I wish I could cite the exact source of this, but I believe it was Bill James who speculated the most important factor in causing a starting pitcher to tire may not be the number of innings pitched or the number of pitches thrown, but the length of time he works. If that’s true, it could be that one of the reasons we have seen such a steep decline in complete games in recent years — not the primary reason, maybe not even one of the biggest reasons, but a contributing factor — is that it’s taking longer to play nine innings these days.
With that in mind, I went to Baseball-Reference.com’s Play Index to look for the longest complete game performances by length of game, in time (as opposed to innings). A few caveats here. First of all, there are plenty of errors in the time of game data. For instance, B-R.com shows the longest game in its database (at least as I write this) is a 1941 nine-inning game between the Red Sox and Yankees that for some reason (surely a typographical error) shows up as 348 hours and 55 minutes. The attendance is also listed as 143, so you can see some odd keyboarding was going on…in fact, it looks like the time and attendance were transposed. Retrosheet, where B-R.com gets is data, shows the time of game as 2:23 with attendance of 20,935. Sometimes the typos are on Retrosheet’s end; for instance, this 1952 nine-inning complete game by Lou Kretlow is listed as 4:47, but in contemporaneous newspaper box scores the time was listed as a more-sensible 2:47.
Second, the only occasions on which time of game will equal what I call a pitcher’s “working time” (the time from his first pitch to his last) is when the home team pitcher throws a complete game and the home team doesn’t bat in the last inning (so the home team pitcher throws both the first and last pitch of the game). Obviously this excludes extra-inning games. Otherwise even a complete-game pitcher either starts pitching after the time-of-game clock begins, or finishes his work before the time-of-game clock stops. So we have no way of knowing for sure what pitcher had the longest “working time” — especially when you include some other factors. Rain delays aren’t included in time of game, so a pitcher who stays in after a rain delay could have a longer “working time” than the official time of game. And it’s possible there are pitchers who didn’t throw complete games — in marathons that went on after they were removed — who could have extremely long “working times.” We’ll look at a few of those later on.
But what we can establish is which pitchers, at least in the B-R.com database, had the longest game times in a complete game. And here’s more caveats: not all box scores include a time of game, and of course nothing before 1914 is in the database at all.
You may be surprised to learn the famous 26-inning 1920 tie in which both pitchers, Joe Oeschger and Leon Cadore, went the distance did not make the list. It was played in a relatively snappy (given the fact that it was the length of almost three normal games) 3 hours and 50 minutes, as the box score from the Boston Herald below shows:
But note the headline above the box score: “1906 Game Slower.” And directly below this box score is the box of what had previously been the longest game, by innings, in major league history, a 24-inning affair between the Athletics and the Red Sox in 1906. This box score is not yet in the Retrosheet or Baseball-Reference.com databases, and it looks like it has the record for longest complete game pitching performance by time, as both Philadelphia’s Jack Coombs and Boston’s Joe Harris went the distance in a game that took 4 hours and 47 minutes to play. Here’s the full box from the Boston Herald of September 2, 1906:
Jack Coombs was a 23-year-old rookie who had joined the A’s just two months before this game after graduating from Maine’s Colby College (hence he was typically referred to as “Colby Jack Coombs”). In this game he struck out 18 and allowed just one run in 24 innings…and note in the box score he even stole two bases!
Joe Harris was 24 and also a rookie, although he had pitched in three games late in 1905. This game was the 18th of the American League-leading 21 he lost in 1906, against just two wins. But in this game he went 21-1/3 innings without allowing a run, the A’s scoring their first run with one out in the third inning (Coombs being the one who crossed the plate) and not scoring again until there were two out in the 24th, when Socks Seybold (0-for-9 in the game to that point) tripled home Topsy Hartsel. I can’t verify this, but I believe that’s the American League record for most consecutive scoreless innings thrown in a single game…a record set by a man who never won a major league game after this and finished his brief major league career with a 3-30 (!) lifetime mark. (Joe Oeschger held Brooklyn scoreless for the final 21-2/3 innings of that 26-inning game in 1920.) Bill Nowlin has more about this game in his biography of Harris as part of the Society for American Baseball Research (SABR) Biography Project.
