The Luckiest Pitching Staff in History
“Colby Jack” Coombs was one of the best pitchers in baseball in 1910. The ace of Connie Mack’s titanic Athletics, Coombs compiled a 31–9 record, leading the American League in wins and shutouts (13), and placing among the top-five in virtually every other important pitching category. Coombs was extraordinarily effective (182 ERA+) and simply indomitable: After working more than 350 innings during the regular season, the right-hander took the mound three times in the World Series, pitching three complete games in a span of five days (including the series-clincher in Game Five).
Coombs went from arguably the best pitcher in baseball to inarguably the luckiest pitcher in baseball the following season: Despite posting an ERA (3.53) 11% worse than league average, surrendering more hits (360), and coughing up more earned runs (132) than any other pitcher in the league, Coombs went 28–12 for those same mighty Athletics. Coombs was the beneficiary of a powerhouse offense that led the league in batting, on-base and slugging percentages, home runs, and runs scored (and this offense averaged nearly 7 runs per game for Lucky — er — Colby Jack). Coombs’ smoke-and-mirrors act didn’t go unnoticed by observers of the game. Wrote the Sporting Life on September 2, 1911: “The Athletics won both games of the double bill, the first by a 7 to 5 count, simply because Jack Coombs had his usual number of horseshoes with him.”
Coombs redeemed himself in the 1911 World Series, pitching to a 1.35 ERA over two starts and 20 innings (Coombs outdueled Christy Mathewson in a thrilling Game Three which saw both pitchers go the distance in a 3–2, 11-inning Athletics victory over the Giants). But there’s no sugar-coating the fact that his 1911 is probably the luckiest season ever by a pitcher, and Coombs had one of the luckiest careers ever, finishing with an excellent 158–110 record (.590 winning %) despite a career ERA+ (99) that suggests he was a league-average pitcher.
As you may remember from a certain supermodel, Boston’s Rick Porcello claimed the 2016 AL Cy Young award largely, it seemed, on the strength of his league-leading 22 wins. Now, let’s be clear: Rick Porcello was very good in 2016. Rick Porcello in 2016 was much better, relative to his league, than 1911 Jack Coombs. Rick Porcello was not better than Cy Young runner-up Justin Verlander, who despite posting (slightly) superior numbers in every relevant category was only able to secure 16 wins.
In other words, there was an element of luck (and league-leading run support, and excellent team defense) to Rick Porcello’s 22 wins. Was it one of the luckiest seasons ever for a pitcher? Not even close. But it underscores the absurdity of assessing a pitcher’s performance on the strength of his record — and got us to wondering about the most illusory records ever for a starting pitcher.
Our criteria is straightforward: We compare a pitcher’s adjusted ERA (which essentially compares a pitcher’s earned run average to the league average, taking park effects into account) and secondary stats to his won-loss record. When the two don’t align, we can reasonably conclude that there was an element of luck at play.
Who then joins Colby Jack Coombs on our “luckiest starting rotation of all time?”
Christy Mathewson, 1914: 24–13, 88 ERA +
College educated, blessed with matinee-idol looks and raised with a gentleman’s disposition, Christy Mathewson stood in stark contrast to the uneducated, rough-hewn men who played the game in the first decades of the 20th century. He was the most beloved player of his day, and remains one of the 25 or so best pitchers of all-time: He holds National League career records for wins (tied with Pete Alexander), winning percentage, and career ERA.
But in 1914, his last full season in the Majors, “Big Six” was a below-average pitcher. The impeccable control was still there (a league-leading 0.7 BB/9IP), but the velocity and command were clearly gone: Opposing hitters crushed 16 home runs off Matty — the most in the league and an extraordinary total for the day (Jimmy Lavender — great name, pedestrian pitcher — was runner-up in this category, surrendering 11 HR). In all, Matty allowed 104 earned runs — most in the league and by far the most in any season of his superb career.
Despite his subpar performance, Matty was rewarded with a sterling 24–13 record (.650 winning %). How did he do it? Well, we might point to a few illustrative games:
- June 1: Allows 16 hits and seven earned runs, but his NY Giants score 13.
- July 11: Allows nine runs (eight earned), but gets the win because the Giants score 13.
- July 14: Surrenders six earned runs in six innings, but his lineup scores 12.
- Aug. 29: Five earned runs on 12 hits allowed, but he’s bailed out again when the Giants score 7.
- Sept. 30: Allows six earned in three innings, but avoids the loss when the Giants rally for a 7–7 tie.
Simply put, the NY Giants scored bushels of runs for their on-field leader. Mathewson’s inclusion on this staff shows that “great” and “lucky” aren’t mutually exclusive.
