I realize this game has been written about quite a few times over the past 81 years, so my goal here is to try to add something to the discussion. The game and performance in question is the Athletics at Cleveland, July 10, 1932, when Eddie Rommel pitched 17 innings in relief (!) and set records that will surely never be broken for most batters faced as a relief pitcher (87), most hits allowed in a game (29) and most runs allowed by a winning pitcher (14).
My thanks to Norman L. Macht, whose article in the Society for American Baseball Research‘s 2008 publication “Batting Four Thousand: Baseball in the Western Reserve” (not available online) is the most thorough contemporary piece I have found about the game. Norman told me the Cleveland newspaper had the play-by-play of the game (which is not yet available through Retrosheet), which led me to find that the digital archives of the Cleveland Plain Dealer are online (for a fee) and allowed me to add to this post.
The three factors that made Rommel’s unusual performance possible were weather, Pennsylvania’s blue laws and the Great Depression (which could share billing with Connie Mack’s frugality). Let’s tackle them one at a time before detailing Rommel’s performance.
WEATHER: The White Sox were scheduled to play a three-game series in Philadelphia on May 9, 10 and 11, but the games on the 9th and 11th were rained out. The next time the Sox were in Philly, for what was originally scheduled as a four-game series, the first game on July 6 was also rained out. That left the A’s with three home games to make up against Chicago — which they proceeded to do, over the final three days of that series, with successive doubleheaders on July 7, 8 and 9.
Meanwhile, the A’s already had a scheduled doubleheader at home against Cleveland on July 11, so they would be playing four doubleheaders in a five-day stretch.
And they would be doing so with what was in essence a seven-man pitching staff. Well, I guess it would be eight if you include Irv Stein, who made his one and only major league appearance in the first game of the July 7 doubleheader, pitching the final three innings and giving up four runs and seven hits. Stein went on to have a long minor league career; I don’t know how much longer he was with the A’s after his one appearance.
The shortage of pitchers led A’s manager Connie Mack to do some things we would consider unusual today. For instance, George Earnshaw was pulled after giving up five runs in the first two innings of the first game on July 7, then came back to start the first game the next day and went six innings. (Earlier in the season Earnshaw had pitched two innings in relief the day after pitching a complete game.) Lefty Grove pitched the final inning of the first game on July 8 (the game Earnshaw started) to close out an A’s victory (his first appearance in three-and-a-half weeks because of a sprained ankle), then started the first game the next day and went six innings. (Grove would pitch on no days rest six times in 1932, including three more occasions later in July. He started a game on July 20, and went the distance, after having pitched in relief on each of the previous two days.)
Fortunately for him, Mack got complete games from Rube Walberg, Roy Mahaffey and Tony Freitas during the three successive doubleheaders. But the demands on his meager pitching staff would lead to a key decision he made about the game on July 10. Of course what made the situation more unusual was…
PENNSYLVANIA’S BLUE LAWS: The A’s, Phillies and Pittsburgh Pirates were prohibited from playing home games on Sundays under Pennsylvania’s blue laws that were repealed for the 1934 season. But with Sunday games typically drawing large crowds where they were played, the Pennsylvania teams could ill afford financially to be idle for the day, and of course none of the other major league teams wanted to miss out on a Sunday payday either. So the Pennsylvania teams would travel to play out of state on Sundays, even in the middle of a homestand.
On five occasions in 1932 the A’s played a road game on a Sunday when they had games scheduled both the day before and the day after in Philadelphia. Those games were:
|April 17 at Washington|
|May 8 at Cleveland|
|May 15 at Detroit|
|July 10 at Cleveland|
|July 17 at Detroit|
The A’s did play a long homestand from August 16 through September 3 that saw them idle on both Sundays, August 21 and August 28. I checked the American League’s preseason schedule and the A’s were not scheduled to play either of those days.
So after playing a doubleheader on July 9, the A’s took a train to Cleveland to play on July 10 and then took a train back to Philadelphia to play on July 11. That brings us to the third factor that led to the unusual events of July 10…
THE GREAT DEPRESSION: If the Sunday game on July 10 had been the start of a road trip, then of course the entire team would have traveled together. But money was tight. Why haul the whole group to play a single game? Why pay to transport and feed players who weren’t going to play in the game anyway? Why not let them just stay at home in Philadelphia, on their own dime?
