September 20, 2007
Lies, Damned Lies
The Best Player in Baseball, Part One
Who is the best player in baseball right now? You can make credible arguments for two players-Albert Pujols and Alex Rodriguez. Who then will win the MVP awards in their respective leagues? A-Rod will probably win the American League's, but Pujols is unlikely to reciprocate in the NL. More probably, it will be someone like Prince Fielder, who is certainly very good, and who might have had the best season in his league, but is certainly not the best player in baseball.
The point is that these are somewhat different questions. Justin Morneau won the MVP award last year; nobody thought that he was the best player in baseball. Ned Garver had the highest WARP in baseball in 1950, Norm Cash did in 1961, and it was Wilbur Wood in 1971. These were good players who had great seasons, not the best players in baseball. Sample sizes in baseball take a fairly long time to converge, and so what we really need is some sort of weighted multi-year average.
To address that very point, I devised a weighting scheme that evaluates a player's WARP3 score over a rolling, six-year period. The weights are assigned as follows:
Year N-3 7% Year N-2 13% Year N-1 22% Year N 31% Year N+1 18% Year N+2 9%
I am not going to take a large amount of space to try and explain these weights, because to a significant extent they are arbitrary; I went with numbers that looked and felt right. The basic premises are that we ought not to place too much emphasis on any one season, because we are looking more for a sustained peak. In addition, we place somewhat more emphasis on the backward-looking years than the forward-looking years. This is based on the idea that a player must be "dethroned" from his position as the reigning best player in baseball (BPIB). At the same time, we don't throw out the future seasons entirely, because they provide some important confirmatory evidence for whether a player's numbers were legitimate, or he was just a one-hit wonder. I'm calling the resulting total SP-WARP, for Sustained Peak Wins Above Replacement.
There are a couple cases where individual seasons are ignored. Years in which a player spent all or most of his season in the military, or was stuck in the Negro Leagues, are not figured into the calculation (however, a player is not eligible for BPIB status in a season that is disqualified for one of these reasons). In addition, for seasons that precede a player's major league debut or proceed his retirement, we assign him credit for half of his weighted average SP-WARP from the seasons in which he was eligible. Certainly, if a player was too old or too young to be playing in the major leagues, this is something of an indictment of his capacity to be the best player in baseball in a previous or subsequent season, but we want to provide some balance for players like Sandy Koufax, who might have retired while still at the peak of their abilities, or those like Albert Pujols who came right out of the gate at full blast.
These rankings are not perfect. What we'd really want to know is the true talent level of every player in baseball at every point in history. This is not easy to do, because baseball career paths are like snowflakes: everyone's development pattern is unique. While I am not inclined to believe that Ned Garver was truly the best player in baseball in 1950, you can make a fairly strong argument that Dwight Gooden was in 1985. Acknowledging the imprecision in the process, I am calling the BPIB race a tie for any players that have SP-WARP scores within five percent of that of the leader. However, it's also important to note that if I were doing this list subjectively, you would wind up with somewhat different results.
The flip side of these rankings erring toward the conservative side is that they tend to be fairly stable. There are 63 players who can make a claim to being the BPIB at some point in history, however for only 28 of these players was this claim undisputed. We start the wayback machine in 1871. You can follow along with the discussion with this handy field guide, which tracks the progress of the BPIB and his four closest competitors in spectacular, retina-destroying color.
