Baseball’s Losingest Pitchersby Craig Burley
April 19, 2006
"It takes a pretty good pitcher to lose 20 games" — Bill James
The reason I am writing about the losingest pitchers in baseball history isn't to tear anybody down. After all, a guy who goes 0-10 is better than 90% of the guys who go 0-0. Besides, most of these guys are dead, or else are Mike Morgan. (If Mike Morgan wanted to avoid being mocked in print, he should have done better out of the pen for the '83 Blue Jays when I was first starting to watch major league baseball. If he cries himself to sleep at night and the huge pile of money he sleeps on gets damp, so be it.)
I decided to focus on the "modern" period (since the two-league system began in 1900), because pitching records are more interesting if the huge numbers of decisions by pre-1900 pitchers are factored out. I had the happy notion, as well, of inventing a new stat, so I'd have something more to write about than just "Nolan Ryan, 292... Walter Johnson, 279..." The cheapest way to invent something, of course, is to take a process that someone else invented and turn it around. So in thinking about baseball's losingest pitchers, it was a natural to take Bill James' conception of "Fibonacci Win Points" and turn it into "Fibonacci Loss Points."
Fibonacci Win Points attempt to measure not just a pitcher's compilation of wins, but also includes a measure of his quality. Invented as a quick measure of Hall of Fame quality by a pitcher, Fibonacci Win Points (or "FWP") are calculated as (wins x winning percentage) plus games over .500. A pitcher with a .618 winning percentage will have FWP equal to his wins. A pitcher with a .500 winning percentage will have FWP equal to half his wins. The measurement tries to collapse longevity and quality into a single stat, with an emphasis on quality. At any rate, there's a full breakdown of Fibonacci Win Points in James' book Whatever Happened To The Hall of Fame? including a discussion of why it's named after medieval mathematician Fibonacci.
Fibonacci Loss Points ("FLP"), on the other hand, are determined by losses times "losing percentage" (or one minus winning percentage, if you prefer) plus games under .500. A pitcher with a .382 winning percentage will have FLP equal to his losses. A .382 pitcher is a truly horrific beast, and a rare one: rare only because there are lots of pitchers out there with the stuff to go .382 and managers are always seeking to replace them.
What makes the Fibonacci method much more interesting for losses than for wins is that while a list of baseball's winningest pitchers is most made up of pitchers with very good win-loss records, a list of baseball's losingest pitchers has many of those very good pitchers on it, but also a large numer of guys who were just so-so and had long careers. The second-losingest pitcher since 1900 is Walter Johnson, but when we think of the "losingest pitchers" we don't think of Walter Johnson, who won 417 games and had a .599 winning percentage. We don't even mean Phil Niekro, who is third in losses with 274 but did have a .537 winning percentage. On the other hand, the guys with the sub-.500 records don't usually pitch for 20 seasons. We walk a fine line here.
I only looked at the top 400 pitchers in losses, so a couple may have slipped past me, but I doubt it. Everyone with 102 or more losses after 1900 was in the group.
Somewhat to my surprise, the list is dominated by very bad pitchers. Or rather, not very bad pitchers, but pitchers with very bad records. Many or even most of these guys are pitchers who handled "workhorse duties" (as Steve Treder put it to me) for some truly awful teams. Most of these unfortunate players pitched considerably better than their records indicate, but were let by bad teams with awful defenses. In that sense, even the biggest losers of all time weren't really losers except by an accident of bad luck.
I had been expecting to see about an even mix between bad pitchers and okay pitchers with long careers, but very few pitchers with good records made the list. In fact, of the top 50 pitchers in Fibonacci Loss Points, only 10 also were in the top 50 in losses, which shows you how difficult it is to be a bad pitcher and keep a job.
The top 10:
10. PEDRO RAMOS (117-160, 135 FLP). The lanky Cuban was a mainstay of the Washington Senators' and Minnesota Twins' rotations from 1955 to 1961, in which time he piled up a record of 78-112. He was nearly a .500 pitcher the rest of his career bouncing around both leagues. Known as the fastest-running pitcher of his time.
9. SID HUDSON (104-152, 138 FLP). Another long-time Washington Senator. Like most pitchers on this list, Hudson wasn't as bad as as his record shows (according to Lee Sinins, Hudson would have had a 124-132 record if he's had average luck and support from his teams). Had the bad luck to miss the war years in the military, or he would have a better record. Great Sid Hudson trivia: on "Babe Ruth Day" at Yankee Stadium in 1947, Hudson started for the Senators and beat the Yankees 1-0, scoring the game's only run himself.
8. JACK FISHER (86-139, 139 FLP). A big righthander who played with the Orioles before they were good and with the Mets when they were still bad, Fisher's most notable year was 1965 when he went 8-24 despite a pretty good 3.93 ERA.
7. LONG TOM HUGHES (132-174, 141 FLP). Hughes went 20-7 with the World Champion Red Sox in 1903, so you can imagine how bad the rest of his career was (still not as bad as numbers one and two on this list, though). Hughes is the third long-time Washington Senator in our first four pitchers.
6. JACK POWELL (184-212, 141 FLP). Powell's overall career record was 246-256, but some of that (and his best seasons) came before 1900. Powell was a strikeout pitcher without very good control, a recipe that wasn't enough to win consistently in the deadball era either. Most of his post-1900 career was spent with the St. Louis Browns.
5. JACK RUSSELL (85-141, 144 FLP). Russell was a durable right-handed starter for the Red Sox during one of their worst periods, the late 1920s and early 1930s. However, he also had a four-year stint with the hapless Senators as a reliever (he was 24-27 in Washington). Russell should have a better record than he does, but he played for such awful teams that he wound up 24 wins worse than he deserved.
