by David Nemec
In most cases, pitchers’ ERAs prior to the 1912 season have been reconstructed by dedicated researchers after laboriously poring over old box scores and game accounts. In some cases, a statistic that was computed well over a century ago was taken as fact for the lack of any method for verification. No one can really be sure now that Tim Keefe, by present scoring rules, really posted an all-time record-low 0.86 ERA by a qualifier in 1880 with Troy. It seems hard to imagine that Keefe could have allowed only 10 earned runs in 105 innings and yet lose six of his twelve decisions, whereas Troy’s other Hall of Fame pitcher, Mickey Welch, won more than half his games despite a 2.54 ERA that was 0.17 runs above the National League average that season.
Even Dutch Leonard’s twentieth-century record-low 0.96 ERA in 1914 has been scaled downward from 1.01 relatively recently as new information has come to light revealing that Leonard hurled 224 ⅔ innings that season rather than 222 ⅔, the total with which he was credited for well over half a century. Indeed, the career and single-season ERAs for almost every pitcher active prior to 1920 have undergone some adjustments since the first Macmillan baseball encyclopedia appeared in 1969.
Incidentally, the career and single-season ERAs for even some contemporary pitchers differ from one record book to another. Tommy John is an example. The final edition of the Macmillan encyclopedia in 1996 lists John with 4708 ⅓ career innings pitched and ERAs of 2.97 in 1979 and 2.64 in 1981. However, Baseball-Reference.com presently credits John with 4710⅓ career innings and ERAs of 2.96 in 1979 and 2.63 in 1981.
A mistake on either’s part? Not at all. To simplify the math work, for the 1970 through 1981 seasons major-league statisticians were unwisely directed to round off the innings a pitcher worked to the nearest whole inning, only to revert in 1982 to the original rule that counted each third of an inning. Baseball-Reference.com incorporates all the thirds of an inning John had lost to the new ruling in his career and single-season stats, whereas the final Macmillan edition continued to deduct a third of an inning from John’s stats in six different seasons. This sort of discrepancy has now been resolved for all pitchers from John’s era. But as has been already discussed earlier, another quirk that developed as a result of the rule to round off innings pitched at the end of the season cost Baltimore’s Sammy Stewart the AL ERA crown in the strike-shortened 1981 campaign. Stewart allowed 29 earned runs in 112 ⅓ innings for a 2.323 ERA, while Oakland’s Steve McCatty was touched for 48 earned runs in 185 ⅔ innings for a 2.327 ERA. When the innings were rounded off to the nearest whole number, however, McCatty won 2.32 to 2.33.
9.16 (g)
When pitchers are changed during an inning, the official scorer shall not charge the relief pitcher with any run (earned or unearned) scored by a runner who was on base at the time such relief pitcher entered the game, nor for runs scored by any runner who reaches base on a fielder’s choice that puts out a runner left on base by any preceding pitcher.
One oddity of the formula in Rule 9.12 (g) for determining when to charge a relief pitcher with an earned run is that it makes it possible for a reliever to be assessed an earned run for a tally that is unearned in his team’s totals. If, say, a reliever enters in an inning when all the runs his predecessor allowed are unearned because of errors, and then promptly gives up a home run to the first batter he faces, the dinger is an earned run charged to his account, but an unearned run to the team’s. Consequently, in many instances a team will have a season ERA that is lower than the aggregate season ERA totals of its pitchers.
9.17 Winning and Losing Pitcher
(a) The official scorer shall credit as the winning pitcher that pitcher whose team assumes a lead while such pitcher is in the game, or during the inning on offense in which such pitcher is removed from the game, and does not relinquish such lead, unless (1) such pitcher is a starting pitcher and Rule 9.17 (b) applies; or (2) Rule 9.17 (c) applies.
Rule 9.17 and its five contingencies have recently replaced Rule 10.19, which some authorities feel was more straightforward and simpler to grasp. In any case, the 1950 rule book was the first to formalize what previously had only been a custom not to award a starting pitcher a victory unless he worked at least five innings in all games that went six or more innings. Before then, exceptions had occasionally been made, especially when a pitcher had to be removed after being injured. One of the most flagrant exceptions occurred in the 1924 World Series, when Giants starter Hugh McQuillan was awarded the win in Game Three even though he worked only 3 ⅔ innings because his club led Washington, 3–2, when he departed in favor of the more deserving Rosy Ryan, who would have gathered the Giants’ 6–4 win as per the 1950 rule change. McQuillan pitched again in relief twice later in the Series, negating the possibility that he suffered a disabling injury in his lone start.
