Rank this reliever

(I’ve been at this one a while, and it’s just too long — so here’s Part 1. Click here for Part 2.)

One name has been removed from these tables of relief WAR leaders. Most of you will twig it quickly, but let me be coy for now. Where would you rank this guy among the best relievers of all time? (All WAR figures are from Baseball-Reference, unless specified.)

 

Relievers – Career WAR leaders

(Min. 80% of games in relief to qualify;*
Hall of Famers in bold italics)

Rk Player WAR IP WAR/
100 IP
1 Mariano Rivera 52.7 1219.2 4.2
2 Hoyt Wilhelm 47.4 2254.1 2.1
3 Rich Gossage 39.9 1809.1 2.2
4 ? ? ? ? ? ? ? ? 29.6 1242.0 2.4
5 Lee Smith 27.9 1289.1 2.2
6 Trevor Hoffman 27.0 1089.1 2.5
7 Billy Wagner 26.9 903.0 3.0
8 Lindy McDaniel 26.2 2139.1 1.2
9 Stu Miller 25.5 1693.1 1.5
10 Kent Tekulve 24.8 1436.2 1.7
11 Dan Quisenberry 23.9 1043.1 2.3
12 Bruce Sutter 23.6 1042.0 2.3
13 Rollie Fingers 23.3 1701.1 1.4

* See the wisdom of the “80% of games” standard: All on this list had at least 65% of their innings in relief, and all but Miller had at least 78%. Even a slightly lower threshold would admit some pitchers to this leaderboard whose innings as starters comprised more than half their career total.

 

Mystery man doesn’t just benefit from bulk innings. His WAR/100 IP also ranks 4th among this group (above all four Hall of Famers), and 16th among the top 200 relievers in total WAR. Only Mariano can match or beat him on both counts.

 

Relievers – Best 3-year WAR

(90% of games in relief during the span;*
for each player, only the top value among overlapping spans is listed)

Rk Player WAR Years WAR/
100 IP
1 Dick Radatz 17.0 1962-64 4.1
2 ? ? ? ? ? ? ? ? 14.8 ????-?? 4.3
3 Rich Gossage* 14.0* 1975/77* 5.1
4 Dan Quisenberry 12.6 1983-85 3.2
5 Bruce Sutter 12.5 1977-79 4.1
6 Ellis Kinder 12.4 1951-53 3.7
7 Mariano Rivera 11.8 2004-06 5.1
8 Mariano Rivera 11.1 1996-98 4.6

* I raised the line from 80% to 90% for this list and the next, to keep the focus on relief work. The 80% line lets in those like Derek Lowe 2000-02, John Smoltz 2003-05 and others who actually had most of their innings as starters and got significant WAR from that role. I made an exception for Gossage, a starter in ’76 sandwiched between two fantastic relief years that by themselves totaled 14.0 WAR. His stats shown are for those years only. There is no other pitcher similarly situated.

 

Relievers – Best 6-year WAR

(same rules as above)

Rk Player WAR IP ERA+ W L SV Years WAR/
100 IP
1 Rich Gossage 24.9 790.1 150 50 50 131 1975-80 3.2
2 ? ? ? ? ? ? ? ? 23.3 683.1 161 58 48 100 ????-?? 3.4
3 Mariano Rivera 21.4 440.1 236 28 23 243 2004-09 4.9
4 Mariano Rivera 20.3 466.0 217 32 20 215 1996-01 4.4
5 Dan Quisenberry 20.1 724.2 165 41 33 212 1980-85 2.8

 

In these short-span tables, too, it’s more than just innings that lift our mystery man. His WAR/100 IP ranks 4th of 8, and 3rd of 5.

(By the way, his 4-year total of 18.7 WAR is the best ever by a reliever, but I didn’t want to stack the deck too much.)