1871-1875 Ross Barnes 1876 Ross Barnes or Deacon White
Ross Barnes was arguably the single most dominant player in Major League history. There are eleven seasons in which the best player in baseball had a SP-WARP at least 25 percent higher than that of his next closest competitor. Ross Barnes is the owner of four of those seasons, all of which rank in the top six overall:
Year #1 Player SP-WARP #2 Player SP-WARP Edge 1871 Ross Barnes 10.96 Levi Meyerle 8.08 36% 1913 Walter Johnson 15.39 Eddie Collins 11.48 34% 1914 Walter Johnson 15.21 Eddie Collins 11.42 33% 1872 Ross Barnes 11.09 Davy Force 8.41 32% 1873 Ross Barnes 11.25 George Wright 8.56 31% 1874 Ross Barnes 10.83 Lip Pike 8.53 28% 1892 Amos Rusie 12.23 Kid Nichols 9.58 28% 1992 Barry Bonds 12.90 Roger Clemens 10.09 28% 1907 Honus Wagner 12.62 C. Mathewson 9.99 26% 1981 Mike Schmidt 11.60 Bobby Grich 9.22 26% 1893 Amos Rusie 14.04 Cy Young 11.22 25%
Barnes' reputation has been besmirched to some degree because once he fell from his peak, he fell incredibly quickly, and this somewhat coincided with the rapidly improving quality of competition in professional baseball. Still, Barnes hit .429 in the National League's inaugural season in 1876; his next closest competitor hit .366. His batting line was .429/.462/.560 in a league with norms of .265/.277/.321. He scored 126 runs in 69 games; George Wright finished second in that category with 72. This was nothing new for Barnes, who had been just as dominant in the National Association.
In 1877, Barnes fell ill. The fair-foul rule was also changed that year-now a ball had to stay fair past first or third base to count as a hit, whereas previously all it had to do was land in fair territory; it didn't matter if it then went to the backstop. It was unclear which of these two factors was the more important in Barnes' demise, but either way, he was never really the same.
I think Ross Barnes should be in the Hall of Fame. The way that I look at this is as follows. If, as is highly likely, any number of hitters were taking advantage of the fair-foul bunt, then Barnes' hitting prowess can't be reduced to some loophole in the rules; the rising tide was lifting all boats. Instead, Barnes' decline from 1877 onward probably had more to do with his illness. If, on the other hand, Barnes was uniquely able to take advantage of the fair-foul bunt such that baseball specifically had to change its rules to reel him in, doesn't that almost definitionally qualify him for enshrinement? "Ross, you're too good. We're changing the rule. Some day, sixty years from now, somebody is going to set up a baseball museum in Cooperstown, New York, and this is the sort of thing that will be mentioned on your plaque." How many players in sports history have precipitated a major rule change all by themselves? The closest analogy I can think of is the NCAA banning the slam dunk in preparation for Lew Alcindor's career at UCLA, although that story is probably apocryphal, as is likely the case with Barnes.
To get him in the Hall of Fame, you would have to forgive Barnes the fact that he didn't really have a career after the age of 26. But if a player can be 25 percent better than anyone else in a recognized major league for a sustained period of a half-dozen seasons, and that player isn't in the Hall of Fame, then what the hell is the Hall of Fame for?
1877-1878 Deacon White 1879-1880 Paul Hines 1881 Paul Hines, Cap Anson, or Fred Dunlap
Apart from my advocacy for Ross Barnes, I have no particular passion for 19th century baseball, but White and Hines also probably ought to be in the Hall. Each of them have plenty of black ink on their career records, and unlike Barnes they coupled that with long careers; Hines was one of the better players in baseball by age 18, while White played regularly into his forties.
1882 Fred Dunlap, Jim Whitney, or George Gore 1883 Jim Whitney, Dan Brouthers, or Charley Radbourn 1884 Charley Radbourn 1885-1886 Dan Brouthers or Roger Connor 1887 John Clarkson 1888 John Clarkson or Roger Connor 1889 John Clarkson 1890 John Clarkson, Roger Connor, or Silver King
Now we get into the era of the 600-inning pitcher. Jim Whitney's name is perhaps the least familiar among this group, but the thing is that he was a fantastic two-way player who was penciled in at outfield or first base on those rare days that he wasn't pitching. His lifetime OPS+ was 112, and in 1882 he hit .323/.382/.510 over 251 at-bats. He was almost literally a one-man team.
