Last season, Baltimore closer Kevin Gregg turned in a season unique in the annals of relief pitching:
With 22 saves and 48 games finished, Gregg became the first pitcher to log 20+ saves and/or 25+ games finished without inheriting a single runner all year. (Inherited runners have been tallied only since 1950, but I’ll personally guarantee that this never happened before that, either.)
There were 28 pitchers with at least 20 saves in 2011. As a group, they averaged 34 saves and 12 inherited runners; 16 of them inherited 10 runners or less. John Axford, the NL Saves co-leader, inherited just 1 runner all year, and that was in a non-save situation. (He stranded the runner, but lost the game in the next inning.)
Among 30-save men last year, the high in inherited runners was 17, shared by Mariano Rivera, Juan Oviedo (f/k/a/ Leo Nunez) and Carlos Marmol. Rivera inherited at least one runner in 6 of 49 save tries, Oviedo 9 of 42, and Marmol 7 of 44 save chances, for a combined rate of 16% of their save tries.
As shown in the following charts, the rate of inherited runners for closers has been declining since the mid-’70s. The rates are shown in two forms, Per 30 Saves and Per 60 IP (the average workload of a modern closer). Since closers of past eras tended to pitch more innings and record fewer saves than their modern counterparts, the rates Per 30 Saves tend to be much higher than Per 60 IP until the ’90s.
Each chart covers a different set of pitchers from 1961-2011 — pitchers with 10+ saves, pitchers with 20+ saves, and the yearly MLB saves leaders.
Chart 1: Pitchers with 10+ Saves
Chart 2: Pitchers with 20+ Saves
(Note that 1961 and ’67 saw just one pitcher with 20+ saves)
Chart 3: MLB Saves Leaders
In Chart 3, the 1989 spike is Mark Davis, who inherited 75 runners (stranding 79%) en route to 44 saves and the NL Cy Young Award (followed by one of the most unproductive free-agent contracts ever seen). Four other closers that year had between 54 and 65 inherited runners; no closer since then has had more than 49.
Other landmarks in saves and inherited runners:
- Ted Abernathy notched the first 30-save season in 1965. He inherited at least one baserunner in 26 of his 38 save chances. Over all (including non-save outings), he inherited 59 runners, of whom 18 scored.
- In 1969, Cincinnati’s Wayne Granger, working for Dave Bristol, set the all-time mark by inheriting 112 runners over 90 games. In 62 of those 90 games, Granger came on with at least 1 man base; 11 times he entered with the bases full. Granger finished 55 of 90 games and logged a hefty 145 relief IP in ’69, with 27 saves.
- In 1972, Rollie Fingers pitched in 9 of the 12 postseason games, inheriting at least one runner in every game but one. Fingers inherited 12 runners in all, stranding 11; the only one he let in was already on 3rd base with no outs when Rollie entered. During the regular year, Fingers stranded 38 of 44 inherited runners (86%).
- John Hiller had 38 saves in 1973, a new record that would stand for a decade, with just 4 blown saves. Hiller inherited at least one runner in 35 of his 42 save chances, and 52 of his 65 games. In all, he inherited 84 runners — equal to the combined total for last year’s Top 10 in Saves — and stranded 85% of them. Hiller inherited 2 or 3 runners 29 times.
- Bruce Sutter won the 1979 NL Cy Young Award with 37 saves. Sutter inherited one or more runners in 29 of 47 save tries, and 36 of 62 games. He inherited 55 total runners, and 13 scored.
- The following year, Dan Quisenberry helped the Royals to their first pennant, with 33 saves (just 4 blown) and 12 wins. Quis inherited a runner in 29 of 37 save chances, and 54 of 75 games (and 2 or 3 runners in 30 games). His 89 inherited runners in 1980 remains the record for a 30-save season (shared by Mike Marshall in ’73).
- In 1992, Tom Henke logged the first 30-save season while inheriting less than 10 runners. In 37 save chances, Henke inherited just 5 runners (7 all year).
