Major League Baseball Record Keeping, Part 2
As promised yesterday, here is a guide to variances between the Baseball Hall of Fame plaques and the official record of Major League Baseball. There is truly little controversy here, as the Hall has elegantly placed a rubric in the Gallery reading: “The information on these plaques was taken from sources believed to be reliable and accurate at the time it was written.” Fair enough, but there are errors galore. As Alan Schwarz wrote nearly a decade ago in the New York Times, “many of these errors wound up on the best players’ Hall of Fame plaques. Walter Johnson was believed to have won 414 games when he was inducted in 1936, but several corrections later, he was left with 417. Eddie Collins’s plaque says he collected 3,313 hits from 1906 to 1930, but the record-keeper back then apparently switched one game of Collins’s statistics with those of his teammate Buck Weaver, so he actually had two more. In the mid-1970’s, when an addition error was discovered on Tris Speaker’s official stat sheets–which are preserved on microfilm–his official career average went up to .345 from its longtime .344, two decades after his death.” History is set in wet concrete.
Major League Baseball Record Keeping, Part 2
John Thorn, Pete Palmer, and Joseph M. Wayman
Nineteenth Century Hall of Famers
Cap Anson Plaque, four batting titles; Total Baseball (TB), three
In 1879, Anson appears to have been the beneficiary of 20 extra hits, either by error or, as is commonly believed, a civic-minded Chicago official scorer. This error was found by John Tattersall in his review of newspaper box scores. In addition he (and we) incorporated the player records of four Chicago tie games, of which Anson played in two. His traditional mark of .407 was really only .317, and was so recorded in Total Baseball; however, although a modern accounting would result in an 1879 batting title for Paul Hines, with .357, we credit the championship that year to Anson, as an instance of the official Major League Baseball policy cited in Part 1 of this article (records may change, titles/awards do not).
The singular season of 1887 presents a different case. Following the long-standing directive of the Special Baseball Records Committee, we did not count walks as hits, the practice which had been the sole basis of Anson’s fourth batting championship. Note that only for this year, in which walks were aberrationally recorded as hits, and 1876, when they were aberrationally recorded as outs, didwe overturn the scoring practice of the time in favor of a modern reinterpretation of who was the batting leader. However, when Jerome Holtzman, MLB’s official historian, ruled to revers the Special Records Committee, we saw reason in his stance and went along. [His article on the subject is appended here at the conclusion.] Today Holtzman’s edict is observed largely in the breach.
Jake Beckley Plaque, .309 lifetime, also for fielding at first base, 2368 games, 23696 putouts, 25000 chances; TB, .308, 2377, 23709, 25024
These are about as close as you can get when comparing several sources prior to 1900.
Dan Brouthers Plaque, .419 average in 1887; TB, .338
This average is inflated by the aforementioned scoring practice of 1887; even if we elected to keep that practice in effect, Brouthers’ walks would have lifted him to .426.
John Clarkson Plaque, 175 losses, 2013 strikeouts, 4514 innings; TB, 178, 1978, 4536
The major difference in strikeouts occurred in 1886, where the guide had 340 and we counted only 313. We recorded two additional losses in 1894, in accordance with the scoring rules of the day. Reviewing our sources, we counted 27 more innings in 1887. His plaque bore a clear error concerning his wins in 1888. He had 33 and did not lead the league, although the plaque says 49. This was a confusion with the 49 Clarkson did win in 1889, which was also listed.
Roger Connor Plaque, hit .300 twelve times; TB, eleven
If we counted walks as hits in 1887, we would also have twelve times.
Ed Delahanty Plaque, hit. 408 in 1899; TB, .410
The discrepancy is a product of newspaper research.
Hugh Duffy Plaque, hit .438 in 1894; TB, .440
This was the highest average all-time. The 1895 guide has 236 hits while the newspaper count is 237; both had 539 at bats. Since no backup data has survived for the guide, there is no way to determine where the difference might be.
Jim Galvin Plaque, 365-311 won-lost; TB, 360-308
The plaque data includes a 4-2 mark in the 1875 NA, which has been denied major league status by the Special Baseball Records Committee. There are other smaller differences.