So Connie Mack — who, of course, owned the A’s in addition to managing them — apparently took only 11 players with him to Cleveland for the July 10 game, the nine players who would start the game plus an extra catcher and an extra pitcher. I don’t know how many players Mack typically took on these one-game Sunday road trips, but no more than 11 players and no more than two pitchers appeared in any of the other such games played in 1932. Norman Macht wrote, in his article I referred to earlier, it was customary for players not expected to play in these single-game road trips to not travel.
I say Mack “apparently” took only 11 players. I’ve read a number of recent sources that make that claim, and certainly only 11 men played in the game. But there’s a line that strikes me as curious in Sam Otis’ game story in the Cleveland Plain Dealer. He wrote about Rommel’s extended performance on the mound, “None of the surplus f[l]ingers on the bench seemed as likely to check the Indians as old man Rommel himself.” That implies there were additional players on hand who could have pitched. I would dearly love to know if that is true.
Mack wasn’t really in a position to use more than two pitchers in the July 10 game anyway. Walberg and Mahaffey were scheduled to start in the doubleheader on July 11. Freitas had pitched nine innings on July 9; Grove had pitched on July 8 and July 9, a total of seven innings. And Earnshaw had pitched on July 7 and July 8, a total of eight innings. Now I don’t know if Irv Stein could have been considered or not, but aside from him Mack had just two other pitchers: Lew Krausse and Eddie Rommel.
The 20-year-old Krausse (who was not yet Lew Krausse Sr.; his son Lew Jr. would be born in 1943 and would go on to pitch in the major leagues as well) would be Mack’s starting pitcher on July 10, three days after he had pitched four innings of relief. It would be just his second start of the season (and the third of his major league career); he had pitched more than one inning in a game just five times in 1932 up to that point. He was the closest thing to a rested pitcher the A’s had (again, excluding the mysterious Stein). Rommel had pitched two innings of relief on July 8 and three innings of relief on July 9, so one might not think Mack was expecting to use him much on July 10…but, as we will see later, Mack had a history of using Rommel in unusual ways. (Rommel would exaggerate the story in later years, claiming he had pitched four innings on the 8th and six innings on the 9th. He also said he pitched batting practice and hit fungoes before the game on the 10th, which is likely true.)
The 34-year-old Rommel had twice led American League pitchers in wins in his younger days, posting a 27-13 record for the A’s in 1922 and going 21-10 in 1925, and he entered the 1932 season with 170 career wins and a .592 lifetime winning percentage. But as the A’s kept acquiring talented pitchers later in the ’20s, Rommel was reduced to a role as a spot starter and occasional reliever. When the A’s played in three consecutive World Series from 1929-31 Rommel made just two appearances, both times coming in with the A’s behind and pitching just one inning (although he got the win in one of those games after being lifted for a pinch-hitter during the A’s 10-run rally that overcame an 8-0 deficit in Game 4 of the 1929 Series). By 1932 he was an afterthought; prior to the July 10 game he had pitched just 11 games on the season, all in relief, a total of 20-1/3 innings (five of them in the previous two days), with no wins and a 6.64 ERA.
The A’s still had hopes for a fourth consecutive pennant as they opened play in Cleveland on July 10, as they were in second place, seven-and-a-half games behind the Yankees. And they got off to a strong start, scoring twice in the top of the first as future Hall of Famer Jimmie Foxx singled in one run and a second scored on an error by Cleveland pitcher Clint Brown. But the Indians moved in front before Krausse retired a batter in the bottom of the inning; Dick Porter and Johnny Burnett opened with singles, and another future Hall of Famer, Earl Averill, followed with a three-run homer. Krausse gave up another single and a walk before the inning was over but escaped further damage.
Krausse batted in the top of the second with one out and nobody on, striking out. But he stayed in the dugout in the bottom of the second, as Rommel took the mound in his place. Until I saw confirmation in the Plain Dealer play-by-play that Rommel started the inning, I figured Krausse must have gotten into trouble in the second inning before being relieved. Why in the world would Connie Mack bring in his only relief pitcher in the second inning, especially when his only relief pitcher had pitched each of the previous two days?
