By In Stuff

What the Eck?

When I started writing that last sprawling piece about John Hiller and the modern reliever, I fully expected Dennis Eckersley to play a major role in it. The story has long been told that it was Tony La Russa’s managing — and an aging Eckersley’s remarkable ability to pitch one clean inning at the end of games — that sparked the whole revolution of one-inning closers.

In fact, I’m pretty sure that’s one of the reasons Eckersley sailed into the Hall of Fame first ballot despite having a borderline case. I think the idea was that Eck, in addition to being a fine pitcher, was a transformational figure in the game.

But after looking at it a bit more closely … I don’t think that’s exactly right.

I’m going to throw a shocker at you : I think if any pitcher represents the shift from old-fashioned to modern closer it is the overwhelming presence of Lee Arthur Smith. And maybe the Lee Smith-for-the-Hall of folks should have played up that part of his career.

First: Eckersley. So you probably know the basics of Eck. He came up when he was just 20 years old, and he was a good rookie starting pitcher for the Cleveland Indians. He would have won the Rookie of the Year award in many years, but he had the misfortune of coming up in the same year as Fred Lynn (who won Rookie of the Year AND MVP) and Jim Rice. So nobody really noticed Eckersley’s 13-7, 2.60 ERA year.

Eck pitched well in Cleveland but he did not really get noticed at all until he won 20 with Boston in 1978 and finished fourth in the Cy Young voting. He was a good starter until he turned 30, comparable to a Rick Wise or Larry Dierker or someone like that, and then at 31 the roof caved in. The Cubs dumped him on Oakland for three minor leaguers right at the start of the 1987 season. La Russa at first used Eck as a catch-all, starting him twice, using him in long relief, letting him finish some games. Toward the end of the season, La Russa decided to use him a lot more at the end of games. Eck proved to be pretty good at it.

In 1988, Eckersley led the American League in saves with 45. The general thought is that this is when the one-inning closer was born — but it isn’t exactly true. Only 21 of Eck’s 45 saves that year are what we could call a modern save (start the ninth with a lead of three runs or less). Those 21 modern saves did lead baseball, but only barely. He was just one ahead of John Franco and Jeff Reardon. Point is: Eckersley was not being used in a fundamentally different way from other top relievers.

In fact, as mentioned in the original piece, it was probably Franco who began the revolution. In 1987, Pete Rose managed the reds and 25 of Franco’s 32 saves were modern. That was the most one-inning saves in baseball history up to that point.

In 1989, Eckersley saved 33 games, but only 17 of them were one-inning saves. Dave Smith (22) and Franco again (21) had more modern saves.

Then, 1990 was sort of the breakthrough for the modern closer. Eckersley did have 26 modern saves But that was the year Bobby Thigpen had 41 one-inning saves — beating the old record by 16.

Then comes 1991, and Lee Smith enters the picture. Up to then, Lee Smith was a pretty typical fireman, not a closer. He was a guy who would pitch multiple innings, a guy who would get you out of jams, a guy a manager went to whenever the game seemed to be on the line. Only 87 of Smith’s 265 saves up to that point were the one-nning kind. You can see how that compares with some of the great fireman of the age.

Percentage of modern saves:

— Sparky Lyle, 18.9%

— Dan Quisenberry, 21.3%

— Goose Gossage, 22.6%

— Rollie Fingers, 23.8%

— Bruce Sutter, 27.3%

And Lee Smith was at 32.8%. It’s a little higher percentage than the rest but it is in the ballpark.

But in 1991, while playing for St. Louis and manager Joe Torre — yep, here’s Joe Torre — it all changed. Smith set a career high of 47 saves. And 34 of those saves were modern.

It was a stark shift. Smith had never had more than 16 modern saves in a season. Now he had more than double, and he he led the league. Torre obviously figured that for various reasons (perhaps thinking Smith was a big guy who wasn’t great at holding on runners) Smith would be a force starting the ninth with no one on and pitching just the one inning.

In 1992, Dennis Eckersley had his pinnacle year. He had 51 saves. He won the Cy Young Award. He won the MVP. That is supposedly the year that Eckersley and La Russa cemented the idea of the modern closer forever.

And 31 of Eck’s 51 saves were one-inning saves, a career high.

But Lee Smith had more. He had 36 modern saves that year — 36 of 43.

And by 1993, well, by then baseball was completely enamored with the one-inning save. Mitch Williams (40 of 43), Randy Myers (40 of 53), Duane Ward (37 of 45), Rod Beck (35 of 49) and Smith (36 of 46) were all essentially one-inning closers. Then would come Jose Mesa and John Wetteland and others.

Of Eckersley’s 390 saves, 231 were one-inning saves. That’s 59%, which is probably lower than you would have predicted. Yes, perception can be reality. The perception of Eckersley as a one-inning closer — even if it wasn’t exactly so — might have been powerful enough to inspire other managers to start using their closers just to get the last three outs.

But, really, it was Lee Smith and Joe Torre who created the first modern-day closer. Before 1990, as mentioned, only 33% of Smith’s saves were modern.