Billy Hamilton Plaque, 196 runs in 1894, 115 stolen bases and .338 batting average in 1891, 937 stolen bases total, 1893-1895 batting averages of .395,.399 .393, ten times scoring 100 runs; TB, 192 runs in ‘94, 111 steals and .340 batting average in ‘91, 912 lifetime steals, 1893-1895 batting averages of .380, .404, .389, eleven times scoring 100 runs
Whether Hamilton scored 196 runs or 192, he still holds the all-time record. There were small differences in stolen bases in several years. We speculate that when the folks at Cooperstown counted Hamilton’s seasons of scoring 100 or more runs, they overlooked his 1889 A.A. accomplishment.
Hughie Jennings Plaque, once hit .397; TB, .401 in 1896
The guide had 208 hits in 523 at bats, which actually computes to .398, although .397 was shown. The newspaper research showed 209 for 521.
Tim Keefe Plaque, 346 wins; TB, 342
Daguerreotypes also has 342.
Joe Kelley Plaque, .391 in 1894; TB, .393
Newspaper box score research pointed up discrepancies in the figures recorded in the guides.
King Kelly Plaque, .394 average in 1887; TB, .322
If we count walks as hits Kelly would have a mark of .393.
Tommy McCarthy Plaque, 1268 games, 109 stolen bases in 1888, 53 assists in 1893; TB, 1275 games, 93 steals, 28 assists
Newspaper research found the early stolen base figures to be inflated, especially for the American Association. His 53 assists in 1893 included many while playing at second base and shortstop. However, he did have 44 in the outfield in 1888, fourth best all-time.
Kid Nichols Plaque, 360-202 won-lost, 30 wins in 1895; TB 361-208, 26
Daguerreotypes has 361-208 as we do, including 26 wins in 1895. Nichols for years had been credited with winning 30 or more games from 1891 through 1897, but it has been conclusively shown that he had only 26 in 1895. The Spalding Guide showed games pitched (which was usually interpreted to be decisions) and percentage. They had 44 and .681 which would give a mark of 30-14. The 1942 Baseball Register showed him 30-15. ICI’s Baseball Encyclopedia had him at 26-16 in 42 complete games and 0-0 in 5 relief appearances. Ironically, the corrected figures give him 30 wins in 1898, for seven seasons of 30 victories or more—total, but not consecutive.
Amos Rusie Plaque, led in shutouts 5 times; TB, 4: Rusie pitched 7 innings of a 9-inning shutout on July 5, 1897.
Taking this shutout away, in accordance with current scoring rules (there were no official rules for crediting shutouts until 1951), gave him 2 instead of 3.
Sam Thompson Plaque, .336 lifetime, hit .400 twice; TB, .331, once)
If walks counted as hits, we would have him at .407 in 1887 and .334 lifetime.
John Ward Plaque, 158-102 won-lost, 2151 hits; TB, 164-102, 2105
Walks in 1887 accounted for 29 of Ward’s now “missing” hits. An additional 18 came from the 1890 Players League, for which newspaper research offered a lower total than the league figures. Other discrepancies were spread out over several years. We spotted 3 extra pitching wins in 1879 and again in 1883.
Twentieth Century Hall of Famers
Luke Appling Plaque, 11,569 chances accepted at shortstop; TB, 11,616
There was an addition error for his putouts in 1940. He had 307, not 257. Daguerreotypes had 11,566, not 11,569.
Jack Chesbro Plaque, 192-128 won-lost; TB, 198-132
Daguerreotypes now has 198-128. The 1947 Baseball Register—only one year after Jack’s election and the creation of his plaque—had 197-127. Still under review are two other games for which it has been argued that Chesbro should have received wins.
