10 worst reliever seasons ever

Last week, I wrote an article for the mighty THT looking at the best reliever seasons of all time. Well, every yin must have its yang, so looking up the best seasons ever just leads to the natural flip side question: what are the worst seasons? Let’s find out.

The method

I introduced the method last time, so you can click back there to get the full version. For now, let’s just give a brief summation.

It’s based on an old Pete Palmer idea, Adjusted Player Runs (APR). You take the number of earned runs a pitcher actually allowed in his innings pitched and compare that to what a league-average pitcher would do in the same number of innings. Then you adjust for park factor to figure out his APR. (I don’t necessarily have the exact same formula as Palmer, but it’s the same general gist.)

The goal is to reward pitchers with the right combination of quantity and quality. Or, for this column, to punish them for the combination of lacking of quality over the most quantity.

Also, in order to compare across eras better, I adjust APR by league. I take a player’s APR and divide it by the league’s ERA. I call this APG (Adjusted Player runs per Game; I guess it could be APRG, but APG just looks cleaner). The list below is in terms of APG.

One key thing to note is that we’re looking only at how a pitcher did in relief. I took the relief splits for every pitcher from 1916 to 2012, so if he had any starts, it isn’t factored in.

I should note one clear flaw here. This study doesn’t account for inherited runners. It only looks at stats you can find from Baseball-Reference’s Splits Finder in the Play Index, and inherited runners aren’t there.

Anyhow, in last week’s column, none of the best seasons came after 1986, and all but one came from 1973-86. This time, the seasons are more spread out, but there is a clear gap. Shockingly, none of the worst seasons come in the last 30 years.

Here are the worst of them all.

10. Jim Todd, 1979 A’s, -23,39 APR, -5.53 APG.

Not surprisingly, a lot of the guys on this list will be obscure pitchers.

1979 was the last and least season of Todd’s six-year career. Most teams wouldn’t give a pitcher like Todd as many opportunities as the A’s did, but beggars can’t be choosers, and with a 54-108 record, the A’s were beggars, indeed.

Todd’s 6.56 ERA isn’t historically dreadful, but he did it while working in one of the game’s great pitchers’ parks. In 81 innings, he allowed 108 hits. More remarkably, he walked twice as many batters as he struck out: 51 to 26. He even managed to serve up a dozen homers, giving him terrible peripherals all around while also letting a bunch of balls in play fall for hits. It’s amazing his ERA wasn’t worse that 6.56.

9. Al Smith, 1938 Phillies, -21.75 APR, -5.75 APG.

Smith actually had a respectable career. In 12 seasons, he posted a 99-101 record with an almost perfectly league-average ERA. While average isn’t very exciting, it’s not bad, either.

Smith was pretty good for the mid-1930s Giants and early 1940s Indians, but in between he laid an absolute stinker of a season with the 1938 Phillies. I’m sure their team defense didn’t help, as it was typically woeful (negative 6.7 wins fielding according to WAR), but then again, Smith was the worst guy on the staff among those with over 60 innings pitched.

In 36 relief appearances, he threw 77.2 frames, allowing 108 hits, and 35 walks, nearly two base runners per inning. That works out to a .394 on-base percentage by opposing batters.

It all works out to a 6.49 ERA in relief while pitching in a league with a 3.78 overall ERA. Six times, he allowed at least four runs in a game. His low point came on July 28, when he faced four batters, and they all scored.

He actually improved dramatically in the second half. As late as July 21, his ERA was north of 9.00.

8. Dick Welteroth, 1949 Senators, -25.99 APR, -6.19 APG.

This is as good a place to ask as any: why are there so many long-ago pitchers on the list and no recent guys? There are two factors at play.

First, with larger bullpens, teams have fewer innings per reliever, so quantity keeps the more recent guys off. Second, back in the day, some teams didn’t really put much priority on the bullpen. Instead of using it as a weapon, it was often just filler. So you could heap plenty of innings on a really crummy arm and not think too much about it.

As for Welteroth, he was a crummy pitcher on a crummy team. Like the 1938 Phillies and 1979 A’s, the 1949 Senators lost over 100 games. And boy, did Welteroth play his part. In 50 relief appearances, he posted a 7.04 ERA. As bad as that sounds, he was even worse as a starter, with a 10.64 ERA in two failed outings.