Who else has pitched a four-hour game? It turns out it’s been considerably more unusual to pitch for four hours than it has been to, say, pitch more than 15 innings. I used Play Index to search games since 1914 that lasted less than four hours and found 45 pitchers who went at least 16 innings, 88 who went at least 15 and 207 who went at least 14. (By the way, contemporaneous accounts of Red Faber’s 16-inning game in 1920 say the time was 3:33, not 1:33. The Al Milnar-Tommy Bridges game in 1942 came in at 2:27, not 2:00…still remarkable for 14 innings. But Walter Johnson’s 14-inning game in 1918 really was timed at 2:12, and Larry French’s 14-inning game in 1931 really was 2:17.) On the other hand, I’ve found only 20 times in the post-1914 database when a pitcher has pitched a complete game that lasted at least four hours (two of whom pitched less than 14 innings), plus a handful of other pitchers who didn’t go the distance but may have still been in their game when it reached the four-hour mark.
Here is a list generated by Play Index of pitchers who hurled a complete game in a game that lasted four hours or longer. Click on the date to see the box score. I verified the Play Index-listed game time by checking contemporaneous printed box scores or stories. There’s also a qualifying game in the Play Index database that for some reason does not show up when I do this search; I’ll mention that later. (By the way, Tom Cheney‘s 16-inning, 21-strikeout game of 1962 just missed at 3:59.)
|Al Jackson||1962-08-14||NYM||PHI||L 1-3||15.0||6||3||2||5||6||57||4:35|
|Mudcat Grant||1959-06-21 (2)||CLE||NYY||W 5-4||14.0||14||4||4||3||8||58||4:18|
|Stan Williams||1961-05-17||LAD||MLN||W 2-1||11.0||4||1||1||12||11||49||4:15|
|Bob Rush||1957-08-23||CHC||NYG||L 2-3||15.1||11||3||3||2||10||58||4:14|
|Billy Hoeft||1957-07-28 (2)||DET||NYY||L 3-4||14.1||9||4||4||5||7||58||4:13|
|Johnny Antonelli||1955-05-01||NYG||CIN||W 2-1||16.0||6||1||1||5||11||57||4:13|
|Mickey McDermott||1951-07-28||BOS||CLE||W 8-4||16.0||11||4||4||1||15||61||4:12|
|Mike Norris||1980-06-11||OAK||BAL||W 6-2||14.0||12||2||2||2||5||51||4:10|
|Warren Spahn||1963-07-02||MLN||SFG||L 0-1||15.1||9||1||1||1||2||56||4:10|
|Juan Marichal||1963-07-02||SFG||MLN||W 1-0||16.0||8||0||0||4||10||59||4:10|
|Camilo Pascual||1964-10-01||MIN||KCA||L 4-5||12.0||12||5||1||3||14||52||4:09|
|Art Nehf||1918-08-01||BSN||PIT||L 0-2||21.0||12||2||2||5||8||77||4:08|
|Carl Hubbell||1933-07-02 (1)||NYG||STL||W 1-0||18.0||6||0||0||0||12||59||4:03|
|Luis Tiant||1974-06-14||BOS||CAL||L 3-4||14.1||11||4||4||4||5||56||4:02|
|Lew Burdette||1958-07-21||MLN||STL||L 4-5||14.0||10||5||4||5||5||54||4:02|
|Saul Rogovin||1951-07-12 (2)||CHW||BOS||L 4-5||17.0||10||5||4||6||9||63||4:01|
|Ray Fisher||1920-08-27 (1)||CIN||NYG||L 4-6||17.0||18||6||6||5||4||71||4:01|
|Art Nehf||1920-08-27 (1)||NYG||CIN||W 6-4||17.0||16||4||3||2||0||66||4:01|
|Matt Keough||1980-05-17||OAK||TOR||W 4-2||14.0||5||2||1||6||8||48||4:00|
The longest complete game by time in the Baseball-Reference.com database went only 15 innings, but at 4 hours 35 minutes lasted nearly as long as the 24-inning A’s-Red Sox game of 1906. Lefty Al Jackson went the distance for the first-year New York Mets, facing 57 batters and allowing just six hits. The New York Times game story reported Jackson threw 215 pitches.
