Strikeout Tigers

March 14, 2020
 

Strikeout Tigers

 

            Strikeout totals go up all the time, so the team record for strikeouts is always or very often some number from the last few seasons.  Entering our current century the record for strikeouts by a team’s pitchers was held by the 1997 Atlanta Braves, with three Hall of Famers in their starting rotation.   I could name them, but then, so could you.  The fourth starter was Denny Neagle, who was 20-5 with a 2.97 ERA and 172 strikeouts in 223 innings. 

            That record for strikeouts by a team was broken by the 2001 Red Sox, whose starting pitchers were Pedro Martinez, Hideo Nomo, Tim Wakefield, David Cone and Frank Castillo.  Nomo led the league in strikeouts, with 220, since Pedro was hurt and missed half the season.  The Sox struck out 1,259 batters, a total which was an all-time record then, but is no longer one of the 150 highest totals of all time.  Their record was broken by the 2003 Chicago Cubs, a first-place team with a starting rotation of Kerry Wood, Mark Prior, Carlos Zambrano, Matt Clement and Shawn Estes.  Wood struck out 266 batters, which led the league, while Prior had 245, which was second.

            Their strikeout record lasted ten years, remarkably enough, but was broken by the 2013 Detroit Tigers, with a starting group of Justin Verlander, Max Scherzer, Anibal Sanchez, Rick Porcello and Mister Fister.  Their record was broken by the Cleveland Indians in 2014, and also by the Tampa Bay Rays, although I guess the Rays wouldn’t count; they were the Sammy Sosas of the team strikeout record, the also-rans who also broke the old record.  The Indians’ 2014 record was broken by the Dodgers in 2016, and also the Nationals.  The Dodgers 2016 record was broken by five different teams in 2017, led by Cleveland, and then Cleveland’s record was broken in 2018 by the Astros and also the Yankees.  In 2019 there was no new record, and there will be no new record in 2020, because of the Coronavirus, although as of now the 19 highest totals of all time were all from the years 2017 to 2019.  So the 2018 Astros’ team record for strikeouts is now an ancient and venerated record. 

             When Leon Spinks defeated Muhammad Ali for the Heavyweight Boxing Championship in 1978, he said of Ali that "He’s still the greatest.  I’m just the latest."  So it is with the strikeout record; whoever holds it is just the latest, but the greatest ever was

            ???

            Any ideas?

            How about the 1946 Detroit Tigers?   I’m not saying that this is absolutely the correct answer, but I’ve got a methodology, and that is the answer it comes up with.  The Tigers, who finished second with a record of 92-62, were led by Hal Newhouser, then finishing up a 3-year run as the best pitcher in baseball.  The American League’s MVP in 1944 and 1945, in 1946 he went 26-9 with a 1.94 ERA, 275 strikeouts.   The Tigers’ four-man rotation was Newhouser, Dizzy Trout, Virgil Trucks and Fred Hutchinson.   Trout had been second in the MVP voting in 1944, losing to Newhouser by a vote of 236-232; actually Trout had more first-place votes than Newhouser did, 10 to 7.   Trucks, known has "Fire" Trucks, had struck out 418 batters as a 19-year-old in the minor leagues, while Hutchinson, remembered as a Hospital in Seattle, is perhaps the greatest control pitcher of all time.   Newhouser, Trucks, Trout and Hutchinson were 2nd, 4th, 5th and 6th in the league in strikeouts.

            The Tiger pitchers struck out 896 batters, some sources say 898; in any case it was a major league record at the time.  (This was the same team, incidentally, for which Roy Cullenbine had a .477 on base percentage.  He was used as a fourth outfielder, and batted 6th when he was in the lineup.)   Anyway, the argument for the Tigers’ being the greatest strikeout team of all time is this.   The 1946 Tigers had 5,877 batters facing pitcher, and struck out 896 of them.   That is 15%--actually, .15246.   The norm for all pitchers in the years 1940 to 1949 was .09216.  The standard deviation of team strikeouts in those same years was .01445.   The 1946 Tigers thus were 4.1 Standard Deviations above the norm for the Decade.  This is the highest figure of all time—hence, the 1946 Tigers were the greatest strikeout battalion ever assembled. 