- In ’93, Jeff Russell was the first to log 30 saves while tossing less than 50 innings. Russell’s total of 46.2 IP was 90 less than Abernathy worked in the first 30-save season.
- In 2002, Eddie Guardado had 45 saves (AL high) while inheriting runners just once — with a 3-run lead, 2 men on and 1 out to get.
- En route to 46 saves and the career saves record in 2006, Trevor Hoffman dealt with just 1 leftover runner — with a 3-run lead and 1 out to get.
- On September 18, 2008, five days after breaking the season saves mark, Francisco Rodriguez entered with 1 out in the 9th, a 5-run lead and the bases loaded (the only time all year that he entered with the sacks full). He let in all 3 runners, but got the 2 outs for (ahem) “save” #59. K-Rod’s career strand rate of 32% is exactly the same as the MLB average during his career.
P.S. Although this topic cries out for data on percentages of all saves that featured inherited runners, the Play Index cannot search for inherited runners in a game.
As I posted back on the old site when we were discussing the value of low-inning relievers in a league full of specialists/LOOGYs and even the occasional ROOGY (I believe on either a Steve Reed or Eric Plunk post), I really hope that one day a manager will buck the trend, do away with the current CW and bring back the rubber-armed reliever, whether as a closer or set-up man.
Save us, Mike Marshall, you’re our only hope!
This subject just frustrates me to no end. I cannot fathom that somewhere there isn’t a major league manager with the balls and brains to use his best relievers in a way that will provide the greatest payoff.
I think baseball needs a rule change of some sort- like other than the starting pitcher a manager cannot remove a pitcher in the middle of an inning until he has faced x number of batters or given up a run or something of that sort. This nonsense of managers using 5 or 6 relievers a game is does more to detract from the enjoyment of the game than anything I can think of.
If they can’t or won’t change the rules in that manner, then the least they should do is require the manager to do a song and dance routine on top of the opposing teams dugout from the time the pitcher leaves the mound until his replacement is ready to go. Visiting fans should have the option of being able to purchase over-ripe fruit and vegetables from vendors.
That ought to speed things up a bit.
Not sure if you’re aware of this, but Mike Marshall now teaches pitching mechanics. He claims his mechanics will eliminate pitcher injuries. His theories are very controversial though and hardly anyone has adopted them. I remember seeing a tv show about them a few years ago and it sounds like he has a hard time getting anyone to listen to him.
http://drmikemarshall.com/
The rise of the 1-inning reliever is fascinating, and obviously hugely correlated with the lack of inherited runners you cite above. It’s hard to have inherited runners when you’re only used to start an inning!
That reminds me of a post I did last year about the longest streaks of exactly 1 inning pitched. John Axford nearly broke the record last year, but fell 4 games short: http://junkstats.com/mlb/the-longest-streaks-of-exactly-one-inning-pitched/
Great post. I started watching and understanding baseball in 1977, and my home team, the Padres, had signed Rollie Fingers. I was used to him coming in the 7th and 8th to pick up saves (one time, he came in the game in the 6th!). Now that saves have become the meaningless stat they are, I’m glad there is no “Fireman of the Year”. Interesting on how games are managed now, as opposed to 35 years ago.
Before his Hall of Fame induction, Goose Goosage was asked about todays set up relievers and he said: I WAS my own set up reliever. Today, a manager will bring the setup guy during the 8th inning only, and the closer in the 9th. They don’t mind if runners are on base or the opponent’s middle of the order is coming up in the 8th inning or anything. If they blew the lead in the seventh or eight, then, what do you need the closer for? They should bring the strongest arm when there is the bigger threat.
It’s easy to assume that today’s usage of relievers is less efficient than in the past, yet some of the emerging data may disagree. Tango has been looking at this very issue. For those interested, follow the link.
http://www.insidethebook.com/ee/index.php/site/comments/are_relievers_being_used_optimally_compared_to_1980/
That’s a fair point and definitely worth studying — thanks for the link.