Out of curiosity, I made a primitive stab at defining “very good” relief seasons. Since 1961, if we rank by WAR all relief seasons of 60+ innings, the 90th percentile falls at 2.6 WAR. So … The most relief seasons worth at least 2.6 WAR belong to:

  1. Rivera, 12 seasons
  2. Wilhelm, 8 seasons
  3. Gossage and ?????, 6 seasons

His career ERA+ of 134 in 1,242 innings is 23rd among all modern pitchers with 1,000 IP. Among relievers, only Wilhelm and Franco topped him on both counts. (And Franco took 20 years of regular work to get there; our man got his in 13.)

_______________

So, where would you rank him? Top 10? Top 5?

He retired holding the season saves record. Yet he never pitched in an All-Star game, and got 2.6% of the Hall of Fame vote in his one year on the ballot.

This year is the 40th anniversary of John Hiller‘s record-setting season. He’ll turn 70 in April.

_______________

Leaving the Hall out of this

This essay is an appreciation, not a Hall of Fame argument. I do rate Hiller above 2 of the 4 current HOF relievers, but I see that more as a knock on those elections than as a “me-too” case for Hiller.

I’ll say just this about the Hall and let it go: I would be happy if it included only Wilhelm, Gossage and (eventually) Mariano. But I think Hoffman’s getting in, and maybe Lee Smith (he’s been around 50% of the vote lately), perhaps even Billy Wagner. I think Hiller has a better case than all but Rivera, Wilhelm and the Goose, and the Hall of Stats agrees. But I’d be happy just to see his work remembered a little better than it seems to be.

_______________

(For the record, Fangraphs takes a dimmer view of Hiller. I don’t get it. Given Hiller’s raw data for 1973-78 combined — a huge K rate, a good HR rate, a suspect defense, and an actual 2.45 ERA in a park that favored hitters — I don’t see how they peg his FIP at 3.02 for that span.)

_______________

Beyond the Saves record

If you know just one thing about John Hiller on the diamond, it’s his brilliant 1973 season — 38 Saves, a record that stood for 10 years; 4th place in the Cy Young Award and MVP votes. But there are better measures of that magical year:

  • 1.44 ERA, in 125 innings — Best ERA since 1918 by an ALer with 100+ IP.
  • One unearned run. RA/9 ranks 5th all-time with 100+ IP. In the live-ball era, only Gibson ’68 was better.
  • 283 ERA+ ranks 5th all-time for 100+ IP. Since 1881, only Pedro has topped Hiller in both IP and ERA+. (That’s 1881.)
  • Stranded 71 of 84 inherited runners (85%) — Best strand rate ever with 25+ saves and 50+ IR (out of 77 such years). Since Hiller, only Dan Quisenberry had 30+ saves and 80+ IR; no 30-save man since ’89 had even 50 inherited runners.
  • Win Probability Added ranks #2 all-time by a reliever, edged by Willie Hernandez in his ’84 CYA/MVP season. No RP has reached 6.6 WPA since Hernandez. (The “perfect” years by Lidge and Valverde notched 5.3 and 4.1 WPA, and K-Rod’s 62-save record tallied 3.5 WPA. Modern saves just don’t tingle the WPA meter.)
  • 7.9 WAR ranks #2 all-time by a reliever, edged by Gossage in ’75. Just 5 other reliever years reached 6 WAR; none since ’86 have reached 5 WAR.

That would be Hiller’s lone year with more than 15 saves, but not his only great season or notable feat:

  • His 15.4 WAR over the next 5 years has been topped by just 8 relievers in any 5-year span (<40 starts).
  • 17 wins in ’74 — #2 all-time in relief (and tied but once).
  • His 1975 K rate — 11.1 SO/9 in 71 innings — was through 1983 the 2nd-highest ever for 50+ IP, and 124% above the league average. (By comparison, Craig Kimbrel’s insane 2012 rate of 16.7 SO/9, the highest ever for 50+ IP, was 117% above the league average.)
  • In 1967, Hiller threw shutouts in his first 2 career starts, and fell 3 outs shy of a 3rd straight. He finished with 6 shutouts in 43 career starts, the highest career rate of any post-war pitcher with at least 5 shutouts. (Just one of his shutouts came in 1968.)