What sets Welteroth apart is his control, or total lack thereof. In 84.1 relief innings, he walked 80 batters. That’s … interesting. To this day, only one person has more walks in relief in a season. (Be prepared not to be shocked: the record holder is Mitch Williams, with 91.)

Welteroth’s 8.54 BB/9 is the worst by any reliever with more than 70 innings pitched, just beating out the 2009 version of Carlos Marmol. (If you drop it down below 70 innings, Mark Clear tops it with his 1984 campaign.)

7. Fred Heimach, 1933 Dodgers, -21.39 APR, -6.41 APG

1933 was the final season of a 13-year career for Heimach, and it couldn’t end fast enough. Most guys make it on this list due to a combination of plenty of innings and poor quality. Not Heimach, though. He makes it on this list despite through just 29.2 innings all year, and that includes three starts.

In relief, he threw just 15.1 innings in 1933. It’s almost impossible to make this list with such a small number of innings, but Heimach figured out the way. He allowed 26 runs in those innings.

Yeah, that’s incredibly bad, and it’s actually even worse that it seems. The 1933 NL was the most pitcher-friendly league between 1920 and 1940, with a 3.34 average ERA. Plus, Ebbets Field played as a pitchers’ park.

Heimach faced 86 batters in relief, surrendering 32 hits and eight walks and one hit batsman. Boy, those three guys who laid down sacrifice bunts probably feel foolish now.

He didn’t allow any runs in either of his first two outings, which totaled three innings. Then he allowed multiple runs in all of his remaining games. On June 24, he surrendered eight runs in an inning and a third. After that, the Dodgers shelved him until July 17. Somehow, he was even worse in that outing, giving up nine runs—all earned—while recording just one out.

6. Ed Connolly, 1967 Indians, -21.12 APR, -6.54 APG

Connolly is a milder version of Heimach. He had 11 relief outings—four more than our man Heimach—and threw 26.2 innings, more than 10 over Heimach’s total. But like Heimach, Connolly was so incredibly bad in those few games that he works his way onto this list anyway.

In those 26.2 innings, Connolly walked 18 batters (though with 21 strikeouts) and allowed 44 hits (a dozen for extra bases), leading to a 10.46 ERA. It could be worse; if you include unearned runs, his runs-per-nine innings is 11.48. Folks, this came in the 1960s, when pitchers ruled. The league ERA was 3.23, even lower than in Heimach’s big, bad year.

Two August outings really put him on this list. On Aug. 13, he allowed nine runs in two innings of work. Ten days later, he gave up seven runs while retiring four batters. Connolly shows us why no one past 1980 is on this list. Who now, in a higher-scoring era, gets left out there on the mound for two massive poundings like that in a year, let alone in a fortnight? Obviously this wasn’t common then, either, but it’s unimaginable now in the era of multiple pitchers per inning.

5. Larry Benton, 1935 Braves, -26.35 APR, -6.56 APG.

Ah, the 1935 Braves, one of the worst teams of all time, dropping an amazing 115 games. Benton did his part to put that team in the toilet.

Like many here, Benton was at the end of a decent career. In fact, Benton was a star for a bit. He led the NL in winning percentage in 1927 and 1928. In fact, in 1928, he also led in wins (25) and complete games (28).

But in 1935? Yeah, he just sucked. He appeared in 29 games, all in relief. In those outings, he posted a 6.88 ERA in 72 innings by allowing 103 hits and walking 24 batters. He had the highest ERA on the worst staff in baseball. He allowed runs to score in three-fourths of his outings.

Benton shows another reason why older pitchers dominate this list. Can you imagine a manager nowadays letting his worst reliever average nearly three innings per outing? (Hell, can you imagine a manager allowing his best reliever to throw that much per outing?) Well, longer outings were more common then, so a terrible pitcher could do a lot more damage in 29 appearances.

4. Dave Hamilton, 1980 A’s, -27.34 APR, -6.77 APG.

Man, those 1979-80 Oakland bullpens were so incredibly bad they contain the two most recent guys on this list, and those are the two worst reliever seasons since 1967.