I also figured if Mack had decided to make the pitching change to start the bottom of the inning, he would have had Rommel bat for Krausse in the top of the second, because Eddie wasn’t a bad hitter. Just the year before Mack had started Rommel in the outfield in three games; over the previous four seasons (1928-31) Rommel had hit .247, while Krausse to that point in his major league career was 1-for-12. I’ve not seen any comment from Mack about why he made the pitching change when he did. I doubt Krausse was injured since, as we will see, he pitched in both games of the next day’s doubleheader.
Perhaps if he had known the game would go 18 innings Mack would have held off bringing in Rommel…
In any event the Indians loaded the bases against Rommel in the bottom of the second before Joe Vosmik hit into an inning-ending double play. Foxx tied the score with his 31st home run of the year (in the A’s 81st game) in the third inning, then the A’s scored twice in the fourth to take a 5-3 lead (Rommel scoring the go-ahead run after reaching base on a walk).
Cleveland came back with three in the bottom of the fourth, though, to recapture the lead, and the Indians built their advantage to 8-6 after six innings. Then the A’s exploded for seven runs in the top of the seventh (scoring all their runs after there were two out with nobody on) to take a 13-8 lead, as Foxx hit another home run…only to watch the Indians score six off Rommel in the bottom of the seventh to go back in front, 14-13. During that inning Rommel intentionally let a popped-up bunt fall to the ground in an attempt to get a double play but then made a wild throw for the A’s only error of the game, and one of the runs was unearned as a result.
Wes Ferrell — who had already won 16 games on the season for Cleveland, with the schedule barely half finished — had come on to pitch for the Indians in the seventh after Brown was replaced by Willis Hudlin, who in turn was pulled after walking the only two batters he faced. Ferrell had pitched a complete game just two days before and would go on to pitch 11-2/3 innings in relief in this one. Cleveland was almost as hard up for pitching as the A’s, having played doubleheaders on July 7 and 9 and a single game on July 8. I don’t know if the Indians had brought any more pitchers with them for this game; they had played a doubleheader in Washington the day before and were scheduled to play the A’s again in Philadelphia the next day, so perhaps they sent some of their players directly to Philadelphia rather than pay for trips to Cleveland and back.
Ferrell looked like he would record his 17th win after he retired the first two batters to face him in the top of the ninth. That’s when Cleveland first baseman Eddie Morgan let what was described as an easy roller off the bat of Philadelphia’s Jimmy Dykes go through his legs. (Harold Kaese wrote in the Boston Globe years later that Dykes said he was so convinced he would be put out to end the game that he ran to first carrying his bat.) Al Simmons then drew a walk and Foxx drove in both runners to put the A’s back on top, 15-14. Foxx’s hit is listed in the Plain Dealer play-by-play and in the Associated Press game stories I’ve seen as a single, yet Simmons scored from first. The box score shows Foxx hitting a double in the game, but the play-by-play doesn’t show any of his hits as a double. I suspect this hit was eventually ruled a double.
Cleveland’s Willie Kamm doubled to lead off the bottom of the ninth, then Rommel retired the next two batters to find himself — like Ferrell in the top of the inning — one out away from winning the game. But Johnny Burnett singled, his sixth hit of the day, to score Kamm and tie the game. A’s second baseman Dib Williams knocked down Burnett’s hit and, after fumbling the ball, threw home too late to catch Kamm. Averill followed with a single, sending Burnett to third, but A’s right fielder Mule Haas prevented the Indians from winning the game by catching Joe Vosmik’s line drive for the third out. Plain Dealer sports editor Sam Otis, in his game story, wrote, “Haas raced over to his left, lunged, pocketed the ball, somersaulted and came up with it to retire the side and send the game overtime.”
The teams would go on to play the equivalent of another game. But after combining for 30 runs in the first nine innings, the teams would total just five runs in the next nine.
Rommel and Ferrell each held their opponent scoreless from the 10th inning through the 15th. The Indians had two great chances to win the game during that time; they loaded the bases with one out in the 11th but Vosmik hit into a 4-2-3 double play, then Morgan was thrown out at home trying to score on Bill Cissell’s single in the 12th. The A’s got a man to third base with one out in both the 12th and 15th innings but failed to bring him home.