One of my favorite Hiller feats came in a blown save on July 11, 1976. He entered in the 7th with a man aboard, but let in the tying runs. So he pitched on, through the 8th, 9th, 10th, 11th … In the 12th, a walk to his 20th batter filled the bags, with no outs and righty Hal McRae coming up, batting .347. Hiller stayed in the game — and escaped on a couple of grounders. Detroit scored in their half to win. He earned 0.333 WPA, twice that year’s median for converted saves. The 5.1-IP stint didn’t seem to hurt him; from there through year’s end, Hiller tossed 61 innings with a 1.48 ERA.

Hiller had a dozen relief games worth at least 0.500 WPA, 3rd most in the expansion era. No one in the last 20 years has more than 4 such games; there were 6 in the majors last year, by 6 different pitchers. No one’s had 2 in a season since 2004.

In 1973 and again in ’74, Hiller logged 5 straight relief games worth at least 0.150 WPA — the longest such streaks in those years, matched once each by 2 others. It’s been equaled once in the last 6 years. In the 10 games, Hiller had 9 saves and a win, allowing 1 run on 6 hits in 21 IP, stranding all 12 inherited runners. In all 9 saves, he entered with the tying run on base or at bat; he came in before the 9th in all but one.

_______________

*END OF PART 1*

Thanks for making it this far! When we return — if you’re still with us — we’ll trace the secret source of Hiller’s hefty WAR during his prime years, and sketch the differences between Firemen and Closers.

0 0 votes
Article Rating
Subscribe
Notify of
guest

30 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Hartvig
Hartvig
11 years ago

One of the last of a kind- not just a swingman but a swingman who was reasonably successful as a starter and used in high-leverage relief situations in multiple seasons. Bob Stanley- who’s career actually overlapped with Hiller’s by 3 years- was the last that I can remember to do this in multiple seasons. Wakefield really only had one season in which he started more than a game or 2 and at least shared the closer position. The rest of the years he split time between starting and the pen he was used as long-relief or in mop-up roles.Ramiro Mendoza… Read more »

Doug
Editor
11 years ago
Reply to  John Autin

Interesting read, John. Looking ahead to when fireman turned into closers, I was struck by how few pitchers there have been with 10+ starts and 10+ games finished in a season with 60% relief appearances. Since 1961, Gordon leads the way with 4 seasons, Tim Wakefield and Bill Hands have 3, and 27 pitchers have down it twice, most recently Seth McClung for Tampa and Milwaukee in 2006 and 2008. On a yearly basis, the numbers are all over the map, though obviously trending down. The most have been 13 pitchers in 1961 and 1969. The next five years after… Read more »

birtelcom
Editor
11 years ago

Most b-ref WAR by a reliever (80% or more games pitched in relief) over the following periods:
1950-1959: Hoyt Wilhelm 22.5
1960-1969: Hoyt Wilhelm 23.0
1970-1979: John Hiller 26.9
1980-1989: Dan Quisenberry 24.0
1990-1999: John Wetteland 17.2
2000-2009: Mariano Rivera 32.2

The list is the same if you use 90% or more games pitched in relief as the definition of a reliever.

Ed
Ed
11 years ago

John – Any idea why Hiller declined from 38 saves in ’73 to only 13 in ’74? Was he being used differently that year?

I also assume that his 31 decisions in ’73 have to be a record for a reliever both in terms of absolute number of decisions as well as decisions as a % of appearances (31 decisions in 59 appearances).

bstar
11 years ago
Reply to  John Autin

This is one of the reasons I question whether WAR is really the best gauge of reliever pitching ability. It seems a lot of reliever WAR is based on leverage and innings pitched. To a large degree, the player has no control over whether or not he pitches in high-leverage situations. It seems to me to largely depend on luck, or circumstance that particular year. One year a reliever may have a ton of high leverage situations, the next he may have fewer opportunities overall as well as far fewer very high leverage ones. This also leads to innings pitched… Read more »

Ed
Ed
11 years ago
Reply to  bstar

Bstar – What???? You mean St. Rivera may not be worth as much as people think he is??? How can that possibly be??? 🙂

bstar
11 years ago
Reply to  Ed

I didn’t mean to imply Rivera from that comment at all, Ed. Rivera’s been the epitome of excellence for a long time.

