 2

 

 

Since the pitcher is held 100% responsible for Balks, 70% responsible for Wild Pitches, 40% responsible for stolen bases allowed, and 35% responsible for Passed Balls, the catcher is held not at all responsible for Balks, 30% responsible for Wild Pitches, 60% responsible for Opposition Stolen Bases, and 65% responsible for Passed Balls:
            BaSv-Cat = [(OBAE-Budget) - .30 * WP - .60 * OSB - .65 * PB]/2

 

I see on editing this for the 40th time that I should have spelled out that each catcher is held responsible for the TEAM’S wild pitchers, based on his innings as a catcher.   We have data for Passed Balls for each catcher, but not data for Wild Pitches for each pitcher/catcher combination.   We have to pro-rate the Wild Pitches based on catcher innings. 

These are the five catchers in the study who we credit with preventing the most Base Advancements by Wild Pitch, Passed Ball, or Stolen Base (C3):

Team

Year

Player

Budget

WP

OSB

PB

Ba Sv

C3

Yankees

2000

Jorge Posada

193

49

70

11

65.7

24.5

Tigers

1968

Bill Freehan

178

38

66

9

61.6

23.0

Tigers

1984

Lance Parrish

166

47

44

11

61.1

22.8

Reds

1976

Johnny Bench

167

43

57

5

60.2

22.5

Pirates

1960

Smoky Burgess

115

25

22

2

48.4

18.0

 

Posada being first on the list is surprising, since Posada did not have a great throwing arm, and was not regarded as a great defensive catcher.   The result here is heavily influenced by playing time; Posada caught more innings than any other catcher in the study.  If you look at his Runs Saved (C3) relative to his Budget (which is based on playing time), you can see that Posada actually has a LOWER rate of Runs Saved, relative to the budget, than any of the other four catchers listed above—a lower rate, but a higher total.  

But also, while Posada DID have some very poor defensive seasons later on, his 2000 season data is actually pretty good.  He gave up 70 stolen bases while catching almost 1200 innings.  It’s actually a pretty decent ratio.  His record against the running game in his early years is actually not bad.   And at this time, Posada is credited with saving more runs than any other catcher in the study:

City

Team

Year

Player

C1

C2

C3

Total

New York

Yankees

2000

Jorge Posada

4.3

1.6

24.5

30.4

Detroit

Tigers

1968

Bill Freehan

4.5

2.0

23.0

29.5

Detroit

Tigers

1984

Lance Parrish

3.4

1.9

22.8

28.1

Cincinnati

Reds

1976

Johnny Bench

2.9

2.0

22.5

27.3

Toronto

Bl Jays

1992

Pat Borders

3.8

1.7

16.8

22.3

 

 

 

 

 

 

             

Formula 50:  BaSv-RV-C3  (Bases Saved Run Value, 3rd Catcher’s Value) 

            Each Base Saved by the Catcher has a Run Value of .373 Runs.

            BaSv-RV-C3 = BaSv-Cat * .373

 

            Formula 51:  BRR-1-Catcher (Baserunners Removed by a Catcher)

            A catcher gets credit for a Baserunner Removed when he (a) picks a runner off base, in which case he gets 100% for that baserunner removed, or (b) throws a runner attempting to steal, in which case he gets 60% of the credit, the other 40% going to the pitcher:

            BRR-1-Catcher = PkOff(c) + OCS

            This is a "positive" category in which high numbers are good, so it’s easy to figure. 

 

            Formula 52:  RR1-RS- Cat-C4   (Runners Removed-1, Runs Saved, Catchers, 4th Catcher’s Value)

            Each baserunner removed has a value of .516 runs:

            RR1-RS-Cat-C4 =  BRR-1-Catcher * .516

           

            These are the five catchers credited with the most Runs Saved by Baserunner Removal:

City

Team

Year

Player

CS

PkOff

BRR-1

C4

Toronto

Blue Jays

1992

Pat Borders

51

2

32.6

16.8

Cincinnati

Reds

1976

Johnny Bench

42

3

28.2

14.6

Detroit

Tigers

1984

Lance Parrish

38

3

25.8

13.3

Detroit

Tigers

1968

Bill Freehan

38

2

24.8

12.8

New York

Mets

1964

Jesse Gonder

33

3

22.8

11.8

 

            Let me explain something.   The catcher who plays in a low-stolen-base era has an advantage in Base Advancements Prevented.  Since he allows fewer stolen bases, he will be credited with NOT ALLOWING more stolen bases.   But the catcher who plays in a high-stolen-base-attempts era has the advantage in baserunners removed.   If a catcher allows 50 stolen bases and throws out 30 runners, he’ll show up strong in "Base Advancements", but do poorly in Baserunners Removed.  If he allows 100 stolen bases but throws out 60 runners, he’ll do poorly in Base Advancements, but do extremely well in Baserunners Removed.   Thus, Smoky Burgess (1960) was on the leaderboard in "Base Advancements Prevented), but disappears from the board in "Baserunners Removed", while Pat Borders (1992), who was not on the board in "Base Advancements", moves to the top of the list in "Baserunners Removed".  Overall, the system is fair to both eras; it just depends on the total of the two values. 

