What Do We Learn From Doing This?
The most obvious things we are going to learn, of course, are the things we already know but can now measure.
1) There is a strong bias in the voting toward players who play on winning teams.
Really; I’m not kidding. Treating "team wins" as part of the individual’s stat line, the won-lost record of Team Wins in competition with other stats is .692 (525 wins, 233.5 losses).
This bias has lessened over time. In the first half of the MVP era, the winning percentage of team wins in this format was .749; in the second half it has been .627. In the first 20 years of the study (1931-1950) the winning percentage of team wins in this format was .845 (160.5 – 29.5).
There have been 59 MVP contests which were won by a player on a team with more wins in preference to a player who was as good or better, but played for a team that was not as good. There have been 28 MVP awards won by a player who played on a worse team and was also a weaker player (or no better player).
In 1934 Lou Gehrig hit .363 with 49 homers, 165 RBI, while Mickey Cochrane hit .320 with 2 homers, 76 RBI. Cochrane was a fine player, but we credit him with 23 Win Shares in 1934, as opposed to 41 for Gehrig—but Cochrane’s Tigers won the pennant, with Cochrane as the player/manager, so Cochrane was voted the MVP. Gehrig won the triple crown, but actually finished fifth in the MVP voting, behind three Tigers (Cochrane, Gehringer and Schoolboy Rowe) and a teammate (Lefty Grove, who was 26-5 with a 2.33 ERA.) Since Gehrig was 18 Win Shares ahead, we credit "Team Wins" with 21 "wins" for that award. Overall, the category has 525 wins, so those 21 are not really a large element in that total, although that is the worst individual case.
On the other hand, in 1937 Charlie Gehringer won the American League MVP Award although in our view Joe DiMaggio was a significantly better player (39-30) and the Yankees won the pennant by 13 games. We count that as 12 losses for the category Team Wins. In 1938 Ernie Lombardi won the National League MVP Award although Mel Ott was (we think) a much better player (36-24) and Ott’s team (the Giants) won the pennant. In 1952 Hank Sauer won the National League MVP Award although Stan Musial, even having an off season, was a much better player than Sauer (37-28) and the Cardinals’ record was eleven games better than the Cubs’ (although neither team won the pennant.) In 1958 Jackie Jensen and Ernie Banks won the MVP Awards although Mickey Mantle and Willie Mays were quite significantly better players (39-27 and 40-31) and also played for better teams. In 1959 Banks won the Award again although Hank Aaron was a better player (38-33) and the Braves’ had a better record. In 1966 Roberto Clemente won the MVP Award although Mays, again, was a significantly better player (37-29) and the Giants had a better record than the Pirates. In 1987 George Bell and Andre Dawson won the MVP Awards although Alan Trammell had a much better year than Bell (35-26) and Tim Raines a much better year than Dawson (34-20) and Trammell and Raines also played for better teams than Bell and Dawson. In 1998 Ivan Rodriguez won the American League MVP Award although Derek Jeter was much more valuable in fact (35-28) and the Yankees had a better record (although both players played for division-championship teams.) In 2006 Justin Morneau won the American League MVP Award although we believe, again, that Jeter was a significantly better player (32-26) and the Yankees’ record was a game better, although (again) both teams won their divisions.
So it’s not an absolute rule that the better team gets the award; it’s merely a strong bias in the voting. There are other biases that compete with that one, and sometimes the other bias prevails.
Of course, there remains the issue of whether this is legitimate credit to winners or illegitimate credit to the winner. The voters have a right to their opinion; by no means do I wish to dismiss their point of view; merely to understand it. But arguing the issue from our side. …our analysis is based on wins. It’s not like we ignore the team’s won-lost record. It’s the foundation of what we do. Sometimes we give what appears to be inexplicable credit to individual players based on the fact that their team won more games than they ought to have won—as in the 2008 National League, when Albert Pujols won the MVP Award and anyone would have to agree that Pujols’ numbers appear to be better than Lance Berkman’s, but our system believes that Berkman was better (36-34) essentially because Berkman’s team was more efficient at producing wins. St. Louis scored 67 more runs than Houston and also allowed 18 fewer runs—yet Houston’s record was a half-game better. We all know that that’s kind of a fluke, something that just happened, but the assumption of our analysis is "they won the games; they have to be given credit for winning the games, fluke or not." If you adjust for the efficiency of the runs, Berkman was better than Pujols, even though Pujols had what look like better numbers.