In the nineteenth century, when teams often had only one standout hurler, pitching aces were frequently removed whenever they held a seemingly insurmountable lead. But removing a pitcher prior to 1889 meant having him swap positions with another player, who would then finish the game in the box. Sometimes a pitcher would be lifted as early as the second inning and sent to first base or right field on the premise that he could always be brought back into the box if the complexion of the game changed. No thought was given then to denying a hurler a win in such cases. In reviewing old records, however, some researchers have taken it upon themselves to deduct victories whenever a starting pitcher made only a token appearance.
One example where an effort has been made to rewrite history resulted in Providence’s Charley Radbourn now being listed in record books with only 59 wins in 1884 rather than 60, an all-time mark that for over a century went pretty much unchallenged. But closer examination reveals that the win in question, in a game on July 28 at Philadelphia, was, via the rule as of 1950, properly assigned to the starting pitcher Cyclone Miller, who trailed, 4–3, when he left the box after the fifth inning but led, 7–4, when Providence scored four runs in the top of the sixth, after which Radbourn came in to pitch four scoreless frames to produce an 11–4 Providence win. The issue remains, though, that in many other similar games throughout pre-1950 baseball history, the wins were given to the winning team’s most deserving pitcher, not necessarily the starter—even when he left the game with his club ahead. An even larger issue is that pitching wins and losses were only an informally kept statistic, as was a pitcher’s ERA, in the game’s embryonic years. When Cy Young finally took off his pitching toe plate for the last time in 1911, there were no headlines that proclaimed, “Young Departs with 511 Wins.” His total, in fact, has been revised several times over the years.
9.17 (c)
The official scorer shall not credit as the winning pitcher a relief pitcher who is ineffective in a brief appearance, when at least one succeeding relief pitcher pitches effectively in helping his team maintain its lead. In such a case, the official scorer shall credit as the winning pitcher the succeeding relief pitcher who was most effective, in the judgment of the official scorer.
Formerly Rule 10.17 (c), Rule 9.17 (c) can lead to some bizarre scoring decisions and in at least one instance, a decision that impacted on a Hall of Fame reliever’s career record saves total.
In a night game at Camden Yards on September 12, 2013, between Baltimore and the Yankees, with his club ahead, 5–2, Yankees manager Joe Girardi called on setup man David Robertson to pitch the bottom of the eighth inning. Robertson retired the first two Orioles hitters he faced, but then yielded two successive singles followed by a three-run homer by Danny Valencia to tie the game at 5–5. Despite next serving up a double to J. J. Hardy, Robertson escaped the frame with the score still knotted. After the Yankees eeked out a run in the top of the ninth, Yankees relief kingpin Mariano Rivera came on in the bottom of the ninth to seal a 6–5 win for his club by setting the Orioles down 1-2-3 for what virtually everyone expected would be his 653rd career save, embellishing his already record total.
But instead of charging Robertson with a blown save and
crediting him with the win, since he was still in the game when the Yankees went ahead, 6–5, in the ninth, the official scorer, Mark Jacobson, awarded the victory to Rivera, thus denying him what would have been his league-leading 44th save of the season. Jacobson’s decision was correct on all counts but one that is nonetheless not often seen, as Rule 9.17 (c) is perhaps more inconsistently applied by official scorers than any other. It gave Rivera his 82nd and final career win, but froze his record save total at 652 since he retired after the 2013 season. When and if someone approaches Rivera’s all-time saves record, it will be interesting to see how vividly this game, his last win instead of his last save, is remembered.
