Bert, Again (and Again, and Again…)

I didn’t really have high hopes this year for being able to write, “our long national nightmare is over … Bert Blyleven is in the Hall of Fame.” I had certainly hoped to be able to do so, as like many others I feel that Bert’s election is overdue.

Not for the first time, the election spurred some discussion over at Batter’s Box, where I hang out frequently. Like everywhere else, Bert is a controversial subject over there, with his supporters and detractors. One thing that caught my eye in particular was the assertion by Mike D of Batter’s Box that if Bert were elected,

He’d be in about the 30th percentile of Hall starting pitchers, and lower than that in the Hall generally.

There are only 49 starting pitchers in the Hall of Fame. This ignores the relievers (Dennis Eckersley, Rollie Fingers, Hoyt Wilhelm, Bruce Sutter), and the Negro Leaguers (for whom comparable data is lacking) and 19th-century pitchers (who were playing a different game). It also excludes Babe Ruth, George Sisler and Satchel Paige who were primarily inducted as non-pitchers (or as a Negro League player in Satch’s case), but includes Cy Young as a guy who pitched a significant amount of time in the 20th century.

Hall of Fame starting pitchers obviously better than Bert: 18.

Pete Alexander
Mordecai Brown
Steve Carlton (a stretch, yes)
Bob Feller
Whitey Ford
Bob Gibson
Lefty Grove
Carl Hubbell
Walter Johnson
Juan Marichal
Christy Mathewson
Jim Palmer
Gaylord Perry (stretch some more)
Eddie Plank
Tom Seaver
Warren Spahn
Ed Walsh
Cy Young

Guys who are hard to compare to Bert but are probably better: 7.

Dizzy Dean
Lefty Gomez
Addie Joss
Sandy Koufax
Hal Newhouser
Dazzy Vance
Rube Waddell

Hall of Fame starting pitchers obviously worse than Bert: 7.

Chief Bender
Jack Chesbro
Burleigh Grimes
Jesse Haines
Catfish Hunter (somebody has to say it)
Rube Marquard
Herb Pennock

That leaves a body of 17 pitchers who might be better than Bert, and might be worse. Where you rank Bert in relation to these players really determines whether you see Bert Blyleven as a Hall of Famer or not. If you’d take these 18 guys (17 plus Bert) and rank Bert in the bottom five, you’ll think he’s out, since he would clearly be among the guys who were not good choices. If you think he’s in the top five, he’s definitely in, unless you are really arguing for an absurdly small Hall. Top 10, you probably think he’s in too. But these are the guys who determine where Bert Blyleven ranks for you—whether in the middle of the Hall of Fame (if you put him first), or somewhere near the bottom (if you put him last).

But if you’re being consistent in any way, it is really hard to rank Bert near the top of that group, and say he should be out of the Hall of Fame. To argue this is to say that half the pitchers in the Hall of Fame shouldn’t be there, and while a person might want a small Hall of Fame, to do so is to flout standards that have been clearly established for the standard required of a Hall of Fame pitcher.

Let me say right now before we start that clearly, just because something is obvious to me does not make it obvious to everyone, let alone true. I talked to Mike Green from Batter’s Box about this, a guy who’s spent as much top-quality brainpower as anyone I know on the question of what makes a Hall of Fame pitcher (see his Hall Watch series for more). He felt strongly that Blyleven was clearly better than guys like Bunning and Faber, among several others, and wasn’t at all happy with me listing guys like Carlton, Perry and Walsh as clearly better. Mike is probably right, truth be told, because the one thing I didn’t want to do in this analysis is to take advantage of the cutting edge of our knowledge and understanding about pitchers. It’s pretty clear, for example, that a lot of what we once thought was Ed Walsh’s pitching is in fact the defense behind him.

But I don’t want to throw those kinds of numbers out here because I am trying to reconstruct the rudiments of both a case for Bert Blyleven for the Hall of Fame and a case against Bert Blyleven for the Hall of Fame. I want the case to be easily and simply understood by someone like an average Hall of Fame voter or average reader of this site, who understands baseball statistics but can get lost very easily (as I can myself) in the more intricate reasoning and calculations of cutting-edge analysis. So while I might “know” that Ed Walsh isn’t in fact “clearly” better than Bert Blyleven, his numbers seem clearly better, and that’s going to be good enough for me to put him in that category.

I know that there are four 300-game winners on our list, and that in the minds of some voters those guys are “clearly” established as better than Bert Blyleven. But every reader of this site knows and understands that 13 wins, or even 37 wins (the difference between Blyleven and Ryan or Sutton), does not “clearly” establish any such thing, so I think that we will allow that argument to slide gracelessly past.