There’s an odd coincidence involving the two longest games on that list, the Al Jackson game and the Mudcat Grant game. In both, Marv Throneberry, playing first base for the losing team, committed an error in extra innings that led to a run.
The shortest game on this list in terms of innings is the 11-inning game won by Stan Williams in 4 hours 15 minutes. Stan managed to cram 206 pitches into those 11 innings, as he walked 12 and struck out 11 to run up the pitch count. His is one of only 12 games in the post-1914 database in which a pitcher reached double figures in both walks and whiffs (another one of those games will come up later in this discussion) and one of only two pitchers on that list (along with Herb Score) to have more walks than strikeouts in his double-double game. Williams is also one of just seven pitchers in the database to be the winning pitcher in a game in which he walked 12 or more (the record that category being 13, by Pete Schneider and Bud Podbielan). (ADDED 8/19/17: Now that Baseball-Reference.com has added games from 1913 to its database, it turns out there are nine pitchers who won a game while walking 12 or more, and the record is held by a fellow named Boardwalk Brown, who walked 15 — and gave up 10 hits — in beating the Tigers in 1913. Somehow he gave up only five runs.)
The winning run in the Williams game scored when Bob Lillis, batting for Williams, drew a bases-loaded walk to force in the winning run, as Braves manager Chuck Dressen outsmarted himself in a battle of wits with the Dodgers’ Walter Alston. Dressen had ordered two intentional walks to load the bases after a leadoff triple by Frank Howard; relief pitcher Seth Morehead then struck out pinch-hitter Bob Aspromonte before walking Lillis in an epic 10-pitch at-bat, with two foul balls after the count went full. Had the Braves gotten out of the inning without a run scoring, Williams would have been replaced on the mound in the 12th inning and wouldn’t have shown up in my search.
The losing pitcher was Warren Spahn, who was relieved after yielding Howard’s 11th-inning triple. The game may have already reached the four-hour mark by the time Spahn got the hook, although possibly not. Spahn pitched in another four-hour game that will come up in a moment.
The run that ended Johnny Antonelli‘s 16-inning win after 4 hours 13 minutes came about in an unusual way. Whitey Lockman led off the bottom of the 16th with a single and was sacrificed to second. Bob Hooper tried to issue an intentional walk to Don Mueller, but when what would have been the fourth ball got too close to the plate, Mueller reached out and singled to left, moving Lockman to third. Pinch-hitter Bill Taylor then hit a fly ball to deep right field that Wally Post couldn’t hang onto; it was scored a single, but even if Post had caught the ball Lockman would have scored from third.
Mickey McDermott‘s 16-inning win was likely not even his first four-hour game of the month, although the other one doesn’t show up in this search. Fifteen days earlier he had pitched the first 17 innings of a game against the White Sox that wound up going 19. That game lasted 4 hours 47 minutes, and McDermott was almost certainly still in the game when it reached the four-hour mark. In his 16-inning complete game, McDermott won despite allowing runs in both the 15th and 16th innings. Clyde Vollmer ended the game with a grand-slam home run off Bob Feller, who had come on to pitch in relief in the 15th inning. For Vollmer, that capped a streak dating back to July 4 in which he hit 13 homers and drove in 38 runs in just 24 games.
One team had two pitchers who threw four-hour complete games: Billy Martin’s 1980 Oakland A’s. Matt Keough got a 14-inning win over Toronto May 17 in exactly four hours, and Mike Norris won a 14-inning game against Baltimore June 11 in which he threw 160 pitches in 4 hours 10 minutes. The A’s had two more 14-inning performances that season; Rick Langford beat Cleveland July 20 (in 3 hours 23 minutes) and Steve McCatty lost to Seattle August 10 (in 3 hours 36 minutes). No other team since 1952 has had even two pitchers work 14 innings in the same season, and aside from Martin’s quartet no pitcher has thrown 14 innings in a game since 1974. With help from the designated hitter, Martin got 94 complete games out of his staff in 1980, the most of any major league team since the 1941 White Sox had 106. (All 30 major league teams combined had just 118 in 2014.)