Ed
Ed
11 years ago
Reply to  Ed

No doubt Bstar. On the other hand, Joe Pos had a post out today looking at Rivera’s 9th inning, one run saves. Looking just at those situations, Pos estimated that Rivera only added 6 wins to the Yankees totals over 16 years. Granted those situations only account for a portion of Rivera’s games but they’re also the highest leverage situations, the ones where he had the potential to make the most impact.

http://joeposnanski.blogspot.com/2013/03/you-made-me-do-it.html?utm_source=feedburner&utm_medium=email&utm_campaign=Feed%3A+blogspot%2FaBacB+%28Joe+Blogs%29

bstar
11 years ago
Reply to  Ed

Ed, I was referring more to the differences in Mariano’s seasons over the years. Take 1996. He pitched the most innings of his career that year, and thus has the highest seasonal WAR total of his career. To me, that just suggests more opportunity, not necessarily better pitching. Mo WAS great that year, with a 2.09 RA9, but he’s had several seasons with a mark lower than that.

Ed
Ed
11 years ago
Reply to  Ed

Okay. But I want to make a (perhaps) different point. If you look at IP/WAR, Rivera’s gained one WAR per 23.1 innings. It took Greg Maddux 50.4 innings to gain one WAR. I’m sorry but I don’t buy that. I know it’s because Sean uses Leverage Index for relievers (and Tom Tango agrees with that). But I have to say there’s 0 chance that an inning pitched by Rivera was worth more than an inning pitched by Maddux. That doesn’t even begin to pass the laugh test.

birtelcom
Editor
11 years ago
Reply to  Ed

Ed@21: It seems to me very clear that the average inning pitched by Greg Maddux was not as valuable as the average inning pitched by Mariano. That’s the nature of leverage. You give up a couple of runs in a close game in the third inning, your team has time to come back; you give up a couple of runs in a close game in the ninth inning your team is dead. A high proportion of Mariano’s innings were close games in the ninth inning, ergo….

Ed
Ed
11 years ago
Reply to  Ed

Birtelcom @22 Sorry but I disagree 100% with that. An inning is an inning and a run is a run. If the creators of WAR really believed in leverage, they’d include it in batting WAR as well. But they don’t. BTW, let’s take a bit closer look at Rivera. He gave up 324 runs in 1219.2 innings and has a WAR if 52.7. That implies roughly 527 runs saved relative to a replacement level pitcher. So that puts a replacement level relievers ERA at roughly 6.28. Um okay. I’m pretty sure if I were the GM of a team I’d… Read more »

bstar
11 years ago
Reply to  Ed

Ed, you do make a really good point. In batting, a run is a run is a run. For starters, the same is pretty much true (because the leverage is so low). Not so for relievers. They get a leverage boost. If they do, why shouldn’t pinch hitters also get a leverage boost? Don’t they often hit in the most crucial of all situations? But what would position player WAR look like if we included leverage? It would basically be RE24, or WPA. I think it’s better to have those as separate numbers. I definitely see your point, but since… Read more »

Ed
Ed
11 years ago
Reply to  Ed

Bstar and JA – Thanks for your responses. It’s certainly possible that I’m wrong about this issue. I’ll need to do some more research and give it some more thought. I do want to raise a related point. Only relievers receive a leverage bonus, not starters. So if a starter pitches a complete game, recording the last few outs of a close game by themselves, that’s treated differently than if a reliever is brought in to record those outs. If leverage is as important as you two (and Birtelcom) assert it is, then why should it matter whether it’s the… Read more »

bstar
11 years ago
Reply to  John Autin

Agreed. And that’s exactly what I meant. Hiller basically absorbed all the leverage from innings 7-9 whenever he pitched, or something along those lines.