The 17 pitchers are :

A Hardball Times Update
Goodbye for now.
Player               Played   Inducted  W   L    WL%  ERA    G    IP     H   HR  BB   SO  ERA+
Jim Bunning*        1955-1971   1996   224 184  .549  3.27  591  3760.3 1366 372 1000 2855  114
Stan Coveleski*     1912-1928   1969   215 142  .602  2.89  450  3082.0  990  66  802  981  127
Don Drysdale*       1956-1969   1984   209 166  .557  2.95  518  3432.0 1124 280  855 2486  121
Red Faber*          1914-1933   1964   254 213  .544  3.15  669  4086.7 1430 111 1213 1471  119
Waite Hoyt*         1918-1938   1969   237 182  .566  3.59  674  3762.3 1500 154 1003 1206  111
Fergie Jenkins*     1965-1983   1991   284 226  .557  3.34  664  4500.7 1669 484  997 3192  115
Bob Lemon*          1946-1958   1976   207 128  .618  3.23  460  2850.0 1024 181 1251 1277  119
Ted Lyons*          1923-1946   1955   260 230  .531  3.67  594  4161.0 1696 223 1121 1073  118
Joe McGinnity*      1899-1908   1946   246 142  .634  2.66  465  3441.3 1016  52  812 1068  121
Phil Niekro*        1964-1987   1997   318 274  .537  3.35  864  5404.3 2012 482 1809 3342  115
Eppa Rixey*         1912-1933   1963   266 251  .515  3.15  692  4494.7 1572  92 1082 1350  115
Robin Roberts*      1948-1966   1976   286 245  .539  3.41  676  4688.7 1774 505  902 2357  113
Red Ruffing*        1924-1947   1967   273 225  .548  3.80  624  4344.0 1833 254 1541 1987  109
Nolan Ryan*         1966-1993   1999   324 292  .526  3.19  807  5386.0 1911 321 2795 5714  112
Don Sutton*         1966-1988   1998   324 256  .559  3.26  774  5282.3 1914 472 1343 3574  108
Vic Willis*         1898-1910   1995   249 205  .548  2.63  513  3996.0 1167  66 1212 1651  118
Early Wynn*         1939-1963   1972   300 244  .551  3.54  691  4564.0 1796 338 1775 2334  106
 
Bert Blyleven       1970-1992          287 250  .534  3.31  692  4970.0 1830 430 1322 3701  118

Bert certainly looks comfortable in this company, but in the end you can’t make a sensible comparison with all this data. What I decided to do was to compare the 17 pitchers on a consistent set of eleven data that is relevant to being a good pitcher. These were wins, winning percentage, shutouts, innings pitched, ERA+, Earned Runs Saved Above Average, Strikeouts per 9 innings, Indicated ERA, Walks per 9 innings, Strikeout-to-Walk Ratio, and Hits per 9 innings, which was somehow deleted from the data table but left in the rankings.

Player            W   WL%  ERA SHO  IP    H    ERA+ ERSAA K/9 IERA W/9 K/B Fibo 
Jim Bunning*     224 .549 3.27  40 3760.3 3433  114 191   6.8 2.63 2.4 2.9  163 
Stan Coveleski*  215 .602 2.89  38 3082   3055  127 267   2.8 0.56 2.3 1.2  202 
Don Drysdale*    209 .557 2.95  49 3432   3084  121 236   6.5 2.03 2.2 2.9  159 
Red Faber*       254 .544 3.15  29 4086.7 4106  119 271   3.2 0.81 2.7 1.2  179 
Waite Hoyt*      237 .566 3.59  26 3762.3 4037  111 165   2.8 1.09 2.4 1.2  189 
Fergie Jenkins*  284 .557 3.34  49 4500.7 4142  115 250   6.3 2.38 2.0 3.2  216 
Bob Lemon*       207 .618 3.23  31 2850   2559  119 194   4.0 2.78 4.0 1.0  207 
Ted Lyons*       260 .531 3.67  27 4161   4489  118 305   2.3 1.44 2.4 1.0  168 
Joe McGinnity*   246 .634 2.66  32 3441.3 3276  121 213   2.7 0.36 2.1 1.3  260 
Phil Niekro*     318 .537 3.35  45 5404.3 5044  115 301   5.5 2.99 3.0 1.8  215 
Eppa Rixey*      266 .515 3.15  37 4494.7 4633  115 235   2.7 0.49 2.2 1.2  152 
Robin Roberts*   286 .539 3.41  45 4688.7 4582  113 230   4.5 2.07 1.7 2.6  195 
Red Ruffing*     273 .548 3.80  45 4344   4284  109 165   4.1 2.07 3.2 1.3  198 
Nolan Ryan*      324 .526 3.19  61 5386   3923  112 229   9.5 3.09 4.7 2.0  202 
Don Sutton*      324 .559 3.26  58 5282.3 4692  108 153   6.0 2.27 2.3 2.7  249 
Vic Willis*      249 .548 2.63  50 3996   3621  118 210   3.7 0.50 2.7 1.4  180 
Early Wynn*      300 .551 3.54  49 4564   4291  106 107   4.6 2.88 3.5 1.3  221 

Bert Blyleven    287 .534 3.31  60 4970   4632  118 329   6.7 2.30 2.4 2.8  190 

Average of 17    263 .558 3.24  41 4190   3956  115 219   4.6 1.88 2.7 1.8 

Some of these numbers (ERA+ and ERSAA) were adjusted to era and park, but most were not. Indicated ERA, for those who are not familiar, is a stat invented by Bill James that is described in the 1988 Baseball Abstract. It is calculated by walks * home runs allowed * 100, divided by the square of innings pitched. I used IERA to stand in for DIPS or some similar measure, primarily because it doesn’t use strikeouts and therefore wouldn’t favor Bert over earlier pitchers who pitched in low-strikeout conditions.