Then in the top of the 16th Foxx hit his third homer of the game (he would finish the season with 58), a two-run shot, to put the A’s in front 17-15…but Cleveland scored two of their own in the bottom of the inning to keep the game going. That makes five times in the game the Indians came back to tie or take the lead after falling behind. Mule Haas ended the bottom of the 16th with another notable catch, described as a leaping one-handed stab at the wall of a drive by Cissell that could have brought home the winning run from first base.
In the top of the 18th, after the first two A’s were retired, Foxx singled and came home to score when the next batter, shortstop Eric McNair, got credit for a double on a ball that (according to writer Harold Kaese, quoting Ferrell) bounced over the head of third baseman Kamm and then over the head of left fielder Vosmik. (Believe it or not, 16 of the A’s 18 runs in the game scored with two out.) Rommel retired the Indians in order in the bottom of the inning, striking out Morgan for the final out, and at last the game was over, at 7:06 p.m., after four hours and five minutes of what the Plain Dealer proclaimed in the next day’s front page to be “the greatest ball game in history.”
The Plain Dealer account doesn’t mention McNair’s hit bouncing over Kamm, but the hop past Vosmik was the focus of Sam Otis’ game story:
Out beyond the edges of a deepening shadow that had eked its way almost to the distant left field barrier of League Park — out in the lone patch of sunlight lingering to pester an unfortunate fielder — a crashing ground smash yesterday took a pesky hop into the setting rays of Old Sol to give the Philadelphia Athletics an eighteen-inning triumph over a never-say-die band of Indians in the most dramatic diamond struggle in Cleveland history. The score was 18 to 17….
That bad bouncing ball, hopping over the suddenly upturned arms of Joe Vosmik directly in the glare of the sun, virtually wrote finis to a game that had seen a series of most unusual thrills…
I don’t know if it’s accurate to say that Rommel got stronger as the game went along, but at least he got better results. At least one batter reached base in each of the first 12 innings Rommel pitched (at least two batters in 10 of them), but he retired the side in order in three of the last five innings and allowed just a walk in one other. Perhaps the Cleveland batters were even more tired than Rommel. Twenty of the hits off Rommel came in his first eight innings of work.
The A’s only other substitute on hand (we think), catcher Ed Madjeski, entered the game during the bottom of the seventh inning after the first four Cleveland batters got hits. He replaced Johnny Heving; the A’s regular catcher, Mickey Cochrane, had been left home to rest. After that Rommel had no choice but to stay in the game as there was no one to replace him (we think). Of course, Mack might have moved Rommel to the outfield, where, as was mentioned earlier, he had played the year before (more details about that soon). Perhaps Foxx, a strong-armed former catcher, could have been used to pitch; he did see time on the mound later in his career, pitching a perfect inning of relief for the Red Sox in 1939, then pitched in nine games, starting two (and winning one of them), in his final season in the majors with the Phillies in 1945. His major league career ERA was 1.52.
Rommel pitched 17 innings on July 10, faced 87 batters (meaning he pitched through the order nine times and went six batters into a tenth time through the order), allowed 29 hits, nine walks and 14 runs (13 earned). He struck out seven (two of them in the 18th inning) and threw two wild pitches. He also had three singles (one of four three-hit games in his career) and walked once in eight trips to the plate, scoring two runs and driving in one.
The Indians’ 33 hits (including four off Krausse) remains the major league single-game record; the 58 hits for the teams combined remains a single-game record; and Cleveland shortstop Johnny Burnett still holds the individual single-game record with nine hits (in 11 at-bats, seven singles and two doubles). Hall of Famer Foxx had six hits (two singles, a double and three homers) for 16 total bases to tie what was then the major league single-game record. It wasn’t broken until 1950, and it remained the American League mark until Josh Hamilton had four home runs and a double in a game in 2012.
Norman Macht wrote that Rommel’s nine walks allowed was a record for a relief pitcher, but it was not. That record is held by another pitcher of Connie Mack’s, Carl Ray, who walked 12 Tigers in a seven-inning relief stint in 1916, his fifth and final appearance in the major leagues. There were a total of 30 walks in that game, six issued by Ray’s predecessors for the A’s and 12 by the Detroit hurlers.
