Better to be lucky than good (Part 1)

Born under a bad sign.
I’ve been down since I began to crawl.
If it wasn’t for bad luck,
I wouldn’t have no luck at all.

It’s mostly understood these days that a pitcher’s win-loss record doesn’t necessarily reflect how well he has performed. There are too many factors beyond his control—how his offense performs that day, the defense behind him etc.—that may contribute to success (or failure).

History is rife with examples of pitchers who posted shiny win-loss records in an individual season despite less than stellar efforts, and vice versa. Two that immediately leap to mind are Jim Deshaies (11-6, 4.62 ERA, 85 ERA+) and Nolan Ryan (8-16, 2.76 ERA, 142 ERA+), teammates with the 1987 Astros.

There are many others, though, and today I thought we’d look back at some of them.

The method

The basic approach was to find guys that had a really good ERA+ (150 or better) and a lousy record (below .500). On the flip side, I targeted pitchers with a winning percentage of .600 or better and an ERA+ of 80 or lower. Only pitchers that qualified for their league’s ERA titled were considered.

My first attempt involved pitchers from 1901 to the present, but that created some problems (and pointed out a flaw in using ERA+ as a measuring stick—a flaw that I’m comfortable with, but a flaw nonetheless). Our top two pitchers on this list are Ed Siever (8-11, 196 ERA+ in 1902), and Ed Walsh (18-20, 189 ERA+ in 1910). The trouble is that 45 percent of the runs Siever allowed were unearned, as were 42 percent of those surrendered by Walsh. Baseball doesn’t work like that anymore, hasn’t for a long time.

So, I narrowed my search. I focused on the years 1961 – 2009, which yielded 13 pitchers (including one who reached both extremes—in consecutive seasons, no less!). We’ll look at the former group in this installment, saving the latter for part 2. To the names, presented in order of appearance…

Born under a bad sign

Dave Roberts, San Diego Padres, 1971
14-17, 2.10 ERA, 157 ERA+

Roberts finished second in the National League in ERA in ’71. Unfortunately, his team averaged 3.02 runs per game that year. Oh, what Roberts could have done with so much offense. When he toed the slab, the Padres gave him a meager 2.30 runs per game. Five times in his 34 starts, they failed to score at all.

In his 14 wins, Roberts posted a ridiculous 1.24 ERA. In his 17 losses, he posted a respectable 3.12 ERA (league average was 3.47). Think about that for a moment: Roberts went 0-17 in those games despite having a better-than-average ERA. Here are some of his toughest losses of the season:

May 17 at Astros: 7 IP, 5 H, 1 R, 0 ER, 1 BB, 7 SO
June 16 at Expos: 8 IP, 8 H, 2 R, 2 ER, 1 BB, 3 SO
July 3 vs Dodgers: 9 IP, 6 H, 1 R, 1 ER, 2 BB, 4 SO
Aug. 21 at Mets: 8.2 IP, 3 H, 2 R, 2 ER, 1 BB 7 SO
Sept. 6 vs Reds: 7 IP, 4 H, 1 R, 0 ER, 2 BB, 1 SO

That’s 0-5 with a 1.13 ERA in case you’re wondering. Heck, twice Roberts lost without allowing even a single earned run.

Sammy Stewart, Baltimore Orioles, 1981
4-8, 2.32 ERA, 156 ERA+

This one is a little problematic, as Stewart qualified for the ERA title despite making only three starts. But when Earl Weaver is managing your club, unusual things can happen. Stewart, for example, had seven relief appearances in which he worked five or more innings.

As for Stewart’s three starts, he lost them all. Then again, he didn’t pitch very well, posting a 6.89 ERA. Fortunately, Weaver mainly used Stewart out of the bullpen, where he went 4-5 with a 1.58 ERA.

Because he wasn’t a full-time starter, it’s hard to pinpont any real “tough-luck” outings for Stewart. The only one I could find came on Aug. 31 against the Mariners.

A Hardball Times Update
Goodbye for now.

In that contest, Stewart replaced starter Scott McGregor with two outs in the second and the Orioles trailing, 3-1. Baltimore tied the score on a two-run homer by John Lowenstein in the sixth. Stewart then coughed up a solo shot to Lenny Randle with two out in the ninth, which proved to be the deciding run.