Ty Cobb Plaque, 4191 hits; TB, 4189
See Part 1 of this article for discussion of the doubly entered 2-for-3 game. But that is not the whole story of how Cobb’s lifetime hit total fell by two, nor of how he comes to retain his twelve batting titles. Modern research of the official day-by-day sheets revealed that Cobb had two games in 1906–on April 22 and 23–in which he went 1-for-8 that were not entered on his sheet. Finally, there was a game on July 12, 1912 (the first game of a doubleheader) in which Cobb had a run which was entered in his hit column. (If this had indeed been a hit, he would have had a 34-game hitting streak. It was really only 23 games.) Finally, under today’s rules, Cobb would not have won the batting title in 1914, because he didn’t have enough plate appearances (or at bats). However, he did win it under the rules of the day, as he won the 1910 title.
Eddie Collins Plaque, 3313 hits; TB, 3312
Collins picked up two hits in 1920 in a game for which his stats were switched with those of Buck Weaver. But he lost a hit in 1907 in a mixup with Jimmy Collins, giving Eddie a net loss of one hit.
Stan Coveleski Plaque, 214-141 won-lost, 2.88 ERA; TB, 215-142, 2.89
Daguerreotypes has 215-141. The 1942 Baseball Register has 216-142. Our ERA includes 1912, when Stan had a 3.43 era in 21 innings. The beginning of official status for ERA in the AL was 1913.
Sam Crawford Plaque, 2505 games, 2964 hits, 312 triples; TB, 2517, 2961, 309
Red Faber Plaque, 253-211 won-lost, 3.13 era; TB, 254-213, 3.15
Elmer Flick Plaque, .378 in 1900; TB, .367
The guide had 207 hits in 547 at bats, while the newspaper research showed 200 in 545.
Harry Heilmann Plaque, 2146 games; TB, 2148
Daguerreotypes has 2145 games. Heilmann appeared in 69 games in 1914, not 66. For 1912-14 the official AL averages did not count a game for any player who had all zeros in his entry. These would be pinch-runners or late-inning defensive replacements. There were a total of over 400 such uncounted appearances in this category altogether.
Walter Johnson Plaque, 414 wins; TB, 417
This research was done by Frank Williams and detailed minutely in his previously cited essay in The National Pastime of 1982. Walter had a 16-game winning streak in 1912, but one of the games (August 5) was not marked as a win on his sheet. In 1911, wins were incorrectly marked as losses on June 27 and July 9. For the AL before 1920, there were about twenty or thirty errors in awarding wins and losses in the official statistics each year.
Heinie Manush Plaque, 2009 games; TB, 2008
Joe McGinnity Plaque, won 20 seven times; TB, eight
This looks like an error made by the Hall of Fame. He won 20 in 1902 between two leagues, as shown in the 1941 Baseball Register (McGinnity was elected in 1946).
Herb Pennock Plaque, 161 losses; TB, 162
Daguerreotypes agrees with 162 losses. The 1941 Baseball Register had 161.
Eppa Rixey Plaque, 4494 innings; TB, 4494-2/3
Official stats rounded innings pitched until 1982. Total Baseball recorded thirds of innings pitched in all seasons.
Red Ruffing Plaque, led in shutouts 1938 and 1939; TB, 1939 only
Lefty Gomez had 4 shutouts in 1938, not 3. This included a rain-shortened complete game on July 15. Ruffing had 3. Also, the comment on the plaque about making the all-star team in 1937-38-39 applies to the Sporting News All-Star team, not the All-Star Game team.
Tris Speaker Plaque, .344 lifetime; TB, .345
There was an addition error on his official sheet for 1911. He had 500 at bats, not 510.
Pie Traynor Plaque, one of few players with 200 hits in a season; TB, of course, many players have had 200 or more hits in a season
Zack Wheat Plaque, 2318 games for Brooklyn, 2406 total; TB, 2322 for Brooklyn, 2410 total
There were four unrecorded games in 1911 in which Wheat appeared only as a pinch hitter.
League Batting Leaders
Until the fourth edition of Total Baseball, Major League Baseball had established no official historical record, despite its product endorsement of several statistical compendiums over the years. As a result, writers, historians, statisticians, and fans were offered a choice amongst differing annual league batting leaders, depending on the major record book favored. Those most favored by recent chroniclers have been Total Baseball and The Baseball Encyclopedia because they featured all the player and club records. Today Baseball-Reference.com vies with mlb.com for the best online presentation.