Besides efficiency of reliever usage, though, there’s the question of roster efficiency. Comparing roster composition for 1980 and 2011 (the years of Tango’s study), we find that the average team now carries 2 more pitchers.* So even if the 2011 reliever win value is as much or more than 1980, it’s debatable whether teams actually get more wins from the strategy.
I forgot the footnote to my comment #9….
* Avg. number of pitchers per team with 30+ innings:
— 1980, 11.2
— 2011, 13.6
Agreed, and no conclusion being made. Tango’s work may show that relievers today ar doing a better job in high-leverage situations, yet relievers today may not have to face as deep a bench as relievers did years back. Basically, the extra guy or two on the bench has now been turned into an extra guy or two in the pen. When a weak LOOGY is called in, a manager may not have another strong righthanded bat on the bench to face the lefty, where twenty/thirty years back that probably would not have been the case.
I included the link because there’s been a few intersting articles, yours included, on the subject recently, including Tango’s, Dave Cameron and Steven Goldman. Lots of interesting thoughts and data, but no clear conclusions.
As with many baseball related items, my introduction to the concept of the “Fireman” came in the form of a baseball card:
http://www.checkoutmycards.com/Cards/Baseball/1976/Topps/205/NL_AL_Leading_Firemen_Al_Hrabosky_Rich_Gossage/1853904
The card includes Al Hrabosky – and the first time I ever saw his “Mad Hungarian” routine was on “This Week in Baseball”.
There are still firemen today – they’re just not the guys getting saves. These firemen are the guys who bridge from the starter to the “8th inning guy”, coming in in the 6th or 7th inning if the starter gets in trouble. Typically, they’re a collection of guys, rather than a designated specialist, and usually these are the only guys who commonly come in with men on base.
The practice of having designated 8th inning and 9th inning guys seems akin to the notion of prevention being preferable to cure. Prevent the fire from starting, instead of putting it out. This approach puts less pressure on the entire pitching staff, so probably more likely to yield wins than the true “fireman” practices from the 1970s.
Would be interesting to look at whether win probability when leading after 6, 7 or 8 innings has improved since the 1970s. That could probably be a good measure of which relief approach is preferable.
Good idea, Doug. I’ll see what I can do on that angle.
Doug, I looked at the MLB records when leading after N innings for 1980 and 2011. Very little difference:
(1980 listed first)
Leading after 5 innings: .820, .818
Leading after 6 innings: .848, .860
Leading after 7 innings: .897, .901
Leading after 8 innings: .946, .951
Also, even if there were differences that seemed significant, we still haven’t isolated the effect of modern reliever usage. One consequence of the extra 2 pitchers modern teams carry (compared to 1980) is fewer quality bats available to pinch-hit, which further depresses late-inning scoring.
P.S. I included “after 5 innings” since, in 2011, 34% of all starters came out before completing 6 IP.
I remember a while back Joe Posnanski did a post about the winning percentages of teams leading after 8 innings. IIRC, it’s been incredibly stable throughout history, and also has very little variation from team to team. Intuitively, that makes sense: we expect teams leading after 8 to win at a very high rate, whether their closer is Mariano Rivera or Byung-Hyun Kim.
Thanks for taking a look, John.
Good point about the counter-effect of reduced pinch-hitter usage. That would be worthy of a study too.
Pingback: Orioles 2012 closer: Should Jim Johnson worry about Matt Lindstrom? « RotoDiamond
Wondering if contemporary managers don’t consider using their best reliever (i.e. the closer, theoretically) in 7th-8th inning situations because of the risk of having them warm up several times in a game, and therefore a great many more times over a season than is currently acceptable for such a valuable arm.
It is probably better to have guys pitch fewer innings/game in relief because
1) the pitcher can “bring it” more and not worry about spreading out his effort over several innings
2) hitters wont face the same reliever twice, there is a definite increase in batter production when they face the same pitcher for the second time in a game
3) you can use the pitcher in more games/less rest is needed
The problem is that all managers are too dumb to use the reduced (compared to 1980) innings from their best relievers at the most important time.