One mistake, one loss. Tough luck.

Joe Magrane, St. Louis Cardinals, 1988
5-9, 2.18 ERA, 160 ERA+

I was a huge Magrane fan back in the day. When he first came up, I thought he would turn into something special. By age 24, he’d already won an ERA title, had an 18-win season, and owned a career ERA+ of 129. Then the injuries hit, and a once-promising career never materialized. Alas, ’tis the fate of many a young pitcher.

Magrane led the NL in ERA in ’88, his first full big-league season. He barely (165.1 IP) qualified for the title, but that’s good enough for our purposes.

The Cardinals weren’t terrible (76-86) that year but they didn’t score a ton of runs (3.57 per game). Tom Brunansky was the only real offensive force, and he started the season in Minnesota. As weak as the Cards’ offense was in ’88, it kicked things into another gear with Magrane on the mound, averaging 2.56 runs per game in his starts—a full run less than overall. (Teammate Jose DeLeon, meanwhile, had it good, going 13-10 with a 3.67 ERA thanks to 3.91 runs per game behind him; then again, he’d gone 2-19 a few years earlier, so maybe he had some breaks coming his way.)

Returning to Magrane, the Cardinals didn’t score in four of his 24 starts in 1988. He pitched beautifully in each of those games; his aggregate line: 31 IP, 18 H, 6 R, 5 ER, 10 BB, 25 SO. Yep, he went 0-4 with a 1.45 ERA. Funny what’ll happen when your team’s offense doesn’t show up.

Magrane posted a sparkly 0.61 ERA in his five victories. In his other 19 starts, he went 0-9 with a 2.74 ERA; the league ERA that year was 3.45. Granted, he allowed 17 unearned runs in those 19 games, which puts him at 4.01 runs allowed per game. The league average was 3.88, so you’d think he would have won at least one of those games (if not a handful).

Curt Schilling, Arizona Diamondbacks, 2003
8-9, 2.95 ERA, 159 ERA+

Like Stewart and Magrane, Schilling didn’t work a lot of innings (168). Schilling actually pitched for a winner (84-78) that scored runs (4.43 per game; slightly lower than league average of 4.60 but not unreasonable).

With Schilling on the mound, Arizona managed just 3.40 runs per game. The Diamondbacks failed to score for him twice, and both times he pitched gems, taking one loss and one no-decision.

In the first game, July 27 against the Dodgers, Schilling gave up a single run in eight innings and lost. In the second, Sept. 1 against the Giants, he went eight scoreless and saw his team fall in the ninth. Schilling’s combined line from those two starts: 16 IP, 13 H, 1 R, 1 ER, 0 BB, 17 SO.

Unlike Roberts and Magrane, who pitched well in defeat, Schilling typically did not. He was 8-0 with a 0.43 ERA in his victories, but when he lost, his ERA ballooned to 5.03—not very good, but not necessarily deserving of an 0-9 record.

Ben Sheets, Milwaukee Brewers, 2004
12-14, 2.70 ERA, 162 ERA+

The very next year, Sheets became the fifth member of our exclusive club. He had a monster campaign for a very bad team. The Brewers went 67-94 that year, scoring 3.94 runs per game (against a league average of 4.64 R/G; only the Expos and Diamondbacks scored fewer in the NL). Lyle Overbay was the offensive threat.

Milwaukee’s hitters hibernated a little (3.54 R/G) when Sheets took his turn, but not to such an extreme degree as had occurred with, say, Magrane or Schilling. The Brewers were shut out four times in Sheets’ starts. He pitched well in three of them (June 24 vs Rockies, July 17 at Cubs, Sep 7 at Pirates) and not so well well in the other (July 22 at Cardinals).

Sheets’ toughest loss probably came in a Sept. 17 contest at Houston. He went the distance, but his team couldn’t muster more than one run (thanks to a Geoff Jenkins solo homer) on 10 hits against Roy Oswalt and Brad Lidge. Sheets’ final line for the game: 8 IP, 4 H, 2 R, 2 ER, 0 BB, 9 SO.