There was no official batting championship rule until 1950. Total Baseball and The Baseball Encyclopedia remedied this oversight by formulating—independently—their own guidelines for the many years of omission, each based on their own concepts of fairness and equality.
In previous editions, Total Baseball established the following criteria for batting championships: 1876-1956, qualification by having plate appearances equal to 3.1 per game times the number of scheduled games, thus conforming to the Major League Baseball practice since 1957. The batting championship criteria for The Baseball Encyclopedia over the years have been: 1876-1919, games played equal to at least 60 percent of games the team scheduled; 1920-1949, at least 100 games played, based on acceptance of the unofficial but universally assumed rule requiring appearance in at least 100 games; and 1950-present, the various changing official rule definitions.
The history of the official batting championship rule is as follows: (1) 1950-1951, 400 official times at bat; (2) 1952-1954, 400 official times at bat or, if less than 400 times at bat and by adding enough imaginary hitless at bats so as to total 400, “he still would have the highest batting average in his league, he shall be the champion batter”; (3) 1955-1956, 400 official times at bat; (4) 1957-1966, 3.1 plate appearances per game times number of scheduled games, equaling 477 in a 154-game schedule and 502 in a 162-game schedule; and (5) 1967-present, 3.1 plate appearances per scheduled game, except that “if there is any player whose average would be the highest if he were charged with the required number of plate appearances or official at bats, then that player shall be awarded the batting championship.”
In the strike-shortened season of 1972, a 156-game standard prevailed instead of the 162 scheduled games. In the strike seasons of 1981 and 1994, the rule of 3.1 plate appearances per game was applied to the number of games played by each team, rather than to those scheduled.
The early record tomes, the Spalding Record Book and the Sporting News Record Book, placed Jake Stenzel (NL) in the lead for 1893 and Nap Lajoie (AL) on top in 1905. Both Stenzel and Lajoie were the leaders during the life of the Spalding volumes, 1908-1924, and in the Sporting News volume from its debut in 1921 until 1929, when Hugh Duffy replaced Stenzel, and 1930, when Elmer Flick supplanted Lajoie. The reasoning behind the Sporting News switches was that both Stenzel and Lajoie failed to meet the unwritten criterion of a representative number of games—Stenzel had played in only 60 games and Lajoie in 65, not even half of their club’s scheduled games. Otherwise the early record books’ leaders were those endorsed by the leagues.
The Spalding Record Book in its 1917 edition made two important batting championship changes, both on the basis of mathematical errors which their editors had noted. To that time, Dan Brouthers had been tied with Cupid Childs for the highest batting average in 1892, but had been awarded the title based on his having played in more games, which was the tie-breaking guideline of the day. In 1917, however, Childs went to the front on the basis of extended batting average (calculating to extra decimal places beyond the thousandths that comprise conventional reporting of batting averages). Next, Lajoie’s average of .422 in 1901 was reduced in 1917, because of a Reach Guide typo in the hit column, to a .405 figure. The Spalding management (the once independent Reach company had long since been acquired by Spalding) should have known that in 1892, the criterion for awarding a batting championship was indeed games played, not extended batting average, and that in 1901 they would have found Lajoie’s correct (as then calculated) average in any number of newspapers. Lajoie’s .422 average was restored in 1954 in response to John Tattersall’s research, but Childs remained, until this writing, ahead of Brouthers in the NL’s Green Book. (In the fourth, official, edition of Total Baseball, Brouthers is credited as the champion because he played in more games, which was the criterion of the time—not because modern research has lowered Childs’ batting average from .335 to .317.)
During its formative years, 1876-1919, the National League omitted any mention of batting championship criteria in its published rules. Certainly, this should have been addressed before 1920. Still, batting championships were tacitly acknowledged by the league, with guidelines drawn primarily from the comments of Henry Chadwick.