Basically, the current strategy is the analogous to this scenario:
You have Albert Pujols on your bench for some reason, bottom of the 8th bases loaded down by 1, and you choose to leave Pujols on the bench because you only want to use him in the 9th because the 9th is more important. So you let the current batter hit and use Pujols to lead off the 9th inning no matter who is due up in the 9th even if it is your 2nd best hitter (just like a manager will pull the 8th inning man even if he strikes out the side and bring in his closer even if he is struggling lately).
Very frustrating to watch your best arm sitting in the pen not facing the best hitter with men on base in the 8th inning.
I like your Pujols analogy.
Topper you are right, but Tony LaRussa disagrees (remember the All Star Game?). 🙂
Good post. I fully expected the last graph to finish off the way it did.
I remember a lot of articles about the bullpen usage of the 2007 Indians. Basically, people postulated that the set-up men were used very efficiently, very close to “optimal” usage, but it was likely completely by accident. Joe Borowski, the “established closer,” was only used in save situations, whereas Betancourt and Perez were used in higher leverage 7th/8th inning situations. Borowski’s overall numbers stunk, but he saved games at an average/above average rate, so he didn’t drag the rest of the bullpen down.
So Betancourt, Perez, and even Jensen Lewis served in a “fireman” type role that season, but I doubt that was Wedge’s intention from the beginning of the season.
I remember that teams as well. I think there was another issue. Lots of people felt that Betancourt or Perez should be moved into the closer role. The problem with that is you’d be eliminating one of your set-up guys with no one to replace him with (you certainly weren’t going to use Borowski and his 5+ ERA in a set-up role).
I remember this JoeBo game from 2007. Not a save situation; he started the 9th with a 4-run lead in Yankee Stadium, got the first 2 outs — and then gave up 6 runs, losing on A-Rod’s walkoff 3-run shot.
http://www.baseball-reference.com/boxes/NYA/NYA200704190.shtml
Borowski’s -0.978 WPA from that game is the 15th-worst relief outing of 1 IP or less ever recorded.
One of many games that year that had Indian fans lining up to personally strangle JoeBo.
I remember that game. Borowski was terrible, but IIRC, it was Wedge’s brilliant idea to pitch to A-Rod instead of walking him and loading the bases for Giambi after the wild pitch. Wedge only cared about keeping the righty/righty matchup, and seemingly ignored that A-Rod was the hottest hitter in the AL at that point of the season, coming into the game with a 1.416 OPS.
Ah, yes — the last great A-Rod season. After 18 games, he had 14 HRs, 34 RBI, 26 Runs, and a 1.507 OPS.
Judging by the apparent demographic on this web site, I doubt there are many ovarian cyst sufferers.
That said, K-Rod’s poor strand rate only reinforces my disdain for him. Easy guy to root against.
Lee Sinins (the same man who suggested changing the words Winning Pitcher to Guy Who Had His Butt On The Bench When A Meaningless And Outdated Statistic Occurred) once called Mariano Rivera the greatest pitcher in the history of baseball at getting three outs before giving up three runs. I’ve lived nearly my entire life in the Tony LaRussa era (and he was my favorite manager up until he abandonned the Bay for St. Louis), and I’ve always felt envious of the days when pitchers were expected to pitch, not to get one or two guys out.
I can’t understand why anyone would dog Rivera. He has posted an ERA under 2 in 11 seasons and has a lifetime ERA+ of 206. It’s not his fault that his managers have used him like every other manager in MLB would. Is there any doubt that he has been the best and most valuable reliever of the last 20 years?
I should mention that Sinins said the same basic thing about Rivera, that it’s not his fault that he was used in that way.
Pingback: High Heat Stats » Strand Leaders: The best when you’re in a mess
I hardly leave a response, however i did a few searching and wound up here High Heat Stats
I read this post completely about the resemblance of most up-to-date and earlier technologies, it’s remarkable article.