Only two pitchers in the database threw four-hour shutouts, and both of them are Hall of Famers. The first was Carl Hubbell, whose 18-inning 1-0 win over the Cardinals in 1933 lasted 4 hours 3 minutes and is arguably the greatest pitching performance ever; it is, at least since 1914, the longest shutout by innings in major league history (Walter Johnson also threw an 18-inning shutout, in 1918, but it took just 2 hours 50 minutes). Tex Carleton, pitching on two days rest after winning a nine-inning complete game against the Giants, worked the first 16 innings of that game for St. Louis. Hubbell faced 59 batters, giving up just six hits (no more than one in any inning) and walking no one. Only one Cardinal reached third base. Hubbell’s “game score” (something I’ve written more about here) is the third-highest in the post-1914 database, behind only the 26-inning performances of Leon Cadore and Joe Oeschger. (“Game score” puts a premium on long performances.) Hubbell actually batted in the bottom of the 18th, with one out and runners on first and second; he hit a ground ball that led to a force out at second base, moving Jo-Jo Moore to third, and the next batter, Hughie Critz, drove in the game’s only run.
Hubbell’s win was the first game of a doubleheader. The Giants also won the second game 1-0, but it lasted just 1 hour 25 minutes.
The other four-hour shutout is a game so famous there’s even been a book written about it. Just Google Juan Marichal Warren Spahn 16 innings and you can read a whole slew of writing about this 1963 classic. Marichal came out the 1-0 winner over Spahn when Willie Mays homered with one out in the bottom of the 16th at Candlestick Park after 4 hours 10 minutes. This is actually one of two four-hour games in the post-1914 database in which both pitchers went the distance (of course, the Coombs-Harris 1906 marathon is not in the database). The first came on Aug. 27, 1920, when the Giants scored two runs in the top of the 17th to give Art Nehf a 6-4 win over Cincinnati’s Ray Fisher in 4 hours 1 minute.
That game made Nehf the only pitcher we know of, at least since 1914, to have two complete games that lasted at least four hours. In 1918, pitching for the Boston Braves, he lost a 21-inning game to Pittsburgh, 2-0. Retrosheet shows this game lasted 4 hours 8 minutes, the next day’s Boston Herald game story lists it as 4 hours 7 minutes. This game was scoreless for the first 20 innings; the only game in major league history that has gone longer without a run was the April 15, 1968 game between the Mets and Astros that was scoreless until Houston won it with a run in the bottom of the 24th. Nehf was lifted for a pinch-hitter in the bottom of the 21st, so like Stan Williams he wouldn’t have shown up in my search had his teammates come back to tie the game.
Luis Tiant was the last pitcher to work more than 14 innings in a major league game, a remarkable 1974 affair at Anaheim in which he went the distance, only to lose after 4 hours 2 minutes when Mickey Rivers scored on Denny Doyle’s double with one out in the bottom of the 15th. Nolan Ryan worked the first 13 innings for the Angels and joined Stan Williams on the strikeout-walk “double-double” list, fanning 19 and walking 10. Unfortunately no record of his pitch count survives, but it had to have been huge, as he faced 58 batters, the most of his career. The Los Angeles Times game story does note he threw 84 pitches in just the first four innings, in which he struck out nine and walked six; it’s a shame the game total wasn’t included. Ryan took a 3-1 lead into the ninth inning before Carl Yastrzemski tied the game with a two-run homer. The Angels threatened to win when they loaded the bases with one out in the bottom of the 12th, but Sox manager Darrell Johnson left Tiant in and he retired Bobby Valentine on a short fly ball and Mickey Rivers on a grounder to get out of the jam. Doyle’s double gave California relief pitcher Barry Raziano the only win of his brief major league career.
The four-hour game by Saul Rogovin is notable in that his 17-inning loss to the Red Sox came the day before the 19-inning game in which Mickey McDermott pitched 17. It’s the only time (at least since 1914) pitchers have worked 17 innings on consecutive days — and in this case the same teams were playing.
I mentioned there was a game that didn’t come up in my Play Index search that qualifies. Bob Smith pitched all 22 innings for the Braves on May 17, 1927, losing to the Cubs 4-3 in a game that lasted 4 hours 13 minutes. Smith faced 89 batters in the game, allowing 20 hits and walking nine. In the post-1914 database, only Cadore and Oeschger pitched more innings in a game or faced more batters. Bob Osborn got the win for the Cubs with 14 innings of shutout relief, the longest scoreless relief appearance in the major leagues, at least since 1914. The Cubs and Braves had played previously three days earlier (a Sunday of idleness due to blue laws and a Monday rainout intervened) in a game that went 18 innings, with Guy Bush going the distance for the victorious Cubs (that game lasted 3 hours 42 minutes).