From 1876 to ca. 1888, the criterion was understood to be the best seasonal performance; as expressed by Chadwick in the Spalding Guide of 1887, “an average rating of a player should be on a season’s work.” Seasonal leadership may be deduced when the league’s recognized champion was not listed first in the official averages. His preeminence was based on a representative number of games played over the season in which he excelled, as opposed to the nominal leader’s handful of games.
The yardstick between 1889 and 1919 was playing in at least 100 games. In the 1890 Spalding Guide Chadwick wrote: “With the object in view of equalizing the averages and placing the names of the batsmen who have played in 100 games or over, are given the front rank, while those who have played in 50 games and over occupy second place, and those in 25 games and over third place, and so on.”
The American League rules in its early years, 1901-1919, omitted any batting championship language. In honoring Cobb as its 1914 bat champ, the AL was undoubtedly proceeding from the guideline of best seasonal performance.
For students of the game, Joe Wayman documented the variances among the record books and encyclopedias and highlighted particularly those traditional batting champions whom previous editions of Total Baseball had toppled from the pinnacle. Even though most of these champions have been rethroned, the following thumbnail account of the debatable batting leaders will serve to explain the seeming anomaly of a batting champion whose average is lower than that of a rival.[*] Thompson becomes the official batting champion with a .372 average in accordance with e decision of the Special Baseball Records Committee not to count walks as hits. Anson would be a .347 hitter without the benefit of his walks.
1878: Abner Dalrymple captured the NL’s batting title. Modern record tomes (The Baseball Encyclopedia and Total Baseball) list Paul Hines in the top spot. The NL in 1878 did not include tie games in its official averages, while today’s record tomes count them. Thus, by counting tie games, Hines emerges with the higher average, but does not take away Dalrymple’s championship.
1879: Cap Anson, the day’s recognized batting champion, has been challenged by the moderns (The Baseball Encyclopedia, Total Baseball, Sports Encyclopedia: Baseball, and The National League Story, by Lee Allen) as to whether his .407 average is legitimate. In fact, the average was disputed as early as the 1880 DeWitt Guide, the averages for which were compiled by William Stevens of the Boston Herald, and Balldom (1914), compiled by George Moreland. Total Baseball keeps the title with Anson in recognition of the league action at the time, but reports his batting average correctly in the Player Register and does not list him in the Annual Record’s listing of top five batting averages in the 1879 NL.
1884: The official batting champion remains Orator Jim O’Rourke. Newspaper box scores, game accounts, and the results of tie games elevate King Kelly to a higher average. Tie games were not included in the official averages at the time, and how to handle them today remains a matter of controversy. However, regardless of how we treat Kelly’s 2-for-4 in his only tie-game appearance (August 11), O’Rourke is the leader.
1887: In only this season, bases on balls counted as hits. One month into the season, the American Association wanted to scrap the rule, but the NL would not consent. Total Baseball gave the championship to Sam Thompson. Without his 60 walks registering as hits, Anson batted .347.
1892: Brouthers and Childs were honored as co-champions at .335, as discussed above. Childs had the higher extended average, .3351, against Brouthers’ .3350 mark. By the day’s reasoning, however, Brouthers is the leader. As Chadwick noted in the Spalding Guide, “the lead in all cases of tie scores in base hits belongs by right to the batsman who has played in the greatest number of games, and in this case Brouthers batted in 152 games to Child’s 144.” Childs’s statistics, as compiled by ICI from newspapers, yield a batting average of .317, placing him third.
1893: Billy Hamilton’s average was higher than Duffy’s, and he would have met modern criteria for plate appearances. The NL, however, honored Duffy because he appeared in at least 100 games, which was expected of the leading players of that day. The title is thus accorded to him.
1910: Cobb is the champion, for reasons discussed amply above.
1914: Cobb, due to his proven hitting excellence, was awarded the championship because, in all reasoned probability, he would have been the leader over the full season. Consider, also, that the batting championship for the Chalmers car in 1910 for position players was based on 350 at bats. Cobb would have easily captured the hit crown on a mythical 100 game requirement, even though he was three games shy.