Our system gives a player full credit for his contributions to the team’s wins, and sometimes, frankly, gives him too much credit for his contributions to the team’s wins (like Berkman). But what we don’t do is give him credit for his teammates’ contributions to wins. Not as much now, but in the early years of the voting, writers would overlook vast differences in individual contributions to victory to choose a player from a championship team. In 1942 Ted Williams won the American League triple crown, hitting .356 with 36 homers, 137 RBI. Joe Gordon won a triple crown of a different kind; he led the league in strikeouts, errors and grounding into double plays—but was voted the Most Valuable Player. Don’t get me wrong; Gordon was a fine player (despite the strikeouts, errors, and double plays.) Gordon was a good player, but we score it Ted Williams 46, Joe Gordon 31—but the Yankees won the pennant, so they gave Gordon the MVP Award. To us, that seems like giving Gordon credit because he had a much superior group of teammates. The 1935 National League Award was just as bad. Arky Vaughan hit .385 with 19 homers, 99 RBI, and a fairly incredible strikeout/walk ratio of 18 to 97. Gabby Hartnett played 116 games, hit .344 with 13 homers, 91 RBI, more strikeouts than walks—but because Hartnett’s team won the pennant, Hartnett won the MVP Award. We score it as Vaughan 39, Hartnett 24—thus, we credit that as 18 "wins" for Team Wins in the MVP bias detector.
2) MVP Voters Appear to Signficantly Over-Value Playing a Key Defensive Position.
Again; MVP Voters appear to us to significantly over-value playing an "up the middle" defensive position. It could be that they’re right and we’re wrong, but in any case, they place much more value on playing a key defensive position than does my analysis.
In the history of the MVP vote, the winning percentage of "Defensive Position" in competition with other values is .700 (462.5 Wins, 198.5 Losses.) Many of the votes that contribute to this score are the ones we have already discussed—Cochrane (c) over Gehrig (1b) in 1933, Hartnett (c) over Vaughan (ss) in 1935, Lombardi (c) over Mel Ott (3b) in 1938, Gordon (2b) over Williams (lf) in 1942, Banks (ss) over Mays (cf) in 1958, Banks (ss) over Aaron (rf) in 1959, Maury Wills (ss) vs. Willie Mays (cf) in 1962, and Ivan Rodriguez (c) over Jeter (ss) in 1998. (Defensive positions are evaluated, of course, by the Defensive Spectrum, and, in this version, Catcher is considered to be the #1 position on the Defensive Spectrum, even though at other times I have argued that catching is so distinct from the other defensive positions that it’s really not a part of the defensive spectrum. Pudge is considered to be playing a "better" defensive position than Jeter in 1998 because catching is the top dog to MVP voters. The spectrum that we used to score these votes was C-SS-2B-CF-3B-RF-LF-1B-DH.)
This bias has lessened over time even more significantly than the better team bias. In the first 40 years of the Award, the winning percentage of Defensive Position in competition with other values was .754—even higher than the value given to playing on a better team (.749). Since 1971 the winning percentage of Defensive Position in MVP competitions is .627.
Other votes that have contributed significantly to the .700 winning percentage of Defensive Position are Frankie Frisch vs. Wally Berger in 1931, Marty Marion vs. Stan Musial in 1944, Joe D vs. Ted Williams in 1947, Lou Boudreau vs. Ted Williams in 1948, Jackie Robinson vs. Stan Musial in 1949, Roy Campanella vs. Stan Musial in 1951, Roy Campanella vs. Eddie Mathews in 1953, Yogi Berra vs. Mickey Mantle in 1954 and again in 1955, Roy Campanella vs. Willie Mays in 1955, Dick Groat vs. Mays or Mathews in 1960, Thurman Munson vs. George Brett in 1976, Kirk Gibson vs. Will Clark in 1988, Kevin Mitchell vs. Will Clark in 1989, and Jimmy Rollins vs. David Wright in 2007.
I will note, of course, that there are also many awards which are deservedly won by players because of their defensive contributions. The 1950 Award given to Phil Rizzuto was, in our view, well deserved. The famous 1941 vote for Joe DiMaggio over Ted Williams is really not an unreasonable vote when you look at the parks and factor in the value of DiMaggio’s defense (we score it 42-41 for Williams, but it’s not unreasonable to think DiMaggio was actually better.) In 1947 it’s pretty unreasonable to think that DiMaggio was actually better. In 1942 it’s pretty unreasonable to think that Joe Gordon was better than Williams, but again, the voters have a right to their opinions; we can’t assume that they’re wrong just because our best efforts to calculate everything reach a different conclusion.