Hamilton was Todd’s teammate in 1979 and did well for himself that year, but my, how thinks had changed by the next season. Hamilton belongs in the Heimach-Connolly Axis of Ineptitude.

In just 20 relief stints, he posted a comically pathetic 12.79 ERA. In 25.1 innings, he walked 26 batters—yes, more than one an inning—while allowing 39 hits. Oh, and five of those hits were homers. As a general rule of thumb, relievers don’t want to give up a dinger every fifth inning.

Batters posted an OPS of 1001 against Hamilton. That’s, uh, neat. Just imagine what the numbers would be if the A’s weren’t working in a great pitchers’ park. Oakland’s starters completed 94 games that year, the most by any team since WWII.

While I don’t mean to defend manager Billy Martin‘s pitcher abuse, it’s worth noting that one reason he did that was because Oakland’s bullpen was so horrible. The overall ERA of AL relievers in 1980 was 3.80, but the A’s relievers combined for a 5.22 ERA. Yuck.

Oh, and like seemingly everyone here, Hamilton never pitched again after his terrible campaign.

3. Nick Strincevich, 1940 Braves, -26.08 APR, -6.77 APG

If you take APG to the thousandths place, Strincevich scores a bit worse than Hamilton: -6.774 versus –6.767.

Strincevich is different in that his horrible season came at the beginning of his career instead of the end. It wasn’t a great career, but he did last eight seasons.

On the face of it, his numbers don’t look too bad. Go to Baseball-Reference, and you’ll see him posting a 5.53 ERA. Okay, that’s bad, but what’s he doing here? Simple, those numbers include 14 starts. This study looks solely at his relief splits. Strincevich was actually a decent starter, posting a 3.77 ERA in 86 innings.

He ended up with that gaudy overall ERA because he was that bad out of the bullpen. In 42.2 innings, he allowed 43 runs for a 9.07 ERA. In 18 relief appearances, he allowed teams to score 15 times. His scoreless outings totaled a combined four innings.

What really put him on this list was his final appearance. Asked to throw mop-up innings in a meaningless late-season contest, Strincevich allowed nine runs in six innings for Boston in a 14-0 defeat. Again, this is why older pitchers dominate. Even in garbage time, a team would never let a guy get bombed for six innings and nine runs nowadays.

2. Art Decatur, 1927 Phillies, -27.29 APR, -6.98 APG

This Phillies are the most inept franchise in history, so I guess it makes sense that the worst season by any NL reliever came from them. In 1927, Decatur was a veteran in his sixth year who would never see season No. 7. Part of the Phillies 103-loss campaign that year, Decatur threw 82.2 innings in 26 relief outings.

Wait, let’s stop there for a second. He threw 82.2 innings in 26 relief outings! I know I’m beating this point to death, but there is a reason why no recent pitchers are on the list.

For comparison, in the last 30 years, the smallest number of bullpen outings by any pitcher with at least 82 innings pitched in relief was 32 by Alan Mills in 1992. But one of the worst relievers in all of baseball averaged more innings per stint in 1927.

Anyhow, Decatur sucked, giving up 67 runs over those innings for a 7.29 ERA. Rather surprisingly for the era (and the quality of defense he pitched in front of), none of the runs was unearned.

1. Connie Grob, 1956 Senators, -30.03 APR. –7.22 APG.

There can be only one. And that one is Connie Grob. Unlike the others on this list, Grob was a one-year wonder. His first, last, and only season was in 1956, but that’s all he needed to make an unwanted sort of history.

Grob strode out of the bullpen 36 times for the 1956 Senators and threw 72.1 innings. In that time, he allowed 66 earned runs off 116 hits and 24 walks. The best part: every eighth batter he faced smacked an extra-base hit against him (12 homers, eight triples, and 24 doubles in 354 batters faced out of the bullpen).

As bad as his –30.03 APR looks, that might actually underrate his awfulness. For this rating system, I choose to use earned runs allowed instead of all runs allowed, and Grob actually allowed more than his share of unearned runs: 10. If you were do this by all runs allowed, Grob’s 76 runs in 72.1 frames would give him an APR of –35.30 and a APG of –7.51.