1926: Bubbles Hargrave topped three other questionable contenders—those not credited with at least 400 at bats. All qualified for the title, though, based on the period’s acceptance of appearance in at least 100 games. In Hargrave’s favor was his position—catcher. Over the years, catchers were considered somewhat differently when it came to handing out awards, because of the demands of their position. In fact, in order for a catcher to qualify for the fielding championship at his position, he need only have appeared in at least 77 games. Hargrave caught in 93 games.
1932: Dale Alexander had the games and at bats to satisfy the AL as to his claim. If the 400 at bat rule had been on the books, Alexander no doubt would have been inserted into the lineup until he secured eight additional at bats. If Jimmie Foxx were today recognized as champion, he would have the first of two consecutive Triple Crowns.
1938: Foxx was the AL leader—no ifs, ands, or buts about it. The AL had a rule in 1938 requiring the batting “…winner to be at least 400 times at bat.” (Reach Guide, 1939) Taffy Wright is the trivia-question champion, batting .350 in 100 games.
1940: Debs Garms raised a few eyebrows as a come-from-nowhere champion. If enough imaginary at bats to reach 400 were to be added to Garms’ total, his adjusted average would still be one point higher than that of Stan Hack, the runnerup.
1942: Ernie Lombardi was the recognized NL batting king. As a Sporting News headline advised, “Ernie’s 105 contests suffice to qualify him for a second title.” The announcement called attention to the “inquiries” which had been made regarding a Lombardi award since the AL had put in place a 400 at bat requirement. Bill James noted that the NL announced a “meritorious 400 at bat” requirement after the problematic Garms award two years earlier. NL President Frick, however, contended there was no specific bat rule and catchers, because of their demanding position, deserved special consideration. The catcher’s fielding championship by this time was based on 100 games, lessening the Frick contention. Frick may have had in mind the prior, 77-game catching requirement for fielding leadership. Thus, Frick’s reasoning could have been to create a proportion: 77 games for catchers is to 100 games for position players what 77 percent of 400 at bats (308) for catchers is to 400 at bats for position players.
1954: Ted Williams, the batter with the highest average in the AL, did not meet the official qualifications to claim the title. The 1954 official Rule 10.17 spelled out the champion as one who, with at least 400 at bats, or with fewer than 400 at bats plus enough imaginary ones to equal 400, has the higher average. Under the official rules definition, Bobby Avila is the batting leader. During the closing weeks of the campaign, Williams was aware of the rule but continued to be selective of pitches, rather than swing at those outside the strike zone simply to reach the required 400 at bats. Williams’ extended average based on adding imaginary at bats to his 386 is a .331 figure.
1981: Bill Madlock is the official batting champion by the rules of the day. Due to the strike-shortened season, the games a team played, rather than the scheduled games, were the basis for individual championships. Pittsburgh, Madlock’s club, played 103 games. Thus, 103 x 3.1 = 319 plate appearances to qualify for the batting championship. Madlock topped the required 319 PA’s by one, with 279 at bats, 4 sacrifice flies, 34 bases on balls, 3 hit by pitch, and no sacrifice hits or interference calls. Total Baseball awarded the title to Pete Rose in editions 1 and 2, based on average games played per team, then corrected the procedure in its third edition.
The record of the four defunct major leagues shows only the American Association had batting champions not agreeing with Total Baseball champions. This happened twice: in the 1884 AA, the official winner was Dude Esterbrook though his teammate Dave Orr actually batted 50 points higher; and in 1886 AA, the champ was officially Orr though Guy Hecker’s recomputed average nips him by a point. (Pete Browning also surpassed Orr and was listed as champion in earlier editions of Total Baseball, for which a guideline of 3.1 plate appearances per game was employed throughout.)
An Important Change to the Official Record of Major League Baseball
Major League Baseball’s Official Historian
Major League Baseball is pleased to announce that, beginning with this seventh edition of Total Baseball, all batting averages are recorded as they were at the time they were reported, and not in accordance with the decision of a 1968 Special Baseball Records Committee. For the sake of conformity, the committee ruled that the 1887 batting averages be recalculated and that walks not be counted as base hits (as they were that year) or as outs (as they were in 1876).