However, if one start captured the essence of Sheets’ 2004 campaign, it would have to be his June 8 outing at Anaheim. The Brewers ended up winning in 17 innings, but Sheets got hung with one of the toughest no-decisions of the decade (9 IP, 1 H, 0 R, 0 ER, 0 BB, 5 SO). Since 2000, only Arizona’s Randy Johnson (May 8, 2001, against the Reds) and the Cubs’ Mark Prior (Sept. 30, 2004, against the Reds) have posted higher Game Scores (Johnson, 97; Prior, 92) in a no-decision than Sheets’ 90. Not that he’ll get any sympathy from Vern Law, but still…

Sheets posted a fine 1.25 ERA in his 12 wins. In his 14 losses, it shot all the way up to 3.70 (league average was 4.31; Sheets’ ERA in losses would have placed him 19th in the NL, just ahead of Jason Marquis, who went 15-7 with his 3.71 ERA). With any kind of support, Sheets easily could have won 20 games pitching the way he did.

* * *

Those are the pitchers born under a bad sign. Next time, we’ll look at the fortunate sons.

References & Resources
Baseball-Reference.


8 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Gilbert
14 years ago

I presume Roberts was the #1 starter for SD and since they had a more balanced schedule might have faced a lot of opposing #1’s, so his team might have been playing at the same level but just scored fewer runs due to the opposition’s pitching quality.

I remember Stewart/McCatty as the rule changer for calculating ERA; they used to round off the fractional IP which put McCatty ahead.  Looking at BB-Ref I hadn’t remembered that they lowered the qualifications due to the midseason strike and that Yankee callup Righetti (105.1 IP, 2.05 ERA) beat both Stewart (112.1 IP, 2.32) and McCatty (185.2, 2.33).  If you rounded Stewart’s IP to 112 and McCatty’s to 186 then Stewart was at 2.33 and McCatty at 2.32.
McCatty only gave up 140 H in those 185.2 IP (only Righetti had a lower rate in the AL that year) with only 91 strikeouts, a testament to the range of the Henderson/Murphy/Armas OF as their infield was nothing special.

Lucas A.
14 years ago

Sheets’ K/BB rate in 2004 was 8.25, 10th best single-season all time.  2.60 ERA, 2.65 FIP … and all he got was one third-place Cy Young vote.

Brandon Tingley
14 years ago

Hard to believe you left out Nolan Ryan’s 1987 season, where he led the league in ERA and SOs, only gave up 10 unearned runs—and was second in losses, 1 behind teammate Bob Knepper, who had an ERA more than TWICE as high as Ryan’s. He only gave up more than 3 ERs TWICE in 34 starts!

Brandon Tingley
14 years ago

OK, my bad, 3 times. And he had two games where he pitched less than five innings (although he gave up only 1 ER in each)

Brandon Tingley
14 years ago

OK, nevermind, I just can’t read :>

David Zientara
14 years ago

The 150 ERA+ cutoff just barely excludes one of the outstanding campaigns of the Mets’ ultimate hard-luck pitcher, Jon Matlack. In 1974, he posted a record of 13-15, with a 149 ERA+ (2.41 ERA). Amazingly enough, more than half of Matlack’s wins were shutouts (7), and he had more complete games (14) than wins. He was third in ERA in the NL and 4th in strikeouts. Presumably, the fact that he posted a losing record was the reason why he received no votes for the Cy Young Award that year.

Paul
14 years ago

Don’t feel too bad for Sheets … he parlayed that season into a $42+ million contract and hasn’t pitched a full season since.  Of course, Brewers fans are all expecting him to put up a monster 2010 … for Oakland.

Paul
14 years ago

Yeah Sheets (12-14, 2.70) deserved better, probably would’ve done better this year as wins are becoming less important.

But, remember 2004 was the year of Clemens 18-4 with a 2.98 era and RJ 16 wins at era of 2.60, then you had a 20 game winner (Oswalt) and Zambrano at 16-8 and 2.75 era & 2 dominant reliever years from Gagne and Lidge.

At least he got a CY vote, as opposed to Josh Johnson this year who had a great year, but gets no recognition in the awards.

CY should be at least 5 places for votes, better 10, that way a voter can recognise and vote for the great Vasquez/Haren/Johnson seasons without affecting the outcome (like what nearly happened this year when there was a near top-3 consensus)