            There are problems with this method, as there are problems with every method.  The 1946 American League was a high-strikeout league compared to the decade; thus, the Tigers would be less impressive if compared to their own league than if compared to the decade.  But there are problems with comparing the team only to their own league, too.  Eight teams is a small group of teams.  Standard Deviations are not designed to describe small populations, like "eight", and don’t work well in them.  Leagues, particularly then but still now, are small enough groups that individuals who may be historic outliers can be artificially normalized by comparing them to norms that they contributed heavily toward creating.  Two pitchers—Newhouser and Bob Feller—accounted for 12% of the league strikeouts in 1946.  Feller in that season set what was at the time regarded as an American League strikeout record, with 348 (subsequent research has shown that Rube Waddell may have struck out 349 in 1904.  The record listed at the time was 343.)  Newhouser struck out more batters as a percentage of batters faced and more per inning than Feller did.  Newhouser, Trucks and Hutchinson were 1st, 3rd and 4th in the league in strikeouts per inning, with Feller second.

            My point is that the standard deviation relative to the LEAGUE is problematic, because the league is a relatively small group of teams, and the Tigers’ four strikeout kings have a large impact on that group.   The standard deviation relative to the DECADE is also problematic, for different reasons, but in my view it is less problematic.  Using the decade as the standard, the 1946 Tigers are the most outstanding strikeout team of all time. 

            The 1946 Tigers are the only team which was Four Standard Deviations above the norm; we will represent 4.1 standard deviations above the norm as 141, and you can probably figure that out.  There are eight teams in history which were at least three standard deviations above the decade norm for strikeouts.  Those are:

1.      1946 Detroit—Newhouser, Trout, Trucks and Hutchinson

2.     1959 Dodgers—Koufax and Drysdale; oddball World Championship although the team was not great; 3.6 standard deviations above the norm (136).

3.     1924 Dodgers—Dazzy Vance, the MVP with a 28-6 won-lost record, had 262 strikeouts.  The only other pitcher in the league who had even one-third that many was his teammate, Burleigh Grimes, who had 135.   3.4 standard deviations above the norm (134).

4.     2018 Astros—(133).  Verlander, Gerrit Cole and Charlie Morton.  Red Sox cleaned their clock in the League Playoffs.

5.     1971 Mets—(133).  Tom Seaver, Jerry Koosman and a young kid named Nolan Ryan, who was traded to California after the season.

6.     1905 Philadelphia A’s—(132).  Rube Waddell and Eddie Plank were 1-2 in the majors in strikeouts.   Also lost in the World Series.

7.     2003 Cubs—(131).  Wood and Prior.

8.     2019 Astros—(131). 

 

Ninth and 10th on that list, if anybody cares, would be the 1904 Philadelphia A’s (129) and the 2017 Cleveland Indians (129). 

 

This research is actually part of a larger project.   The goal of the project is to estimate the runs saved by fielders and pitchers, runs saved against zero as opposed to runs saved against the league average.   It’s a long story, and we’ll get into that later, possibly, God willing and the creeks don’t rise and the Coronavirus doesn’t kill us all.  Anyway, what I really need to know, for that project, is not what the BEST teams are, but what the WORST teams are.  I need to know where the floor is, where the bottom is. 

The WORST strikeout team of all time, relative to their era, is the 2003 Detroit Tigers.  That was the team that went 43-119; you probably remember them.  Their three top starters were Mike Maroth (6-21, 5.73 ERA), Jeremy Bonderman (6-19, 5.56 ERA), and Nate Cornejo (6-17, 4.76).  Bonderman led the team in strikeouts with 108.   Bonderman was a 20-year-old rookie who was pulled out of the rotation late in the season so that he wouldn’t lose 20 games.  The team struck out 764 batters, which would have been a good total in the 1950s, but which was 2.8 Standard Deviations below the norm in their own decade (72).  

The most interesting team on the low-strikeout side, by far, is Bambi’s Brewers.   The 1977 Milwaukee Brewers were 67-95, a mere 33 games out of first place.  They hired as their manager that winter George Bamberger, a highly respected pitching coach.   Under Bamberger’s guidance the team ERA dropped by 57 points in 1978, and the team went 93-69.   They followed that up with several more excellent seasons, finally making the World Series in 1982, although that was two managers later. 