I’ve not found any comments Rommel made about his 17-inning outing at the time, but years later he told the Boston Globe’s Harold Kaese about the effect it had on him. “I didn’t throw a ball for the next three days,” Kaese quoted Rommel as saying in his January 29, 1957 story. “I couldn’t raise my arm over my head. I never was so stiff and sore.”
The win was Rommel’s first of the season and the last of his major league career. It would be nearly six weeks before he pitched again, and then it was another marathon relief appearance. On August 20, A’s starter George Earnshaw gave up two hits and two walks to the first four White Sox batters and got the hook. Rommel came on and pitched eight innings, facing “only” 33 batters. He pitched in just four games after that, including long relief appearances on consecutive days: 5-2/3 innings on September 4 followed by 7-2/3 innings on September 5.
It’s worth mentioning that Rommel was known as primarily a knuckleball pitcher, which may have reduced the wear and tear on his arm, but he was used in ways that are highly unusual even for a knuckleballer. Connie Mack seemingly had the idea that Rommel could pitch any time for any length of time.
Thanks to SABR member Tom Ruane for pointing this out, in his Retro-Review of the 1930s on Retrosheet.org (all of Tom’s Retro-Reviews are highly recommended reading): On September 6, 1931, Rommel pitched a complete game in beating the Red Sox, 5-3, in Boston. The next day, in the first game of a Labor Day doubleheader in New York, the first eleven Yankee batters reached base (eight via walk); Mack replaced starter Roy Mahaffey after five batters, yanked reliever Hank McDonald after four batters, then pulled Jim Peterson after he walked the only two men he faced. The fourth pitcher of the inning — still with nobody out — was Rommel, who had pitched a complete game the day before. And Rommel went the rest of the way in this one, pitching nine innings, facing 42 batters and allowing 14 hits in a 15-3 New York win. Then in the second game of the doubleheader, Rommel played a complete game in left field, as he did each of the next two days…probably the oddest four-day stretch of player usage in major league history.
Earlier in the 1931 season Mack had used Rommel in relief on four consecutive days (June 22-25), pitching a total of 9-2/3 innings. Later in September Rommel pitched two innings of relief in the first game of a September 14 doubleheader, then hurled a complete game victory the next day.
Going back to 1928, Rommel started and pitched seven shutout innings against the Red Sox on September 7, then pitched in relief the next day and came out of the bullpen in both games of a doubleheader the following day. Over three days he pitched 12 innings in four games (the A’s played a doubleheader each day). An earlier iron man performance came in 1925; Rommel pitched a complete game to defeat the Yankees on May 29, then picked up another win the next day when he pitched the last five innings of a 14-inning game in the second game of a doubleheader. On August 16, 1924 Rommel started both games of a doubleheader at home against Cleveland, pitching a 12-hitter to win the first game then going the first five innings of the second contest (although the box score looks wrong, if Rommel actually did pitch five innings he should have allowed five runs; the box score as published in The Sporting News says Rommel was lifted for a pinch-hitter in the bottom of the fifth and the line score indicates Cleveland had scored five runs).
But to see some really unusual usage, go back to Rommel’s big 27-win season in 1922. He started 33 games, completing 22, and frequently made relief appearances between starts. Twice, on June 27 and July 11, he started the first game of a doubleheader, pitched in relief in the second game, and was the winning pitcher in both games. On three other occasions he pitched in relief the day after pitching a complete game; on one of those occasions he had pitched in relief on each of the two days BEFORE pitching the complete game, then pitched in relief again the day after the complete game (June 8-11). To end the season, Rommel started the first game of a doubleheader on September 28 and was knocked out after an inning, came on the next day in the 12th inning of the first game of a doubleheader, gave up a run, then got the win when the A’s scored twice in the bottom of the inning, then the next day in the final game of the season (the second game of yet another doubleheader) pitched a complete game 17-hitter for his 27th win.