4. MIKE MORGAN (141-186, 151 FLP). Mike Morgan has played for every team in the history of organized baseball, usually when they were desperate for pitching. He did pitch briefly for the successor of the second Washington Senators incarnation, the Texas Rangers. I wonder if, given the Rangers' franchise-long struggle to find good pitching and the historical futility of both Senators clubs, that there might be some Curse Of The Big Train at work?
3. BOB FRIEND (197-230, 157 FLP). I can't do any better describing Friend than Bill James does in the essay on him in The Neyer/James Guide To Pitchers. Friend is easily the best pitcher in this list, and he had horrible luck in his career. He, Jack Powell (who won 23 games three times) and Tom Hughes are the only 20-game winners on the top 10 list.
2. SI JOHNSON (101-165, 166 FLP). Si Johnson should have been a Senator, but he pitched in the wrong league his whole career (for the Reds, Cardinals, Phillies and Braves) in the 1930s and 1940s. Johnson had one central skill that big-losing pitchers don't usually have: above-average control and above-average strikeout-to-walk ratios. He also gave fewer home runs than average. The defenses behind him, though, were generally abysmal, and his offenses couldn't score many runs. In the middle of his career, he spent two-and-a-half years with the remnants of the Gas House Gang Cardinals, still a decent team. There he had his best season, going 12-12 in 1937 with a good ERA.
1. MILT GASTON (97-164, 170 FLP). Our Fibonacci Loss Points champion is, naturally, another former Senator. Still, Milt Gaston spent only one year in the nation's capital, and for most of his career toiled for the Browns, Red Sox, and White Sox. He also put in one year (1924) with the Yankees, meaning that in 11 major league seasons (during which he averaged nearly 15 losses a year) Gaston played for five of the eight American League teams! All that bouncing around, incidentally, allowed Gaston to set a record for having the most Hall of Fame teammates or managers in his career—18 of them in all, despite playing for some repellently awful ballclubs. As befits the losingest pitcher on our list, Gaston was also the unluckiest...his "neutral record" according to Lee Sinins was just 128-133, another decent pitcher undone by his teams.
The Top 50 Pitchers Of All Time in FLP
Rk W L FLP Rk W L FLP 1 Milt Gaston 97 164 170 26 Bob Smith 106 139 112 2 Si Johnson 101 165 166 27 Ron Kline 114 144 110 3 Bob Friend 197 230 157 28 Eddie Smith 73 113 109 4 Mike Morgan 141 186 151 29 Charlie Hough 216 216 108 5 Jack Russell 85 141 144 30 Bob Rush 127 152 108 6 Jack Powell 184 212 141 31 Mike Moore 161 176 107 7 Long Tom Hughes 132 174 141 32 Eppa Rixey 266 251 107 8 Jack Fisher 86 139 139 33 Nolan Ryan 324 292 106 9 Sid Hudson 104 152 138 34 Danny Darwin 171 182 105 10 Pedro Ramos 117 160 135 35 Ed Brandt 121 146 105 11 Bobo Newsom 211 222 125 36 Ray Benge 101 130 102 12 Ken Raffensberger 119 154 122 37 Jose DeLeon 86 119 102 13 Buster Brown 51 103 121 38 Joe Oeschger 82 116 102 14 Slim Harriss 95 135 119 39 Vern Kennedy 104 132 102 15 Harry McIntire 71 117 119 40 Murry Dickson 172 181 102 16 Jim Clancy 140 167 118 41 Tom Zachary 186 191 102 17 Kaiser Wilhelm 56 105 117 42 Ray Burris 108 134 100 18 Don Cardwell 102 138 115 43 Bob Harmon 107 133 100 19 Rick Honeycutt 109 143 115 44 Kevin Gross 142 158 99 20 Bob Groom 119 150 115 45 Tom Candiotti 151 164 98 21 Ned Garver 129 157 114 46 Bobby Witt 142 157 97 22 Jimmy Ring 118 149 114 47 Harry Howell 116 138 97 23 Danny MacFayden 132 159 114 48 Dick Ellsworth 115 137 96 24 Earl Hamilton 116 147 113 49 Claude Osteen 196 195 96 25 Frank Tanana 240 236 113 50 Al Hollingsworth 70 104 96
It may be that the "Fibonacci" method, in which pitchers get most of their value from their games under .500, makes the baseline for getting Loss Points too awful. The highest pitcher over .500 was Frank Tanana, whose 240-236 record gave him 113 FLP for 25th on the list. Nevertheless, this is a fun exercise that produces some interesting answers. Strangely, no active pitchers placed in the top 50; the closest is Terry Mulholland whose 124-142 record is about an 0-2 record away from the top 50 (he is 55th). Next is Steve Trachsel, whose 119-135 record gives him 88 FLP; he needs another 6-9 or so. The top active pitchers after Trachsel are Jason Johnson (52-86), Jose Mesa (77-101), and Glendon Rusch (57-86). Given today's huge pitching staffs and 30 teams, one might expect to see more big losers going forward, but there are three factors that counteract this (greater competitive balance, large bullpens where most of the bad pitchers are put, and free agency making it easier to escape an awful team) which seem to be enough to keep current pitchers away from threatening these sorts of records.
Unless Sir Sidney Ponson somehow stays out of the penitentiary, of course.
References and Resources
Many thanks to Bill James for providing the inspiration for this article and to Lee Sinins' Sabermetric Encyclopedia for providing the data.
Craig Burley can be contacted via e-mail.