There are always multiple biases at work; what we’re trying to do is sort them out and balance one against another to test which is stronger. Let’s focus on two contests between teammates—1944 and 1954-1955. In 1944 Stan Musial hit .347 with 51 doubles, 14 triples and 12 homers. His teammate Marty Marion hit .277 with 27 doubles, 5 triples and one homer—yet Marion won the MVP Award, probably one of the weakest players (and weakest seasons) ever to take home the trophy.
Marion was not merely a shortstop; he was a great defensive shortstop. He had the reputation as the best defensive shortstop of the 1940s, and our statistical analysis agrees that he was. However, the MVP voters appear (to us) to be placing a fantastic weight on the defense. Musial had 78 more hits and 51 more walks, had easily more than twice as many extra base hits, and drove in and scored not far from twice as many runs.
It has to do with how the human mind calculates things intuitively. Hitting isn’t everything; you also have to give credit to the fielders, which means that sometimes you have to overlook the fact that the outfielder is a better hitter. Those are true statements; no one would argue that they’re not. The thing is that once you make the decision to overlook the fact that the outfielder is a better hitter, you can overlook any amount of difference; you can overlook a 70-point difference in batting average as easily as you can overlook a 20-point difference.
To return to the point that I make several times a week, the world is billions of times more complicated than the human mind; therefore, the human mind is totally unable to deal with the complexity of the external world. We simplify the external world in an effort to make sense of it. The value of science, the value of research, is that it gives us a better system by which to simplify the world. Research tells us how much offense can reasonably be ignored to give credit to defense. But in 1944 they had no good way to say how many runs Musial had created, or how many Marion had created. They had no framework with which to estimate how many runs a good shortstop was worth. Those voters were every bit as smart as we were, and they knew vastly more than we know about the players that they were watching in 1944—but their system of simplifying the world allowed them to throw out an outfielder’s batting stats, no matter how many games the hitter might have won for his team with his bat.
In 1954 we credit Yogi Berra with 34 Win Shares, Mickey Mantle with 36, yet Yogi won the Award. That’s fair enough; Berra had a tremendous season, hitting .307 with 22 homers, 125 RBI. We calculate that Mickey was a little bit better, but only a little bit, and. . .we could be wrong.
But the next year, Mickey was up to 41 Win Shares, Yogi went down to 24—and Yogi won the vote again. The intuitive calculation—the calculation that you would make with arguments but no numbers—was the same in 1955 as it had been in 1954. Yogi and Mickey were still teammates, Yogi was still the catcher, Mickey still hit more homers and scored more runs, but Yogi still drove in more runs. What was different was the numbers. Yogi’s advantages had grown smaller; Mickey’s had grown larger. The intuitive calculation did not adjust for this.
The 1947 MVP race is like that; the intuitive calculations of 1941, which had reasonably concluded that DiMaggio was better than Williams, were carried forward to 1947, even though it was no longer really reasonable to argue that DiMaggio was still better than Williams. In 1941 Williams created 24 more runs than DiMaggio, which you can reasonably trade off against park effects and defense. In 1947 Williams created 52 more runs than DiMaggio. That’s different—but it’s only different if you can calculate the difference; otherwise it’s the same.
There have been 49 MVP Awards that were won by a player playing a more difficult defensive position in preference to an equal or better player (almost always better) playing a less critical defensive position. On the other hand, there have been 22 MVP Awards that went the other way. In 1941 Dolph Camilli and Pete Reiser were teammates on the Dodgers, and the Dodgers won the National League pennant. Reiser—basically a rookie—won the batting title at .343, and also led the National League in doubles (39), triples (17), and runs scored (117). Dolph Camilli, a veteran, led the National League in home runs and RBI. Reiser played center field, Camilli first base, and because of this we see Reiser as being the more valuable player (34-29), and as the most valuable player in the league—but Camilli, playing the less demanding defensive position for the same team, won the Award.
When I started my writing career there were a lot of old Brooklyn Dodger fans who remembered Pete Reiser’s 1941 season, and would write glowing tributes to Reiser’s memory and to the career that he lost to injuries. Anything you wrote about Reiser was radioactive—much like writing about Dick Allen today, or Tony Oliva. There are some players who have that effect on people. I reflect with much sadness that those guys are almost all dead now, I suppose; there aren’t many people left to remember Reiser. Anyway, that’s one award that went against the grain, against the player playing the "up the middle" defensive position, even though Reiser probably should have won the award.