Again, the more you look at Grob’s season, the more it’s clear why no one from the last 30 years shows up on this list. Who lets their worst pitcher average two innings per bullpen outing these days? Groh once allowed 10 runs in a six-inning relief outing that year. That ain’t the way bullpens are handled in 2013. But it’s how Grob was handled in 1956, and it’s why he ends up with the worst reliever season ever.

Of course, having now done two of these columns looking at the best/worst relievers ever and finding no top finishes from the last 25 years, the question becomes, what are the most extreme reliever seasons of recent times? I’ll get to that in a few weeks when I have a chance.

References & Resources
Info comes from Baseball-Reference.com, primarily its Splits Finder. All stats here are just relief stats unless specifically noted otherwise.

Print Friendly
 Share on Facebook4Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Fantasy Waiver Wire: Week 25, Vol. I
Next: State of the NL Central: three teams in October »

Comments

  1. Phil said...

    I think that baseball was managed foolishly in the past.  And this article on relief pitchers is a great example.  Why was the bullpen such a low priority?  Why were bad relievers given so many innings? 

    Baseball fans in the 1950’s might have thought, “7 relievers?  How do they get enough work?”  I feel the same way about bench players from the past, “15 hitters?  How did they get enough work?”

  2. Bob M said...

    This article brings to mind the way that NFL teams, even into the ‘60’s, picked their placekickers. They were often position players who really had questionable talent for placekicking. Look up some of the unbelievably poor rates of success in a position that so often decides the outcome of a game, and you have to wonder what took front offices and coaches so long to figure out that reserving a roster spot for a true kicker would make a lot of sense.

  3. Marc Schneider said...

    In fairness to the NFL, I believe that during the 50s/60s, teams could only carry 40 players so there was not much room for a specialist who could do nothing else.  And some of those position players were actually pretty good kickers, such as Lou Groza.  But, yes, it took a long time for the NFL to place much emphasis on the kicking game. 

    As for bullpens, it was expected that the starter would complete most of his games.  Bullpens were generally used when the team was losing badly.  To the extent that teams needed a relief pitcher in a close game, I think they typically had a guy who could pitch several innings at a time-so, they might have one good relief pitcher and the rest was just filler to be used in blowouts.

  4. Steve Treder said...

    “As for bullpens, it was expected that the starter would complete most of his games.  Bullpens were generally used when the team was losing badly.  To the extent that teams needed a relief pitcher in a close game, I think they typically had a guy who could pitch several innings at a time-so, they might have one good relief pitcher and the rest was just filler to be used in blowouts.”

    That’s exactly right, and not only that, in high-leverage relief situations, it was common that one of the team’s top starters would be used.  The last couple of pitchers on the staff were designated only for blowouts and emergencies.

  5. littlelucas said...

    Great article. Call me an old time traditionalist but I am not a fan of how bullpens are used today. I know the game is evolving into more of a specialist role but does a team really have to carry as many pitchers as position players? It is no wonder there are no pitchers on your list from the recent era. The games take forever. I saw a game the other day when a team used SIX pitchers in a very low scoring game (something like 2-1). This makes for a lot of dead time for the fans. Bring in the lefty to get one out and then bring in the righty to get one out, etc.

    Another HUGE difference is 5 man rotations vs traditional 4 man rotations in the pre 1980’s. Even with this there is a huge priority placed on pitch counts. Starters only pitch every 5th day tops and there is still this pitch count mindset. I was watching the Tigers-Mariners game last night and pitch count is a constant stat that is displayed in the top corner of the screen like balls, strikes and outs. The pitcher stat CG (complete game) is almost extinct. It used to be CG’s were a badge of honor for a pitcher. Nowadays it means nothing. Last night Mariners pitcher Iwakuma had 4 hit shutout with 8-0 lead and they took him out after 8 innings!!! And he was pitching a shutout!!! You try taking the ball away from Gibson, Seaver, Koufax, Morris and others when they had a shutout going into the 9th!!!!! I would hate to be that manager.

    The biggest change in baseball over the last 50 years is the use of bullpens. Which has also changed the way starters are used.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Current ye@r *