After 1990: About 71% of Smith’s saves were modern ones.

* * *

By the way, here are your leaders for modern saves in a career:

1. Trevor Hoffman, 498 (83% of career total)

2. Mariano Rivera, 491 (75.3% of career total)

3. Billy Wagner, 369 (87.4% of career total)

4. Francisco Rodriguez, 359 (83.5% of career total)

5. Joe Nathan, 345 (92% of career total!)

6. Jonathan Papelbon, 315 (86% of career total)

7. Troy Percival, 300 (84% of career total)

8. John Franco, 291 (69% of career total)

9. Jose Mesa, 283 (88% of career total)

10. Huston Street, 282 (87% of career total)

Among active players, Craig Kimbrel is the ultra-essence of the modern closer. More than 93% of his career saves have been one inning saves. It’s kind of sad think of how often the awesomeness of Kimbrel has been wasted just getting three outs with a three-run lead.




Print Friendly, PDF & Email

14 Responses to What the Eck?

  1. Gordon C Hewetson says:

    At some point these numbers dull my senses. Who came first — Lee Smith and Joe Torre or Dennis Eckersley and Tony LaRussa. I much prefer the John Hiller narrative.

  2. Chris H says:

    I think Ron Davis deserves a mention here. In 1978, Gossage averaged a bit over 2 innings per appearance with the Yankees, and led the league with 27 saves in 134 innings pitched.

    In 1979, the Yankees and manager – guess who – Billy Martin realized they had this kid named Davis who could also throw hard, and also didn’t seem to be a starter. Davis appeared in 44 games that year, finished 21 of them, and earned 9 saves. He also had a 14-2 record – 14 wins in relief! Gossage meanwhile dropped to 36 appearances and 58 innings – a bit over an inning and a half per appearance. (That’s not many appearances; maybe he was injured part of the year.)

    In 1980, Davis appeared in 54 games, logged 131 innings, and earned 7 saves. Gossage appeared in 64 games, threw 99 innings – again, about 1.5 innings per appearance – and again led the league in saves with 33. And by this time, if memory serves, everyone was talking about the fearsome duo in the Yankees’ pen, with a “setup man” followed by a closer.

    (I was young then, so memory may in fact not serve; maybe I’d just never heard of setup men before. But the way the Yankees finished games with a one-two punch was definitely much discussed as a model bullpen.)

    One more year of that: 1981, Davis with 43 games, 73 innings, 6 saves; Gossage with 32 games, 46 innings, and 20 saves.

    And again, from memory: people talked about the way Davis was saving Gossage some of the workload, presumably making him more rested and more effective. And, it seemed Davis might be being undervalued, as *only* a setup man. The following spring, the Yankees shipped him off to Minnesota along with Greg Gagne in exchange for Roy Smalley – that seems like it might have been a terrible trade – and Davis became a closer in his own right, picking up 106 saves for the Twins in the next four years.

    I remember seeing Davis and Gossage pitch, 8th and 9th innings, at Cleveland. And I remember thinking Gossage threw noticeably harder. I can’t possibly have actually seen the difference. But he was by far the more intimidating presence on the mound.

    • invitro says:

      “And by this time, if memory serves, everyone was talking about the fearsome duo in the Yankees’ pen, with a “setup man” followed by a closer.” — The thing is, the stats from 1980 (or nearby years) disqualify Davis from being a setup man. Davis had only two holds in 1979, only two again in 1980, and only seven in 1981. If he entered the game in a save situation (21, 11, and 16 times for the three years), he either got the save, or a blown save.

      I’ve been interested in this. A couple of articles ago, someone said La Russa had invented the setup man. I don’t know a standard definition for a setup man based on stats. But we could start with just having a lot of holds.

      Rick Honeycutt had 18 holds in 1988, with 7 saves and 2 blown saves, for the A’s. That’s good enough for me to call him a setup man, for now, anyway. Greg Cadaret had 13 holds the same season for the A’s.

      But Dave Leiper had 15 holds in 1987 for the A’s, with 1 save and 3 blown saves. Does that make him a setup man (to Eckersley and Jay Howell)? Two guys had 7 holds for the A’s in 1986; I don’t think that’s enough to be called a setup man.

      What about the 1983 White Sox? Dennis Lamp had 15 SV, 5 BS, and 0 H. Salome Barojas: 12 SV, 4 BS, 10 H. Juan Agosto: 7 SV, 1 BS, 11 H.

      I’ll stop with the 1987 Reds. Franco had 32 SV, 9 BS, 0 H. Rob Murphy: 3 SV, 4 BS, 16 H. Frank Williams: 2 SV, 5 BS, 15 H.

      I’m very interested if anyone has an objective definition of “setup man”, or other ideas on who the first one was.