What about “hidden” four-hour games, in which a pitcher who did not work a complete game may have been still in the game when it reached the four-hour mark? To look for candidates, I used Play Index to find pitchers who threw at least 14 innings in a game that lasted more than four hours.
Gaylord Perry pitched 16 shutout innings (making him the last major leaguer to go 16 innings; he was also the last to go 15, in a 1974 game) in a 1967 game that went 21 and lasted 5 hours 40 minutes. If all the innings were the same length Perry would have been in the game 4 hours 19 minutes. Vern Law worked the first 18 innings of a 19-inning game in 1955 (yes, he’s the last to pitch 18 in a game, or even 17 for that matter) that lasted 4 hours 44 minutes, making his prorated share 4 hours 29 minutes. Saul Rogovin (him again!) went the first 15 innings of a 17-inning game in 1952; his prorated share of the total time of 4 hours 40 minutes is 4:07. Ray Moore pitched the first 15 innings of a 16-inning game in 1957 that went 4:29; his share is 4:12. In 1934, Dizzy Dean and Tony Freitas each pitched the first 17 innings of a game that went 18; total time 4:26, their share was 4:11. Again, we have no way of knowing for sure if any of these pitchers were still in the game four hours after it started, but I would wager at least some of them were.
If you know of any other four-hour pitching performances, please leave them in the comments. Aside from the 1906 A’s-Red Sox game there could be others before 1914, and as I mentioned earlier there are games since 1914 that don’t have time of game in the Retrosheet/Play Index database, so I could be missing some.
And if you’re wondering who pitched the longest nine-inning game? That would be Mickey Lolich, whose shutout on the opening day of the 1970 season at Washington lasted 3 hours 43 minutes. Lolich faced 39 batters (only four pitchers have faced more in a nine-inning shutout since), allowing seven hits, walking five and striking out 10…and he threw 168 pitches — on opening day! Four Senators pitchers combined to throw 177.
The loss marked the eighth straight year the Senators had lost their traditional season opener in the nation’s capital…and I love the headline and subhead that appeared in the next day’s Washington Evening Star:
Randy Johnson just missed Lolich’s mark in 1990, pitching Seattle to a 13-4 win over the White Sox in Chicago that lasted 3 hours 42 minutes. Johnson gave up 10 hits, struck out 11 and threw 153 pitches; six Sox hurlers threw 192.
And what about the scenario I described earlier…the longest “working time” we can verify, a complete game won by the home team’s pitcher in which the home team did not bat in the last inning, thus the home pitcher threw both the first and last pitches? That distinction would be held by Billy O’Dell, who staggered to a 19-8 win over the Dodgers at Candlestick Park on April 16, 1962 in 3 hours 31 minutes. The Dodgers batted around in the ninth inning, scoring five runs, but Giants manager Al Dark left O’Dell in to finish what he started. O’Dell faced 47 batters in the game; only one pitcher since then has faced as many batters in a game in which he pitched nine innings or less (it didn’t happen all that many times before O’Dell, either). There were 27 hits and 14 walks in the game; O’Dell threw 172 pitches, and five Dodger hurlers combined to throw 199.
Jay Tibbs came up just short in 1989 when he pitched a complete-game 13-hitter (no one has allowed more hits in a complete game win since, and only twice since has a winning pitcher allowed more hits) against Toronto in 3 hours 30 minutes. Tibbs threw “only” 127 pitches…but three Blue Jays pitchers walked 15 Orioles (only five teams since have issued as many walks in a nine-inning game) and combined to throw 225 pitches. The win improved Tibbs’ record on the season with the Orioles to 5-0 with a 2.45 ERA after being called up from the minors. He left his next start in the fourth inning with shoulder trouble, spent the rest of the season on the disabled list and won only three more games in his major league career.
The only game in which the home pitcher threw the first and last pitch of the game to go even three hours since 2003 was Clay Buchholz‘s 2007 no-hitter, which went 3 hours 2 minutes.