Personally, I think it’s difficult to compare ’70s firemen with modern elite closers. The ’70s guys will have more WAR/year but the modern guys will have better ERA+ numbers.

Ed
Ed
11 years ago

Re: Hiller’s FIP. Remember that FIP only considers HRs, Ks, and BBs. The fact that the Tigers had a “suspect defense” is completely irrelevant in the calculation of FIP. (I think HRs are somehow adjusted for ballpark but I’m not 100% sure of this). I suspect that Hiller’s FIP is hurt by his intentional walks. His 13.2% intentional walk rate (71 intentional walks, 535 total walks) is about the 25th highest percentage of all time among pitchers with 500+ walks. As far as I know, FIP treats all walks the same. That being said, it’s hardly like FIP hates Hiller.… Read more »

Ed
Ed
11 years ago
Reply to  John Autin

I don’t really have any answers for you John. Everything you pointed out is why some people (Bstar for example) hate Fangraphs WAR for pitchers (which is based on FIP). Tom Tango, btw, has said there’s merit to both Baseball Reference’s and Fangraph’s pitcher WAR and it’s best to just average them.

Anyway, Jim Crawford had a lower FIP than Hiller in ’76 despite an actual ERA nearly twice as high. That seems like a conclusion that would be hard to justify no matter what you believe (Crawford also pitched for the Tigers).

mosc
mosc
11 years ago
Reply to  Ed

I run into averaging of averages in my line of work and the statistician in my always complains. Taking multiple measurements with known error and averaging them does not improve the accuracy necessarily. More accurate measurement should receive more weight. Also, if you’re using multiple averages based on shared input data (very much so in this case), you can obscure your precision by essentially double counting some of your data. I think the gap in the WAR calculation shows us that we still do not correctly understand the subject. Looking at the variation between the two numbers would be a… Read more »

bstar
11 years ago
Reply to  Ed

Ed, in fact, I’d written a long post disparaging reliever FIP marks right under your #4 comment this morning, but deleted it on my way to work because I felt I was being too harsh. But since you broached the subject… To be clear, Tango has said that rWAR is probably better for CAREER evaluation, and only that FIP-WAR may be better for seasonal evaluation. Obviously, if someone is out- or under-performing their FIP year after year after year, it seems virtually everyone agrees that FIP is not worth much for that pitcher. And that definitely seems to be the… Read more »

Artie Z
Artie Z
11 years ago
Reply to  Ed

Ed and John – the problem is that Hiller gave up 6.9 H/9 innings, while Crawford gave up 9.5 H/9 innings. FIP doesn’t care about that at all. Neither of those is all that different from the pitcher’s career mark – Hiller at 7.5 and Crawford at 9.95. For reference, 6.9 H/9 is about Mariano Rivera’s career H/9. 9.5 H/9 is about Kevin Millwood’s. That’s a huge difference in what type of pitcher we are discussing. Heck, the career FIP for Hiller and Crawford isn’t all that different – 3.39 for Hiller, 3.49 for Crawford, despite Hiller leading 2.83 to… Read more »

bstar
11 years ago
Reply to  John Autin

One other point I would make about fWAR and rWAR for pitchers that people don’t seem to understand. They are BOTH defense-independent, not just fWAR!

bstar
11 years ago

JA, a minor quibble with your comparison of the K/9 numbers for Kimbrel and Hiller. Since they’re relievers, shouldn’t we use K% to include all plate appearances? We’re ignoring all the walks Hiller gave up otherwise.

1975 Hiller K% – 29.5, 129% above league average
2012 Kimbrel K% – 50.2, 150% above league average

But Kimbrel’s mark, despite striking out over 1 out of 2 batters, is not the all-time best relative to league. It’s this guy:

2003 Eric Gagne K% – 44.8, 162% above league average