            These are the updated leaders in Catchers, Total Runs Saved:

Team

Year

Player

C1

C2

C3

C4

Total

Tigers

1968

Bill Freehan

4.5

2.0

23.0

12.8

42.3

Reds

1976

Johnny Bench

2.9

2.0

22.5

14.6

41.9

Yankees

2000

Jorge Posada

4.3

1.6

24.5

11.0

41.5

Tigers

1984

Lance Parrish

3.4

1.9

22.8

13.3

41.4

Blue Jays

1992

Pat Borders

3.8

1.7

16.8

16.8

39.1

 

 

                     We'll finish the catchers tommorrow.  

 
 

COMMENTS (14 Comments, most recent shown first)

DavidHNix
I had a Smoky Burgess model catcher's mitt. Dont recall if it was Rawlings or Wilson. It was like a plywood slab with a little dimple in the center for a pocket. If that's what he used, it's no wonder he had a bad defensive reputation.
1:52 PM Jul 27th
 
W.T.Mons10
Did a little checking. I took Jorge Posada from 2000-2003. He caught over 1000 innings in each of those seasons. His % of POs vs. his % of innings caught were as follows:
2000 82.9 vs. 83.0
2001 76.7 vs. 76.6
2002 82.5 vs. 82.1
2003 81.6 vs. 79.7


Applying the walks saved formula to 2003, which had the biggest difference, I get approximately 2.8 runs prevented using % of POs and 2.6 runs prevented using % of innings caught.

Looking at 1997, when he only caught 479 innings, the %s are 30.5 and 32.7; the runs prevented are .73 and .78. Is that worth the extra work?
7:35 AM Jul 26th
 
CharlesSaeger
You can't use putouts as a measure of walks allowed. They're unrelated, and furthermore, putouts are a measure of success, while walks are a failure.
7:21 PM Jul 25th
 
markkolier
The inclusion of Posada in 2000 surprised me even after understanding he was better defensively early in his career. Makes me wonder how much catching Andy Pettite (not to mention Clemens, El Duque and Cone) and his threat as a pickoff artist deterred base stealers.
12:50 PM Jul 25th
 
bjames
On the strikeouts and walks. . .it would have been simpler, yes, but less accurate. A strikeout by the pitcher is recorded as a putout by the catcher, so it is obvious that putouts have to be used to estimate strikeouts by a pitcher, and obvious that that is an extremely accurate estimate. But there is no reason to believe that putouts would be as accurate to gauge the share of walks as innings.
10:29 PM Jul 24th
 
W.T.Mons10
Well, I should have asked if the different methods produce a meaningfully different result. If not, wouldn't it be simpler just to use the same proportions for both walks and strikeouts? This is complicated enough as is.
8:28 PM Jul 24th
 
MarkBernstein
I remember the old Smokey Burgess, but I have a dim memory that people said he'd been a very different player when younger.
8:13 PM Jul 24th
 
CharlesSaeger
W.T.Mons10: Not Bill, but likely because he really wants to allocate strikeouts by strikeouts and putouts are pretty much strikeouts.
7:28 PM Jul 24th
 
willibphx
Top five is close. I will be interested to see if there is any real separation between the "good" and "bad" catchers when all the primary catchers are seen.
5:47 PM Jul 24th
 
W.T.Mons10
Minor point, I guess, but why are you allocating strikeouts to individual catchers by the share of team putouts they had but walks by the share of team innings caught?
4:57 PM Jul 24th
 
SteveN
I was surprised to see Smokey Burgess's name come up a couple times in the positive categories. That is not his reputation.
3:56 PM Jul 24th
 
bjames
46/47 is awkwardly stated and somewhat misleadingly stated. I'll re-work the way it is stated in the file.
3:18 PM Jul 24th
 
CharlesSaeger
I'm looking at formula 47: from where are you getting W-Av? Is this the same one you just used in formula 46? Didn't you just credit the catcher for not allowing walks in that formula?
2:04 PM Jul 24th
 
evanecurb
I have nothing to add to the process here, other than to cheer Bill on. I hope this project will open up a whole new way to look at quantifying defense. It's likely that this method will lead other researchers to build on it and refine it, just as Bill suggests. It's also possible that somebody will find a big hole in it - an incorrect assumption or something - but one that turns out ultimately to be correctable, then the path would continue after the correction is made. In either case, a foundation has been laid here.
1:17 PM Jul 24th
 
 
©2020 Be Jolly, Inc. All Rights Reserved.|Web site design and development by Americaneagle.com|Terms & Conditions|Privacy Policy