Incidentally, the “Fibo” line in the data table is Fibonacci Win Points, a system developed by Bill James in his book Whatever Happened To The Hall of Fame? that sums up a pitcher’s won-lost record in a single number. Essentially it’s half a pitcher’s wins, plus his games over .500. A pitcher with a .618 winning percentage (as HOFers frequently do) will have Fibonacci Win Points equal to his wins. So Bob Lemon, for example, has 207 wins and 207 Fibonacci Win Points. I didn’t use the Fibonacci Win Points in the ranking, though.

The eleven data points I did use, I felt, were quite well balanced. Two concern a pitcher’s won-lost record (wins and win percentage); two concern his runs allowed (ERA+ and earned runs saved); one concerns straight dominance employing factors from all over the map (shutouts); two concern his control (walks, plus half of Indicated ERA and half of K/W ratio); one and a half concern his strikeout ability (strikeouts plus half of strikeout-to-walk ratio); one concerns directly his ability to prevent hits (hits allowed) and a half concerns his ability to keep the ball in the park (home runs allowed, plus half of Indicated ERA). Then there’s innings pitched, which are an important factor in themselves. Four are count stats (wins, shutouts, innings, earned runs saved) and seven are rate stats (win percentage, ERA+, strikeouts, Indicated ERA, walks, strikeout-to-walk ratio, and hits allowed), so there is balance there too. It’s a good list to measure a pitcher’s quality.

Having these eleven different data points for each pitcher, I used the most simple ranking method I could think of: I just ranked each of the 18 pitchers for each stat and assigned him points equal to his rankings. I then added up each pitcher’s points; lowest score wins.

Player            W W% SO IP E+ ER  K IE BB KB  H Total Rank  
Jim Bunning*     15  9 11 14 12 14  2 14  8  3  6  108   10  
Stan Coveleski*  16  3 12 17  1  5 15  4  7 14 14  108   10  
Don Drysdale*    17  6  5 16  2  7  4  8  5  2  4   76    3  
Red Faber*       11 12 16 11  4  4 13  5 12 15 15  118   13  
Waite Hoyt*      14  4 18 13 15 15 14  6 10 16 17  142   18  
Fergie Jenkins*   7  6  5  7  9  6  5 13  2  1  7   68    1  
Bob Lemon*       18  2 15 18  4 13 11 15 17 17  3  133   16  
Ted Lyons*       10 16 17 10  6  2 18  7 11 18 18  133   16  
Joe McGinnity*    1  3  1 14 15  2 11 16  1  3 10   77    5  
Phil Niekro*      3 14  8  1  9  3  7 17 14  8  9   93    8  
Eppa Rixey*       9 18 13  8  9  8 17  2  4 13 16  117   12  
Robin Roberts*    6 13  8  5 13  9  9  9  1  6 12   91    7  
Red Ruffing*      8 10  8  9 16 16 10 10 15 12 13  127   15  
Nolan Ryan*       1 17  1  2 14 10  1 18 18  7  1   90    6  
Don Sutton*       1  5  3  3 17 17  6 11  6  5  2   76    3  
Vic Willis*      12 10  4 12  6 12 12  3 13  9  5   98    9  
Early Wynn*       4  8  5  6 18 18  8 16 16 11 10  120   14  
Bert Blyleven     5 15  2  4  6  1  3 12  9  4  8   69    2 

This method finds that Fergie Jenkins is the best of the 18 pitchers in the list, with Blyleven a very, very close second. Don Drysdale, despite his short career, is tied for third with Don Sutton. The system thinks Waite Hoyt, Bob Lemon, Ted Lyons and Red Ruffing are the weakest of the 18, and I would agree 100% with that sentiment, while recognizing that Lemon only ranks so low because of his short career. I thought Bert would come in about the middle of this group, and the high ranking surprised me.

Is Blyleven the second-best pitcher on this list? I tried to be both objective and all-encompassing, and just measure almost everything relevant while keeping the method balanced and simple, and I came to that conclusion. Bert might easily be considered the best; he beats Fergie Jenkins in seven of the eleven categories but big losses in winning percentage and walks put him second.

Having crunched the numbers every way I could think of, I can only conclude that Bert Blyleven is one of the two or three best pitchers on that list of eighteen mid-level Hall of Famers. He belongs.

References & Resources
Thanks as always to Baseball-Reference.com and to the Sabermetric Encyclopedia for making this kind of research a breeze.


Comments are closed.