Using Baseball-Reference.com’s Play Index, I compiled a list of the most batters faced in a game by a relief pitcher dating back to 1916 when their database begins. You’ll notice Rommel’s name appears twice:
|1||Eddie Rommel||1932-07-10||PHA||CLE||W 18-17||17.0||29||14||13||9||7||87|
|2||Elmer Jacobs||1917-08-22||PIT||BRO||L 5-6||16.2||17||1||1||3||1||66|
|3||Tom Sheehan||1924-08-18||CIN||NYG||W 8-7||15.2||11||3||2||7||8||63|
|4||Leon Cadore||1921-05-30 (2)||BRO||BSN||L 5-6||14.1||9||3||3||3||6||55|
|5||Eddie Rommel||1923-04-27||PHA||WSH||T 10-10||11.2||14||6||4||4||2||54|
In that 1923 game, A’s starter Walk Kinney faced just five batters, retiring one, before Rommel came on to relieve him, and Eddie went the rest of the way in a game that was called by darkness after 12 innings. Rommel also pitched at least 13 innings as a starting pitcher four times, including a 15-inning game in 1925 in which he went the distance despite allowing eight runs. And he was on the losing end of the longest-ever Opening Day shutout, giving up a run in the bottom of the 15th to lose a 1-0 duel with Walter Johnson on April 13, 1926.
One name on that list above you may recognize is Leon Cadore, who the year before his 14-1/3-inning relief appearance set the all-time record by pitching to 96 batters in a 26-inning game. His mound opponent that day, Joe Oeschger, also went the distance, facing 90 batters, and he shows up on another list topped by Rommel: most hits allowed in a game (again, since 1916).
|1||Eddie Rommel||1932-07-10||PHA||CLE||W 18-17||17.0||29||14||13||9||7|
|2||Hod Lisenbee||1936-09-11||PHA||CHW||L 2-17||8.0||26||17||14||4||1|
|T3||Ted Lyons||1929-05-24||CHW||DET||L 5-6||21.0||24||6||6||2||4|
|T3||Joe Oeschger||1919-04-30||PHI||BRO||T 9-9||20.0||24||9||8||4||2|
|5||Jeff Pfeffer||1919-06-01||BRO||PHI||L 9-10||18.0||23||10||7||3||6|
Oeschger is the only major leaguer (at least since 1916) to pitch 20 or more innings in a game twice — and both games ended in ties. Hod Lisenbee’s 26-hitter (pitching for Connie Mack) was a complete game, as the A’s were on the road so the victorious White Sox did not bat in the bottom of the ninth.
Rommel is the only pitcher on that list of most hits allowed in a game who was the winning pitcher. Let’s take a look at his competition in this list, most hits allowed as a winning pitcher (also since 1916):
|1||Eddie Rommel||1932-07-10||PHA||CLE||W 18-17||17.0||29||14||13||9||7|
|T2||Syl Johnson||1930-06-03||STL||PHI||W 11-10||8.1||20||9||9||0||5|
|T2||Lefty Grove||1929-08-14||PHA||CLE||W 5-3||17.0||20||3||3||4||8|
|T4||Mort Cooper||1944-09-24 (1)||STL||PHI||W 4-3||16.0||19||3||3||5||7|
|T4||Dizzy Dean||1936-05-31||STL||CIN||W 8-7||12.0||19||7||6||2||7|
|T4||Ernie Wingard||1925-05-31||SLB||CHW||W 15-11||9.0||19||11||10||1||0|
And while we’re at it, how about most runs allowed as a winning pitcher (yes, since 1916):
|1||Eddie Rommel||1932-07-10||PHA||CLE||W 18-17||17.0||29||14||13||9||7|
|2||Gene Packard||1918-08-03 (1)||STL||PHI||W 16-12||8.1||15||12||12||3||3|
|T3||Bill Beckmann||1941-05-04||PHA||CHW||W 17-11||7.0||11||11||9||3||2|
|T3||Thornton Lee||1938-09-28||CHW||CLE||W 14-11||9.0||16||11||11||6||3|
|T3||Buck Ross||1938-08-16||PHA||BOS||W 14-11||8.2||13||11||10||3||5|
|T3||Jack Knott||1936-09-02||SLB||PHA||W 13-11||9.0||12||11||11||7||2|
|T3||Pete Donohue||1928-06-02||CIN||BSN||W 20-12||6.1||14||11||11||0||0|
|T3||Ernie Wingard||1925-05-31||SLB||CHW||W 15-11||9.0||19||11||10||1||0|
It’s worth noting three of the pitchers on that list (Rommel, Bill Beckmann and Buck Ross) got there pitching for Connie Mack, who is probably worth a full-fledged study of all the unusual pitching decisions he made. He let more than his share of guys just get hammered on the mound, especially in the A’s lean years. I’m willing to wager a disproportionate number of the ugliest/weirdest pitcher box score lines were in games Mack managed.