Anyway, what I never realized until now—what I don’t think anyone had realized—was the quite remarkable extent to which Bamberger turned around the pitching staff by eschewing strikeouts, and focusing on getting rid of walks.   This will show up on the chart below; these are the "worst" strikeout teams of all time:

2003 Tigers              72  (2.8 Standard Deviations below the decade norm)

1992 Tigers              74

1950 St. Louis Browns                  75

1918 Philadelphia A’s                    75

1980 Milwaukee Brewers               76

1979 Milwaukee Brewers               76

1978 Milwaukee Brewers               76

2002 Detroit Tigers                       77

1961 Washington Senators             77 (A first-year expansion team)

1991 Detroit Tigers                        77

 

Those are all American League teams, in part because of the DH Rule; I didn’t normalize the DH Rule out of existence, again believing that that does more harm than good.  The worst National League teams ever were the 1961 Milwaukee Braves (78) and the 1981 Cardinals (78). 

Anyway, back to the Brewers.  Most of these were terrible teams or at least not very good teams, but the Brewers are on the list not because they were bad, but as an option.  They CHOSE not to use strikeout pitchers.  Their leading non-strikeout pitcher was Lary Sorensen.  In 1978 Sorensen was 18-12 with a 3.21 ERA, but struck out only 78 batters in 281 innings.  I knew that Sorensen could not have a long, successful career with that extraordinarily low strikeout ratio.  NOW this would be common knowledge, but at that time I was the only person who knew that, the only person who had studied it.  My argument that Sorensen would not be able to sustain his success was not a popular argument, and especially was not a popular argument with his agents, the Hendricks Brothers, who I worked with and worked for.  They didn’t want to hear it, and they didn’t believe it. 

The Brewers choosing to eschew strikeouts in order to avoid walks. . .would that work, in 2020?  I doubt that it would.  It might; I don’t know.  I doubt it.  At that time walks were probably more important than strikeouts in the success of a pitcher.  Now, because strikeouts are so much more common, that is clearly not true.  

The best strikeout teams ever are 3+ standard deviations above the norm, but no team is 3 standard deviations BELOW the norm.  Talent in major league baseball is not normally distributed; it is the right-end tail of the bell-shaped curve.  The effects of this on the team strikeout distribution curve are slight, but certainly detectable. 

Well. . . .I’d better put on a record a little stuff about the decade norms before I sign off here.   In the years 1900 to 1909, the norm for strikeouts per 1000 batters faced was 96.  This is how it has changed over the decades:

 

From

To

Strikeout Rate

   

1900

1909

96

per 1000 batters

1910

1919

101

 

 

1920

1929

72

 

 

1930

1939

85

 

 

1940

1949

92

   

1950

1959

114

   

1960

1969

151

   

1970

1979

135

   

1980

1989

140

   

1990

1999

159

   

2000

2009

170

   

2010

2019

206

   

 

So there have been two decades in which strikeouts have gone down, and nine in which they have gone up.    This chart includes the Standard Deviations by decade:

From

To

Strikeout Rate

Standard Deviation

1900

1909

96

21.1

1910

1919

101

17.7

1920

1929

72

10.7

1930

1939

85

15.6

1940

1949

92

14.5

1950

1959

114

17.3

1960

1969

151

18.4

1970

1979

135

16.6

1980

1989

140

19.2

1990

1999

159

18.4

2000

2009

170

18.0

2010

2019

206

23.8

 

The standard deviation is high for the last decade because the strikeout totals went up so much during the decade that the norms for the end of the decade are different than the norms for the start of the decade.   Thanks for reading. 

 

 

 
 

COMMENTS (11 Comments, most recent shown first)

raincheck
Re: the Brewers. I think there is usually some benefit to a bad team going against the grain. Everyone is looking for strikeout pitches and the ones available to a bad team with no assets to trade and no cash to spend may be pretty bad strikeout pitchers.

So you look for an asset that is not in favor, in this case “pretty good” contact pitchers. So maybe you can go from bad pitching to pretty good pitching and achieve middling mediocrity. Which is a huge step up. You won’t find a staff of great contact pitchers, because such a thing doesn’t exist. But you can improve by shopping in a market no one is in.
12:38 PM Mar 24th
 
jrickert
I remember those Brewers teams. There was some awareness at the time that Bamberger was having the pitchers try to get groundball contact - though more attention was paid to the increase in complete games (38->62). The pitching staffs had never been strikeout(SO) staffs - they were last in SO the in 1977, before Bamberger, so the drop in strikeouts form 14th place by 18 SO to 14th place by 53 SO. shift from 11th in walks allowed to 1st in walks allowed was more widely noted.
The Brewers highest strikeout finish in Milwaukee before Bamberger was 8th - The Pilots had been 3rd. The Brewers finished last 4 times in SO in the 8 years before Bamberger. The Brewers didn't do better than 8th in SO until a 6th place finish in 1986. (tl;dr Bamberger accentuated a philosophy that was already in place, leading to historically low SO levels)
The Brewers young pitchers, Lary Sorenson, Jerry Augustine, and Bill Travers were all low strikeout pitchers who were perceived at the time as promising your pitchers who would form the foundation of the staff for many years.
7:41 PM Mar 16th
 
bjames
I don't think there is a correct solution.