1891 Roger Connor or Amos Rusie 1892-1894 Amos Rusie
There are a lot of parallels between Amos Rusie's career and Ross Barnes', and it's curious that Rusie is in the Hall of Fame while Barnes is not.
1895 Cy Young 1896 Cy Young or Hughie Jennings 1897 Hughie Jennings 1898 Hughie Jennings, Cy Young, Kid Nichols, or Ed Delahanty 1899 Cy Young or Ed Delahanty 1900-1902 Cy Young, Honus Wagner, or Nap Lajoie
Cy Young is one of relatively few great players whose careers straddle the boundary between the NA/NL era and the birth of "modern" baseball in 1901. In 1900, along with Honus Wagner and Nap Lajoie, he was one of the three best players in baseball, and the defection of Young and Lajoie to the American League was critical for lending credibility to the junior circuit. We are mostly ignoring Hughie Jennings, who did not have a long peak, but for a period of three or four years, relative to the competition he combined a Derek Jeter-type bat with an Ozzie Smith-type glove.
1903-1904 Honus Wagner or Nap Lajoie 1905-1908 Honus Wagner 1909 Honus Wagner or Christy Mathewson
Wagner's year-to-year batting numbers are a study in consistency, but through the noise he had a somewhat late peak; his best WARP3 score was 15.4 in 1908, at which time he was 34. The trick was that Wagner was able to maintain his numbers while the rest of the league entered the Deadball Era; the National League went from hitting .267 in 1901 to .239 in 1908.
1910 Ed Walsh 1911 Ed Walsh, Walter Johnson, or Ty Cobb 1912-1915 Walter Johnson 1916-1917 Walter Johnson or Pete Alexander 1918 Walter Johnson
…and it shouldn't be any surprise that once Wagner's skills started to atrophy just a little bit, the emerging stars of the Deadball Era were pitchers-Ed Walsh, Pete Alexander, and especially Walter Johnson. I don't know whether it is the result of some mathematical property intrinsic to WARP, but this generally seems to be the case: in offense-dominant eras, the best player in baseball is a hitter, while in defense-dominant eras, the best player in baseball is a pitcher.
One of the major surprises of this study is how rarely Ty Cobb's name appears at the top of the heap; Roger Maynard would not be pleased. Cobb never actually finished the season with the highest SP-WARP total in baseball, the best he can manage is an honorary three-way tie with Johnson and Walsh in 1911. It isn't that WARP doesn't like Cobb; his lifetime WARP3 of 194.3 is the ninth-highest of all time. But he had a little bit of a bifurcated peak because he missed significant playing time in 1913 and 1914, which was right at his age-26/27 seasons. In 1914, Cobb got hit by a Dutch Leonard pitch; he stayed in the game and responded by spiking Leonard when the pitcher tried to field a bunt in a later at-bat; only later was it discovered that Cobb had suffered a broken rib in the process (Leonard would get the last laugh, implicating Cobb in a game-fixing scandal that would force his temporary retirement in 1927). Later that same season, Cobb missed the better part of two months after breaking his thumb in a fight with a butcher's clerk. Basically, Cobb missed a fair amount of playing time throughout his career for reasons that boiled down to his being a jackass-injuries from fights, injuries from rough play, suspensions, contract holdouts, and finally his coerced (and temporary) retirement in 1927 from a gambling scandal. That's really what prevented him from appearing more frequently on this list. Somewhat like Terrell Owens today, you knew that when things were going well with Ty Cobb was right when they were about to turn ugly.
1919-1924 Babe Ruth 1925 Babe Ruth or Rogers Hornsby 1926-1930 Babe Ruth 1931 Babe Ruth or Lefty Grove 1932 Babe Ruth, Lou Gehrig, Mel Ott, or Joe Cronin
There were actually two dominant players during Ruth's era: the Babe and Rogers Hornsby. The gap between Hornsby and the third-best player in the league was generally larger than that between Ruth and Hornsby. Hornsby briefly pulled into a virtual tie with Ruth in 1925, when the Babe played in just 98 contests because of illness.