Lefty Gomez, 1932: 24–7, 97 ERA+
With an adjusted ERA just slightly below league average, Hall-of-Famer Gomez owns the “best” season on this somewhat dubious list (Gomez, in fact, received significant MVP support, placing a misguided fifth on the 1932 ballot — ahead of teammate Babe Ruth, who hit .341/.489/.661).
The 1932 Yankees (107–47) were a remorseless juggernaut of a team, storming the AL pennant by 13 games before dismissively sweeping the Cubs in the World Series. A pitcher with an adjusted ERA of about league average would theoretically finish with a .500 record, give or take a game; Gomez, with an offensive hurricane at his back, finished 17games above .500. Lefty surrendered five or more runs in 13 of his 31 starts; his record in those games? Six wins, four losses, three no-decisions. If you fashion a 6–4 record from an ERA north of seven, you know things are breaking your way (or in this case, you know Ruth, Gehrig, Lazzeri, Combs and Dickey are taking the field behind you).
The 23-year old, who would go on to claim two Triple Crowns for pitching, saved his best effort for his last start of the season, going the distance and surrendering one earned run in Game Two of the World Series.
Roxie Lawson, 1937: 18–7, 89 ERA+
Alfred “Roxie” Lawson’s career year reads as a statistical horror story: Pitching to an ERA of 5.26 (and an adjusted ERA 11 percentage points worse than league average), Lawson allowed 352 baserunners in 217 innings (his WHIP, if you’re counting, was 1.615). When he wasn’t getting hit hard, he was escorting batters to first base, issuing 115 free passes. You might put up with this sort of wildness from a young Nolan Ryan-type trying to harness his stuff and find his command — but Lawson was no Nolan Ryan. He struck out a scant 68 batters the entire season (for a SO/BB rate of 0.59).
Lawson’s ranks among the 31 qualifiers for the ERA title:
- ERA: 27 out of 31
- ERA+: 27 out of 31
- WHIP: 27 out of 31
- FIP: 28 out of 31
- K/BB: 31 out of 31
- BB/9: 27 out of 31
You get it: It was a rocky season for Roxie. Despite this morass of mediocrity, Lawson was rewarded with an 18–7 record (.720 winning %) and a 19th-place showing on the 1937 MVP ballot (a ballot that ignored rotation-mate Tommy Bridges, who produced a 17–9 record with an ERA a run-and-half lower than Lawson’s). Like the others on this list, Roxie was the beneficiary of timely run support: Lawson was clubbed for five runs or more in 17 of his 37 appearances (29 starts) — yet escaped relatively unscathed with a 4–4 record (with nine no-decisions).
Roxie had moxie: He would somehow fashion a career record of 47–39, despite walking twice as many batters as he whiffed, and pitching to a 5.37 ERA that registered 11% worse than the league average over that time (it should be said that the 1930s were a brutal time to be pitcher; it should also be said that even within the context of the toughest offensive environment ever, Lawson was pretty bad).
Storm Davis, 1989: 19–7, 85 ERA+
The 1989 Oakland Athletics are remembered today for two things: 1) They won the “Earthquake” World Series with a sweep of their cross-Bay rivals, the San Francisco Giants; 2) They fielded the game’s biggest (literally) lineup: The Canseco-McGwire “Bash Boys.” While Oakland did field a seemingly endless parade of mesomorphic sluggers, it was their superb pitching that set them apart from the rest of the American League. Led by glowering ace Dave Stewart, Oakland’s team ERA of 3.09 paced the league by a significant margin. The weak link on this formidable staff was Storm Davis — though you wouldn’t know it by his wonderful 19–7 record (.731 winning %).
How lucky was Davis? Among the 39 pitchers who qualified for the ERA title, Davis’ bloated 4.36 mark ranked 33rd in the league; his adjusted ERA (85) ranked 34th. This might be forgiven if he was a workhorse, but Davis pitched only 169.1 innings in 1989 (allowing plenty of baserunners along the way: his 1.506 WHIP ranked 36th in the league). Despite averaging just over five innings per start — the minimum required of a starter to qualify for a win — Davis was somehow involved in 19 winning decisions. His pWAR barely registers at 0.2. He was simply on the right team at the right time.
Davis’ 19–7 mark didn’t fool Manager Tony LaRussa, who was taking no chances in the post-season. Despite having the best record on the team, Davis didn’t throw a single pitch in the 1989 World Series.
Coombs, Mathewson, Gomez, Lawson and Davis: Our all-time luckiest staff registers for a combined 113–46 record, despite an ERA 10% worse than their respective leagues. As the saying goes, it’s sometimes better to be lucky than good.
— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —
[i] Well, as much as a 28-game winner needs to “redeem” himself
[ii] Mathewson twice posted superb W-L records despite a sub-par performance. In 1906, ‘Big Six’ went 22–12, but with an adjusted ERA 12% worse than the league average
[iii] Though we greatly prefer Branch Rickey’s “Luck is the residue of hard work and design.”