That's right. Neither solution is perfect.
9:45 PM Mar 15th
 
W.T.Mons10
You've got more or less the same problem that you run into with park factors. If you use one year factors, there's a lot of randomness. But if you use multi-year factors, then you have to deal with the fact that some of the parks have probably changed. I don't think there is a correct solution.
8:29 PM Mar 15th
 
bjames

Don Coffin

I have no particular conclusions...


I appreciate your research all the same.
5:13 PM Mar 15th
 
doncoffin
"Two pitchers—Newhouser and Bob Feller—accounted for 12% of the league strikeouts in 1946."

In 2019, the 10 pitchers accounted for 6% of the strikeouts in MLB.

I took a little broader look at this. The top 2 in 1946 would be (roughly) equivalent to the top 4 today. Looking at both leagues, not just the AL, we'd compare the top 4 (two from each league) to the top 8 today.

In 1946, the top 4 strikeout pitchers in MLB (Feller, Newhouser, Tex Hughson, Virgil Trucks) accounted for 9.9% of all strikeouts in MLB. They also accounted for 5.5% of the innings pitched.

In 2019, the top 8 strikeout pitchers (Gerrit Cole, Justin Verlander, Shane Bieber, Jacob deGrom, Trevor Bauer,, Stephen Strasburg, Lance Lynn, Max Scherzer) accounted for 5.0% of all strikeouts and for 4.0% of all innings pitched. (I have to say, I have no recollection of ever hearing of Shane Bieber before today.)

That's twice as large a share of Ks for the 1946 group (compared to the 2019 group), and a 40% greater share of IP.

And, the top 4 strikeout pitchers in 1946 . combined, struck out 6.5 per 9 (against an MLB average of 3.9 per 9), while the top 8 in 2019 struck out 8 11.3 per nine (compared to an MLB average of 8.9. per 9).

I have no particular conclusions...
3:43 PM Mar 15th
 
StatsGuru
Is it time to lower the mound again? Lowering the mound in 1969 seemed to keep strikeouts in check for 20 to 30 years.
7:16 AM Mar 15th
 
bjames
Trying again to explain. . A LEAGUE may (or may not) be a sufficient context for a PLAYER, but a LEAGUE is not a sufficient context for a TEAM, just as a TEAM is not a sufficient context for a PLAYER.


Suppose that we evaluated Hal Newhouser, 1946, by comparing him only to his teammates in 1946. Would that create a sufficient context for evaluation?

Obviously it would not. The team consists of only five or six key pitchers. Very often they are not a representative sample of the actual context.

Comparing the Tigers, 1946, to the American League in 1946. . .the American League consists of only 8 teams. They may or may not be a representative sample of the true context, but certainly we cannot ASSUME that they are. A league is just not a large enough entity to be a representative context for a team.
9:08 PM Mar 14th
 
bjames
Trying to help Mons understand the problem. . .suppose that, after dividing each team by the league, you divided each PITCHER'S strikeout rate by the TEAM strikeout rate. Would that give you a better view of the pitcher's ability?

Obviously it would not, for a reason I hope you can see; if not, I'm wasting my time talking to you.

But how is THAT step different than the step that YOU propose, which is to divide each team by the league?


It isn't different at all. It's exactly the same, and has exactly the same problem.
8:45 PM Mar 14th
 
bjames
Would it not work to just divide the team strikeout rate by the league rate? Or maybe use a 3-year rate for the league, so you don't have such a problem with rapidly changing rates?



As was clearly explained in the article, this causes more problems than it solves.
8:37 PM Mar 14th
 
W.T.Mons10
Fred Hutchinson not only had an excellent strikeout to walk ratio as a pitcher, but also as a batter: 66 walks and only 30 strikeouts.
Would it not work to just divide the team strikeout rate by the league rate? Or maybe use a 3-year rate for the league, so you don't have such a problem with rapidly changing rates?

8:15 PM Mar 14th
 
 
©2020 Be Jolly, Inc. All Rights Reserved.|Web site design and development by Americaneagle.com|Terms & Conditions|Privacy Policy