      • nightfly says:

        That was my comment, and it’s kind of you to remember it. I think I wrote something to the effect that because of how he used Eck, LaRussa wound up inadvertently inventing the idea of a bridge – you’re holding back the closer for the ninth, so you need to get to him from the starters. That may not be a single set-up man like Mo Rivera for Wetteland in 1996; it’s a job that would probably be spread around and be dependent on lefty-righty matchups. (Closers tend to be considered able to get anybody out for an inning.)

        Someone with a play index could probably find out which teams had the most hold opportunities and such… since the A’s won a ton of games in the late 80s they’re a likely candidate.

  3. Cliff Blau says:

    Tom Henke. In 1991, he pitched in 49 games. In 42 of those, he entered with the bases empty. He pitched exactly one inning in 41 games. 28 of this 32 saves were one inning.

  4. Sinister Six says:

    Just throwing it out there, Eckersley probably shouldn’t be in the HOF at all. The fact that he got in on the first ballot is nuts.

    • Patrick says:

      Eckersley is a very interesting case. He had some very good years as a starter. His six best seasons by WAR were 7.3, 7.3, 5.3, 5.1, 4.6, and 4.5.) That’s a solid peak. He had a career WAR of 45.7 as a starter. It’s hard to know how he’d have fared from age 32-whenever as a starter, but I don’t think 65 WAR/250+ wins is out of the question given how good Oakland was.. That’s not first-ballot HOF, but depending on how it all shook out there could be a decent argument

  5. steve says:

    I found this line interesting: (Joe Torre) “perhaps thinking Smith was a big guy who wasn’t great at holding on runners”. It made me wonder how many of the modern closer types are pitchers you wouldn’t particularly want to bring in to a close game with zero or one outs and runners on. Or, maybe, a guy who is a bit wild (Mitch Williams) is better off starting an inning so he has at least one batter for whom a pitch to the backstop is irrelevant. Not all closers are fireman. Papelbon, I imagine, would have learned to be a fireman and hold guys on if he had to. He didn’t have to, so he didn’t (Pap being Pap). Kershaw, though, the other night, showed he can be whatever he wants to be.

  6. John Autin says:

    Great stuff again, Joe.

    I’m not sure there’s one turning point for the modern closer. Sparky Anderson’s first 2 NL champs both had the MLB Saves leader getting the bulk of their saves in 1-IP stints — Wayne Granger 23 of 35 in 1970, Clay Carroll 22 of 37 in ’72. In 1969, Fred Gladding had 20 of his NL-high 29 in the 1-IP role.

    It’s safe to say that the use of relief aces blossomed in the mid-60s, and for 20-25 years there were competing models for how best to use them. Starting in the mid-70s, the best ones worked massive innings, especially in the AL with the new DH rule; from 1973-83, 18 guys logged 140 IP with 5 starts or fewer, and 15 were in the AL. The last 30-save guys with 140 IP were Willie Hernandez ’84 and Bob Stanley ’83; “Steamer” and Quis that same year were the last with 30+ saves and at least 2 IP per game for the year.

    There are probably several reasons why the short-save model won out. But it may have been inevitable that rising recognition of relief aces would steer us towards ONE STAT for ranking them. Take Hernandez, the ’84 CYA/MVP: He actually worked far more games and IP in NON-SAVE spots that year, and his 3.9 WPA in just those games would have ranked 3rd this year in total relief WPA — but all folks remember is, 32-for-33 in save tries. It’s kind of the same reason that RBI leaders have won so many MVP awards. And once established, that pattern is self-perpetuating.

    It will be very interesting to see what affect this postseason trend has (if any) on next year’s regular season.

  7. Ross says:

    Joe – Are these numbers for the modern closers only referring to saves that were exactly 1 inning? What about when you include saves that were less than 1 inning (pretty common when team is up 4, gives up a run, brings in the closer)? These percentages, while high, still seem lower than I’d expect for the more recent guys.

    • Ross says:

      I assume you need BBRef Play Index to look it up, so I can’t, but for example Papelbon since he became a full time reliever (his 2nd season) has appeared in 672 games and pitched only 691.2 innings (1.03 innings per appearance). Having watched him pitch over the years, I’d definitely guess that more than 86% of his saves were 1 inning or less.

      • invitro says:

        I counted them up by hand and got 90%. I counted 30 1.1-inning saves, 6 1.2-inning saves, and one 2-inning save. Then, (368-30-6-1)/368 = 90%. Maybe I missed a bunch? I counted only seven saves of more than one inning since the start of 2011.

  8. Nathan says:

    For several years I have been arguing Lee Smith ushered in what we consider to be the modern day closer. It wasn’t necessarily for all of the reasons you stated, but I think you’ve proven the point better than I would have if given the opportunity.

    For whatever reason, Lee Smith often gets lost in the shuffle in these conversations and is often dismissed in comparison to Rivera, Hoffman and Eckersley. Smith should get credit for helping to innovate the closer role, but he will likely only ever be considered a compiler of saves. And that is unfortunate.

  9. Alter Kacker says:

    As a geezer who remembers when Jerry Holtzman invented the save, the construct “modern save” strikes me as both exceedingly awkward — and personally insulting!

Leave a Reply

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