John Thorn, the eminent editor of Total Baseball, has described it as an attempt to normalize baseball’s “gloriously messy” statistical history and bring the abnormal 1887 season in line with modern statistics. It was the only season when walks were considered hits and hence skewed the averages upwards.
For example, there were eleven .400 hitters, all properly listed in the 1888 Spalding and Reach guides, the official statistical compendiums of the time. (An arithmetic check has revealed that Paul Radford, the eleventh and final such batsman, in fact batted “only” .397.) The acknowledged batting champions were Tip O’Neill, at .492, for the St. Louis Browns of the old American Association, and Cap Anson, .421, for the Chicago National League entry. (As with Radford, an arithmetic correction reduces O’Neill’s average to .485, still the all-time record).
The special committee, in deciding walks were not hits, took 50 hits away from O’Neill, dropping his average to .435. Anson, stripped of 60 hits, fell to .347 and lost his batting title, fairly won. Worse, he no longer qualified for the 3,000 Hit Club of which he was the first member.
Revisionist history is admirable when new and undisputed evidence is brought forth. But this was an abomination, an absolute falsehood and twisting of the known facts for the singular purpose of regulating history to conform to previous and subsequent standards. It was a grievous corruption. If a walk was a hit in 1887 it should stand as a hit forevermore.
The committee was formed by General William Eckert, baseball’s fourth commissioner. Eckert always had good intentions but was ill-equipped and didn’t have a schoolboy’s knowledge of the game. The day after he took office in 1965, during his first press unveiling, it was painfully apparent he was unaware the Los Angeles Dodgers had been transplanted from Brooklyn.
The committee was co-chaired by Dave Grote, public relations director of the National League and Robert Holbrook, his American League counterpart. Neither was qualified to rule on such matters. The other members were Jack Lang, secretary-treasurer of the Baseball Writers Association of America; Joseph Reichler, director of public relations of the Commissioner’s Office, and Lee Allen, the historian of the Hall of Fame.
Why the committee was formed remains a puzzle. The general belief is that it was at the request of the Macmillan Company, which was preparing a new encyclopedia, trumpeted as better and more complete than any of its predecessors. It went on sale the next year.
To heighten the launch, the committee mostly reviewed statistics accumulated in the period before 1920, “a time that was somewhat chaotic for record-keeping procedures.” Perhaps the encyclopedia’s editors were eager to find previously published errors; adjustments would strengthen the authenticity and value of the new enterprise.
The only established historians on the committee were Joe Reichler, who had been the national baseball writer for the Associated Press, subsequently elected to the writers wing [sic] of the Hall of Fame; and the distinguished Lee Allen, widely respected, the author of a half dozen noteworthy books, including delightful histories of the American and National Leagues.
Reichler knew his stuff. A stickler for accuracy at any cost, he had edited an earlier encyclopedia, published in 1962 by Ronald Press. Allen was a compulsive researcher and known for his fascinating player anecdotes of the late 19th and early 20th centuries. He also wrote a wonderful weekly column, “Cooperstown Corner,” for The Sporting News and was not concerned with current events. They agreed to the changes. However, a year before he died, Allen admitted to historian David Voigt that “past records ought not to be tampered with.”
The change in record-keeping procedure that commences with publication of this edition of Total Baseball should not be interpreted as a blanket damning of Macmillan’s The Baseball Encyclopedia. In mid-life, it became known, fondly, as the “Big Mac,” and was the final statistical authority, an enormous aid to sportswriters, book-writers, researchers, and super-fans. There were 10 editions. Sales may have approached a million copies.
Nor is this a total condemnation of Eckert’s Special Baseball Records Committee. The committee voted on 17 thorny issues and responded with good reason, with two exceptions: the 1876 scoring of walks as an at bat (if a player drew four walks he was 0 for 4), a practice that has also been restored in this edition; and the 1887 statistical butchery. A listing of the significant 1887 batting averages restored to their proper dimension follows.