Here is an illustration of a more commonly seen application of Rule 9.17 (c). On August 16, 2004, at Arizona’s Bank One Park (now known as Chase Field), Pirates reliever Jose Mesa came on in the bottom of the ninth to protect a 7–3 lead that would have meant a win for starter Sean Burnett (had Mesa been successful). Instead, Mesa gave up five hits and four runs to send the game into extra innings. Pittsburgh tallied a run in the top of the 10th and Mike Gonzalez held the Diamondbacks scoreless in the bottom half. If Mark Jacobson had been the official scorer for this game, Gonzalez probably would have gotten the win, but Diamondbacks scorer Rodney Johnson awarded it to Mesa instead, even though he was colossally ineffective in the single inning he worked and Gonzalez was given his only save of the season.
9.17 (d)
A losing pitcher is a pitcher who is responsible for the run that gives the winning team a lead that the winning team does not relinquish.
Rule 9.17 (d), at a glance, seems so clear that it requires no discussion. Yet until the twentieth century was well underway it was not unusual for an official scorer to saddle a starting pitcher with a defeat when his team lost, even if he left the game with his club leading. Likewise, a pitcher sometimes would collar a win in a game that he left while his team was trailing. In a 1912 meeting on April 20 at the Polo Grounds between the Giants and Dodgers, Giants rookie Jeff Tesreau got one such victory that seemed insignificant at the time but turned out to be of monumental importance. Tesreau was relieved by Rube Marquard in the top of the ninth inning of a game in which the Giants were trailing, 3–2, after he gave up three runs in the ninth. Marquard retired the only batter he faced to end the Brooklyn rally. When the Giants tallied two runs in the bottom of the frame off Nap Rucker in relief of Eddie Stack, the official scorer put a “W” beside Tesreau’s name. Nowadays, the win would go to Marquard, who was in the game when the winning run scored. Had Marquard garnered that extra victory, it would have enabled him to launch the 1912 season by winning his 20th straight decision. Instead, he had to settle for 19 straight wins, tying the then all-time record set by Tim Keefe in 1888 rather than establishing a new standard.
Had Rule 10.19 (a) been on the books in 1912, Rube Marquard would have begun the season with 20 straight wins. As it is, his 19–0 start is the best in history.
There is a strong temptation now among some baseball historians to correct these apparent injustices. As a result, the career won and lost totals of many pitchers, including not only Cy Young but several other Hall of Famers, have been revised in the past four decades. It becomes almost a matter of personal taste whether Tom Hughes had 16 wins for the Boston Braves in 1915 (as per the final 1996 Macmillan edition and Baseball-Reference.com) or 20 wins (as per the 1982 Macmillan edition). Hughes is by no means an extreme example. In any event, many decisions made by official scorers over a span of some fifty years have since been rescinded in an effort to bring a historical uniformity to all records.
9.18 Shutouts
A shutout is a statistic credited to a pitcher who allows no runs in a game. No pitcher shall be credited with pitching a shutout unless he pitches the complete game, or unless he enters the game with none out before the opposing team has scored in the first inning, puts out the side without a run scoring and pitches the rest of the game without allowing a run. When two or more pitchers combine to pitch a shutout, the league statistician shall make a notation to that effect in the league’s official pitching records.
This rule might almost have been tailored to account for Ernie Shore, the perpetrator of the greatest one-game relief stint in history. On June 23, 1917, in the first game of a doubleheader at Fenway Park, Shore relieved Red Sox starter Babe Ruth after Ruth was booted by home-plate umpire Brick Owens for arguing a ball four call to Washington Senators leadoff hitter Ray Morgan. Enraged, Ruth charged Owens and threw a punch at him before he could be hauled off the field by a policeman. Had Ruth’s attempted haymaker landed, the course of baseball history might have been permanently altered. Sox catcher Pinch Thomas was also tossed for objecting to the call. Thomas’s replacement, Sam Agnew, gunned down Morgan trying to steal second, and Shore then retired the next 26 batters in a row and for over half a century received credit for both a shutout and a perfect game.
Shore’s perfect-game honor is still the subject of controversy. Most historians now agree that no pitcher can earn a perfecto in a game where an opposing runner has reached base safely, or a complete game for that matter when he was not the starting pitcher. But Shore’s combined no-hitter and shutout remain firm according to the rule book.