1933 Lou Gehrig, Mel Ott, Carl Hubbell, Arky Vaughan, or Jimmie Foxx 1934-1936 Arky Vaughan 1937 Arky Vaughan or Mel Ott 1938 Arky Vaughan 1939 Arky Vaughan or Joe DiMaggio
Lou Gehrig is generally thought of as Babe Ruth's contemporary, but he was eight years younger and briefly took on the role of the BPIB in 1932 and 1933, not so much because he was getting better but because Ruth's career was beginning to draw to its close. But the best player of the late 1930s was someone who never finished higher than thirrd in the MVP balloting and never won a World Series. That player was Arky Vaughan, who in 1935 became the only shortstop besides Honus Wagner to lead his league in OPS in the modern era; the .491 OBP he posted that season is also the best ever for a shortstop. Vaughan had a relatively brief peak, but for five or six seasons he was the best player in baseball, and that wasn't owing to a lack of competition-Gehrig, Mel Ott, and Charlie Gehringer were all consistent 12-WARP players during this period.
1940 Joe DiMaggio or Bob Feller 1941 Bob Feller or Ted Williams 1942 Ted Williams 1943 Lou Boudreau, Charlie Keller, or Luke Appling 1944 Stan Musial 1945 Hal Newhouser 1946-1947 Ted Williams 1948 Ted Williams or Stan Musial
The early and mid 1940s were one of the richer eras for top-line talent in baseball history: Joe DiMaggio, Ted Williams, and Bob Feller were all right at their peaks. Unfortunately, right in the middle of all that, there was a war going on, and by 1943 all three of these superstars had enlisted in the armed forces; Feller was the first major leaguer to volunteer, immediately after the attack on Pearl Harbor. This denied the league of what might have been some of the most competitive pennant races in baseball history-Williams' Red Sox, DiMaggio's Yankees, and Hank Greenberg's Tigers were all potentially dominant teams, while Feller's Indians and the St. Louis Browns would have been plucky foes. (I've chronicled the 1944 American League at great length in It Ain't Over.)
With that said, Newhouser's BPIB crown in 1945 should not be looked at as a fluke. His SP-WARP score in that season was 11.75, which is nearly identical to the seasonal average highest SP-WARP score of 11.86. Keeping in mind that WARP is adjusted for league quality, so Newhouser's SP-WARP actually rose a bit in 1946 and 1947-no surprise, since he led his league in ERA in 1946 and had a better strikeout rate than Feller.
1949 Stan Musial 1950-1951 Stan Musial or Jackie Robinson 1952 Stan Musial, Jackie Robinson, or Robin Roberts 1953 Stan Musial or Robin Roberts
Stan Musial owes his superiority over this period to his ability to stay healthy. Between 1946 and 1956, Musial played in an average of 153.5 games per season, while Ted Williams is only at 131.8, even after excluding the two seasons where he spent part of his time in Korea. Williams was perhaps five percent better on a rate basis, but the Cardinals were getting 15 percent more Musial than the Red Sox were getting Williams.
Instead, Musial's primary competition for the title was Jackie Robinson, who remains absurdly underrated as a baseball player, and Robin Roberts, who was not only very good, but was being used radically differently than any other pitcher of his era. Between 1950 and 1956, Roberts averaged 319 innings per season. This was not generally an era of starters with 300 IP, something which had a revival in the mid-1960s when offense went into the tank, making it easier for pitchers to get through lineups while keeping their pitch counts down. Between 1946 and 1961, there were a total of 13 seasons in which a pitcher threw at least 300 innings; Roberts was responsible for seven of those, while every other pitcher in baseball combined for six over a period that spanned 16 seasons.
Tomorrow, tune in for Part Two, where Nate concludes his roundup of the best players in baseball.