Another one was the National League in 1945; Phil Cavaretta (1b) won the award in preference to Stan Hack, who was the third baseman on the same team, and who was (in our view) the "true" most valuable player. In the American League in 1958, 1960, 1961 and 1964 the awards went to Jackie Jensen, a right fielder, Roger Maris, a right fielder, and Brooks Robinson, a third baseman, even though a center fielder, Mickey Mantle, was probably the true most valuable player in the league in all four of those seasons (as well as 1954, 1955, 1959 and the three years when he did win the award.) In 1967 the National League award went to a first baseman (Orlando Cepeda) even though a third baseman (Ron Santo) was probably more valuable. In 1973 the National League award went to a third baseman (Pete Rose) even though the second baseman on the same team (Joe Morgan) was probably more valuable. In 1974 the National League award went to a first baseman (Steve Garvey) even though a third baseman (Mike Schmidt) was far more valuable. In 1979 the American League award went to a Designated Hitter/Left Fielder (Don Baylor) even though a center fielder (Fred Lynn) was more valuable. In 1996 the American League MVP Award went to a right fielder (Juan Gonzalez) even though a shortstop (Alex Rodriguez) was far more valuable by our calculations (34-21). In 2006 the American League award went to a first baseman (Morneau) even though a shortstop was more valuable (Jeter). In 2008 the Award went to a second baseman (Dustin Pedroia) even though a catcher (Joe Mauer) had more Win Shares. The player playing the more demanding defensive position doesn’t always win, even when he should win. But the advantage for the player playing the key defensive position over an equally valuable player playing a lesser position is 70-30.
3) MVP Voters Dramatically Under-Value Hitting in General. Yes, there are some kinds of hitters the voters like and some kinds they don’t like as much, but MVP voting, historically, has under-valued hitting by a huge, huge margin—much greater than the margin by which they have over-valued defensive position or over-valued playing for a good team. In essence, both of those biases have come at the expense of hitters, so that the extent to which hitting is under-valued in MVP discussions is the sum extent to which defense, playing for a good team and pitching are over-valued.
Remember, we ignore any MVP contest which is won by the "right" man. But in MVP contests between two hitters which are won by the wrong man, the winning percentage of the lower OPS is .792. The winning percentage of OPS, in an MVP contest, is .208.
There have been 58 MVP contests in history that were won by one player when there was another player who was more valuable in fact and also had a higher OPS. You almost can’t start naming them, because it is so near to being all of them. All of these awards that we have discussed here—Cochrane over Gehrig, Hartnett over Vaughan, Lombardi over Ott, Camilli over Reiser, Gordon over Williams, Marion over Musial, DiMaggio over Williams, Boudreau over Ted Williams, Jackie Robinson over Musial, Campanella over Musial, Hank Sauer over Ted Williams, Yogi Berra over Mantle, and on and on. . .they were all won by the player with the lower OPS, either because the voters over-valued defense or because they attributed to the individual the benefit of playing on a better team.
The very first National League MVP Award, given to Frankie Frisch, is a perfect example. I believe that the true most valuable player in the National League was Wally Berger, who hit .323 with 19 homers, 84 RBI; there were a dozen other hitters who had better numbers, but Berger played in a tough hitter’s park. Berger’s numbers were still far, far more impressive than Frisch’s. Berger had 38 more hits than Frisch, twice as many doubles and triples, almost five times as many homers, more walks, a higher average and on base percentage, and a slugging percentage more than a hundred points higher. When we factor in that Frisch’s park factor was 110 and Berger’s was 97, we credit Berger with 31 Win Shares, Frisch with 21.
The MVP voters chose to set aside the hitting numbers, and honor the player who 1) played for a championship team, and 2) played a key defensive position. That’s not wrong in principle—but the offense that they chose to ignore was probably more valuable than the defense that they chose to honor. This was the pattern for the first fifty years of MVP awards, and, to an extent, remains the pattern today. But from 1931 to 1970, the winning percentage of OPS in MVP competitions was—get this--.066. 14 times in 15, if an MVP Award didn’t go to the obvious guy, it went to the player with the lower OPS.
Since 1971 it’s been .384, so that’s. ..not similar. From 2001 to 2010 the winning percentage of OPS in MVP competitions is .587—meaning that hitting may actually have been over valued in that era.
4) Focusing just on hitting, certain elements of offense are especially under-valued. Offense (as most of you know) has two essential elements—on base percentage, and slugging percentage. Each of those areas has "collateral statistics"—for example, walks, batting average and runs scored are collateral statistics for on base percentage, whereas home runs and RBI are collateral statistics for slugging percentage.