On May 31, 1988, Yankees reliever Neil Allen collected a whitewash in a similar manner when he blanked the A’s, 5–0, at Oakland. Al Leiter started for the Yanks and on his first pitch was struck on the left wrist by Carney Lansford’s smash, which resulted in a double. After Leiter was removed from the game, Allen was given all the time he needed to warm up as per the injury rule. He then preceded to toss nine innings of three-hit scoreless ball to receive credit for a shutout but not a complete game. Allen’s 1988 stats reflect that he had no complete games but one shutout. It appears to be a misprint, until Allen’s game log for the 1988 season is examined.
9.19 Saves for Relief Pitchers
A save is a statistic credited to a relief pitcher, as set forth in this Rule 9.19.
The official scorer shall credit a pitcher with a save when such pitcher meets all four of the following conditions:
(a) He is the finishing pitcher in a game won by his team;
(b) He is not the winning pitcher;
(c) He is credited with at least 1/3 of an inning pitched; and
(d) He satisfies one of the following conditions:
(1) He enters the game with a lead of no more than three runs and pitches for at least one inning;
(2) He enters the game, regardless of the count, with the potential tying run either on base, or at bat or on deck (that is, the potential tying run is either already on base or is one of the first two batters he faces);or
(3) He pitches for at least three innings.
Until 1969, the term save was not even an official part of the game’s lexicon. That season, major league rulemakers—at the urging of sportswriter Jerome Holtzman—first paid formal acknowledgement to a facet of relief pitching that many publications, The Sporting News among them, had already long since championed. But what The Sporting News deemed a save and what is now considered a save are not at all the same. In 1967, for example, The Sporting News named right-hander Minnie Rojas of the California Angels the American League “Fireman of the Year” for his 22 saves, whereas Ted Abernathy of the Chicago Cubs got the National League trophy for netting 26 saves.
Most record books now list Rojas with 27 saves in 1967 and Abernathy with 28. The reason for the disparity is because The Sporting News granted a save only when a reliever faced the tying or lead run during his mound stint or began the final inning with no more than a two-run lead and then pitched a perfect inning. In contrast, both major leagues awarded a save in 1969, the year the term got its official baptism, if a reliever merely entered the game with his team in front and held the lead for the remainder of the game. What this meant is that, by The Sporting News’s definition, a reliever in 1967 who came into a game with his team ahead 4–1 and worked three perfect innings would not get a save, b
ut the major-league rule in 1969 gave a save to a reliever who worked just the final inning of a 10–0 blowout.
Over the years, these inequities have been eliminated. In 1973, the rule was amended to give a reliever a save if he either found the potential tying or winning run on base or at the plate during his stint or else worked at least three effective innings. Two years later, the current save rule was adopted.
In determining retrospective saves for pitchers active before the concept was born, researchers applied the 1969 rule. Hence many early day pitchers are now credited with specious saves. An unfortunate example is Ted Conover, a one-game major leaguer with a career 13.50 ERA, who earned a retrospective save some eighty years later for his effort on May 7, 1889, when he replaced Cincinnati rookie starter Jesse Duryea in the eighth inning of a 16–4 Cincinnati blowout win over last-place Louisville and surrendered all four Louisville runs in his two-inning stint. Why was Conover called on at all by Reds manager Gus Schmelz? Probably to save overtaxing Duryea’s arm, but to little avail. As it was, Duryea not only lost his chance for a shutout but worked 401 innings in his frosh season and won 32 games. He won only 27 more in his brief five-season career.
When center fielder Stevie Wilkerson took the hill at Anaheim for the pitching-depleted Orioles on July 25, 2019, in the bottom of the 16th inning and notched a save in Baltimore’s 10–8 win over the Angels, the media made much to-do of his being the first position player ever to record a save. That may have been the case ever since the save became an official stat in 1969, but is otherwise untrue. In 1883, Cap Anson finished two games in the box for Chicago and collected a retrospective save in one them. But unlike Ted Conover’s save six years later, Anson’s emanated from a solid performance. In two relief appearances that year, he logged an 0.67 WHIP. Two years later, Anson’s teammate, second baseman Fred Pfeffer, bagged two saves and two wins in five appearances. Prior to 1887, when the pitching rules tightened, making it problematical for a non-pitcher to enter the box and throw with whatever delivery he pleased, Anson frequently used his position players in relief or even as spot starters. Wilkerson’s save may be the last of its kind if MLB’s roster-size increase to 26 in 2020 enables teams not to have to use position players as pitchers in crucial situations.