One record Rommel did not set on July 10 was most innings pitched in a game in relief. That honor still belongs to George Washington “Zip” Zabel, who entered a game on June 17, 1915 with two outs in the first inning for the Cubs against Brooklyn and pitched the last 18-1/3 innings in a 4-3 Cubs win. The late, great SABR member Bob Davids wrote Zabel entered the game without having had a chance to warm up after starting pitcher Bert Humphries lost a fingernail on his pitching hand when he was hit by a line drive. Zabel faced “only” 78 batters in his effort. (Bill Lamb wrote Zabel’s biography for the Society for American Baseball Research’s Baseball Biography Project.)
Rommel also did not break the record for most runs allowed as a relief pitcher. Three players gave up 16 runs in a relief appearance, including a remarkable performance by Lefty O’Doul, who later won two National League batting championships after converting to the outfield. On July 7, 1923, O’Doul pitched three innings, retired just nine of the 28 Cleveland batters he faced and allowed 16 runs, although only three were earned.
And now for the coda: what happened to the A’s after the July 10 marathon. The win in Cleveland, combined with the Yankees’ doubleheader loss to the Browns, put the A’s just six games out of first. They would get no closer. The A’s actually had a better record from that point on than they had up to that point, but the Yanks played at a phenomenal .720 clip the rest of the way.
As mentioned earlier, there was a doubleheader (one that was on the original season schedule, not caused by a rainout) back in Philadelphia on Monday, July 11, against their opponent from the 10th, the Cleveland Indians. The A’s scored 15 runs and stroked 36 hits in the double dip — and lost both games. Rube Walberg went the first eight innings in the opener, giving up eight runs, then Sunday starter Lew Krausse pitched the ninth in a 9-8 loss. (Winning pitcher Jack Russell gave up all eight A’s runs on 18 hits.) In game two Roy Mahaffey, pitching on two days’ rest, struggled through seven innings, allowing 12 runs, before Krausse came on again to pitch the last two in a 12-7 defeat. (The A’s had an individual pitcher give up 14 runs and 12 runs on successive days…wow.) Mel Harder pitched a complete game 17-hitter for the victory. Think about that…Cleveland’s winning pitchers in the doubleheader gave up 18 and 17 hits.
The next day, July 12, was Lefty Grove’s turn to pitch — on two days’ rest. It didn’t help that Earnshaw, who had three days’ rest, was not available and wouldn’t pitch again for another week; I’ve not been able to find out what his situation was. Anyway, with the pitching staff depleted, Grove had to take one for the team, going all the way, facing 46 batters and allowing 18 hits, in a 7-6 loss. It was the most hits the Hall of Famer ever allowed in a nine-inning game in his career. Grove actually took a 6-5 lead into the ninth before Eddie Morgan homered to tie the game, Luke Sewell doubled and Bill Cissell drove in the winning run. And Grove still nearly won the game anyway, but Joe Vosmik made a running catch to end the game that robbed Doc Cramer of what had all the appearances of an extra-base hit that would have scored two runs. (An item in the next day’s Plain Dealer said Cleveland’s Dick Porter led off the game with a 14-pitch at-bat, fouling off eight pitches with a full count before drawing a walk.)
You’d think that would earn a man a little rest, but no, Grove came back again the next day in relief of Mahaffey — who was starting with one day’s rest — and pitched the final four innings, allowing two runs in the top of the tenth to take a 5-3 loss that finished a four-game series sweep by the Indians. Despite going more than three weeks without pitching in late June and early July, Grove worked a career-high 291-2/3 innings during the 1932 season, completing 27 of 30 starts and making 14 relief appearances.
My sincere thanks to Retrosheet, Baseball-Reference.com and SABR for providing the research tools that made almost all of this post possible. This game may have been mentioned in the June 1936 issue of Baseball Magazine; if you have that article, please leave a comment to let us know if there is any information there that would add to or contradict anything I’ve written.