I assumed, going into this study, that we would find that on base percentage and its collateral statistics are systematically undervalued in MVP voting, whereas slugging percentage and its collateral statistics are systematically overvalued. This turns out to be not exactly true. It is certainly true that on base percentage and its collateral statistics, historically, are tremendously, tremendously undervalued in MVP voting.
Through the history of MVP voting, the winning percentage of On Base Percentage in MVP contests is .128. And it hasn’t gotten better over time. It was .134 for the first 40 years of MVP voting; it has been .119 for the last 40 years.
What I thought we would find here is a very low winning percentage for on base percentage, but a much higher winning percentage for slugging percentage. That’s not really true. What happened instead is that, in the early years of MVP voting, the voters were so willing (and eager) to simply throw batting stats out the window that it results in a very low winning percentage for any and every batting stat, with one exception which I’ll get to later. In the first 40 years of MVP voting, the winning percentage of on base percentage (in MVP voting) was .134, the winning percentage of slugging percentage was .118, and the winning percentage of OPS was .066.
Then, in the second half, the winning percentage of slugging percentage goes way up, but the winning percentage of OBA stays down. Since 1971 the winning percentage of OBA is .119, but the winning percentage of slugging is .547. The winning percentage of OPS—which combines the two, obviously—is .384. OBA/Slugging/OPS, 1931-1970: .134, .118, .066. OBA/Slugging/OPS 1971-2010: .119, .547, .384. OBA/Slugging OPS, 1931 to 2010: .128, .310, .208.
So that was really interesting; I actually came to understand something there that I didn’t understand at all going in. The first half of the history of the vote, the voters under-valued ALL offensive numbers indiscriminately. The second half, they respect slugging but not its brother, OBA.
This split is reflected as well, of course, in all of the collateral statistics. The winning percentage of runs scored in MVP contests is .204--.092 in the first half, .338 in the second half. The winning percentage of RBI is much better—.486 in the first half, .647 in the second half, .564 overall. RBI is one of the few offensive statistics which has managed to be overrated by MVP voters, despite their general tendency to underrate all categories of batting performance. But while RBI certainly are the most overrated batting stat (with one possible exception, which we’ll get to later), the impact of RBI in all MVP races has been nothing like the impact of playing a key defensive position or playing for a championship team.
Home Runs—a collateral statistics of slugging percentage--have an overall winning percentage, in MVP contests, of .405 (.226 the first half, .630 since 1971.) But walks—a collateral statistic of on base percentage—have an overall winning percentage of .139--.129 the first half, .151 the second half. Basically, walks don’t exist so far as MVP voters are concerned. That’s one reason Mickey Mantle won "only" three MVP Awards, one reason that Eddie Mathews won no MVP Awards and Ted Williams did not win several that he should have won. His walks basically did not exist insofar as the MVP voters were concerned.
The Big Surprise Finding
The big surprise finding in this research was: Strikeouts. Strikeouts, over the course of MVP history, are by far the most favored statistic for an MVP candidate. Players who don’t strike out have been treated very, very well by MVP voters.
From 1931 to 1970, the winning percentage of (fewer) strikeouts in MVP competitions was .651—by far the highest for any batting statistics. Since 1971 the winning percentage for strikeouts has dropped to .549, but still. ..combining the two, that’s .605 (463-302), still easily the highest percentage for any offensive category. A player who doesn’t strike out, historically, has a better chance of winning the MVP Award than a player who drives in runs. And if you’re a player who doesn’t strike out AND drives in runs AND plays a key defensive position, like Yogi Berra, you’re in good shape.
In the history of MVP voting there have been 51 Awards won by less-deserving players who struck out less often than the more-deserving player who might have won.
Frankie Frisch struck out less than Wally Berger (1931 NL).
Mickey Cochrane struck out less than Lou Gehrig (1933 AL).
Charlie Gehringer struck out even less than Joe DiMaggio (1937 NL).
Ernie Lombardi struck out less than Mel Ott (1938 NL).
Joe DiMaggio struck out less than Ted Williams (1941 and 1947).
Lou Boudreau struck out less than Ted Williams (1948 AL).
Jackie Robinson struck out less than Stan Musial (1949 NL).
Yogi Berra struck out less than Ted Williams (1941 AL) and less than Mickey Mantle (1954 and 1955 AL).
Roy Campanella struck out less than Eddie Mathews (1953 NL), and less than Willie Mays (1955 NL).
Jackie Jensen (1958), Nellie Fox (1959), Roger Maris (1960 and 1961) and Brooks Robinson (1964) all struck out less than Mickey Mantle, and all won MVP awards in seasons when Mantle was actually the best player in the league.
Dick Groat struck out less than Willie Mays or Eddie Mathews (1960 NL).
Maury Wills struck out less than Willie Mays (1962 NL).
Ken Boyer struck out less than Richie Allen (1964 NL).
Orlando Cepeda struck out less than Ron Santo (1967 NL).
Harmon Killebrew struck out less than Reggie Jackson (1969 AL).
Pete Rose struck out less than Joe Morgan (1973 NL).
Steve Garvey struck out less than Mike Schmidt (1974 NL).
Don Baylor struck out less than Fred Lynn (1979 AL).
Don Mattingly struck out less than Rickey Henderson (1985 AL).
Terry Pendleton struck out less than Bonds or Sandberg (1991 NL).
Barry Larkin struck out less than Barry Bonds (1995 NL).
Juan Gonzalez struck out less than A-Rod (1996 AL).
Ivan Rodriguez struck out less than Derek Jeter (1999 AL).
Justin Morneau struck out less than Derek Jeter (2006 AL).
By "struck out less" I mean that he had fewer strikeouts. I didn’t figure strikeouts per at bat. But with astonishing consistency—there are many examples besides these—disputed MVP Awards have gone to players who did not strike out. Even the newest example (Ryan Braun rather than Matt Kemp) is a continuation of this trend. There really is very little difference between them—other than the fact that Braun doesn’t strike out as much.
I do not know whether this is a direct or an indirect effect—in other words, do players who don’t strike out much do very well in MVP voting because voters prefer players who don’t strike out, or is it rather than players who don’t strike out tend to do something else that MVP voters like? I don’t know. I suspect that it’s both. Players who don’t strike out much also tend not to walk very much—think of Berra vs. Mantle, or Garvey vs. Schmidt—and, as we know, walks are a big turnoff to MVP voters. If you don’t strike out or walk you get more at bats, and you may have more RBI. You may also make more outs, but. .. .who’s counting the outs, other than the strikeouts?
Category Summary
This is a summary of the winning percentage of all categories of offensive performance:
Games Played, 1931-1970: 124-254, .327.
Games Played, 1971-2010: 159-174, .478.
Games Played, 1931-2010: 283-428, .398.
At Bats, 1931-1970: 189-236, .445.
At Bats, 1971-2010: 193-152, .559.
At Bats, 1931-2010: 382-388, .496.
Runs Scored, 1931-1970: 37-367, .092.
Runs Scored, 1971-2010: 115-225, .338.
Runs Scored, 1931-2010: 152-592, .204.
Hits, 1931-1970: 106-319, .249.
Hits, 1971-2010: 200-145, .580.
Hits, 1931-2010: 306-464, .397.
Doubles, 1931-1970: 150-260, .366.
Doubles, 1971-2010: 127-219, .367.
Doubles, 1931-2010: 277-479, .366.
Triples, 1931-1970: 96-303, .241.
Triples, 1971-2010: 136-150, .476.
Triples, 1931-2010: 232-453, .339.
Home Runs, 1931-1970: 96-327, .226.
Home Runs, 1971-2010: 212-124, .630.
Home Runs, 1931-2010: 308-451, .405.
RBI, 1931-1970: 202-214, .486.
RBI, 1971-2010: 228-118, .659.
RBI, 1931-2010: 440-332, .564.
Walks, 1931-1970: 55-368, .129.
Walks, 1971-2010: 52-293, .151
Walks, 1931-2010: 107-661, .139
Strikeouts, 1931-1970: 274-147, .651.
Strikeouts, 1971-2010: 189-155, .549.
Strikeouts, 1931-2010: 463-302, .605.
Stolen Bases, 1931-1970: 128-298, .300.
Stolen Bases, 1971-2010: 162-183, .470.
Stolen Bases, 1931-2010: 290-481, .376.
On Base Percentage, 1931-1970: 57-368, .134.
On Base Percentage, 1971-2010: 41-304, .119.
On Base Percentage, 1931-2010: 98-672, .128.
Slugging Percentage, 1931-1970: 50-375, .118.
Slugging Percentage, 1971-2010: 189-156, .547.
Slugging Percentage, 1931-2010: 239-531, .310.
OPS, 1931-1970: 28-397, .066.
OPS, 1971-2010: 133-213, .384.
OPS, 1931-1970: 161-610, .208.