Previous entries:

The 1870s, '80s, and '90s

The 1900s and '10s

The 1920s, '30s, and '40s

The 1950s

The 1960s

The 1970s

The 1980s

The 1990s and 2000s

20003 Notes: Part I & II

Final Analysis: I, II, III, and IV.

Base Relief I, II, and III

2006 Edition: Part I, II, III, IV, and V

In what I hope is the last post to this series prior to my presenting the results at SABR36 on Thursday, I wanted to take a look at the future of relief pitching. I have been anticipating some change to the way that closers are employed since I started the series a few years ago. Why not? Every dozen or so years throughout baseball history baseball has fundamentally changed its approach to the reliever role.

In the 1870s, Harry Wright, without the ability to make substitutions, invented relieving by rotating himself from center field to the mound. In the 1890s, starting pitchers like Cy Young and Kid Nichols began to be used to also close out games. Around 1903, John McGraw invented the relief role. In the 1920s, Firpo Marberry became the first quality pitcher used mostly in relief. In the Forties, the Yankees brought relievers into the limelight with two stars, Johnny Murphy and Joe Page. In the Fifties, pure relievers started to take hold. In the Sixties, managers pushed the boundaries of what relievers were capable of doing on almost a yearly basis. In 1979, Bruce Sutter is used almost exclusively in save situations establishing a new role for relievers. In 1988, Dennis Eckersley is used almost exclusively in one-inning save situations and the rest of the A's bullpen is used to pick up the slack, establishing the current mold for relief pitching.

So it's been twenty years, but despite all the talk that using your best reliever in save situations exclusively is a waste of talent, baseball is still in love with the Eck paradigm. Closers are saving a higher percentage of their teams games than ever before and are collecting fewer innings per appearance than ever.

I've been anticipating a shift whenever I see the slightest harbinger of change, be it the 2003 Red Sox so-called "bullpen by committee" or the existence of Brooks Kieschnick, I start expecting the sky to fall. Then it occurred to me that I might be looking in the wrong place.

Could it be that the closer role has reached some level of maturity? Maybe Bill James' prediction that closers will someday collect eighty saves a year will never come true. Could it be that there is a major paradigm shift afoot that is going unnoticed?

Well, I took a look at the percentage of team saves that the average closer is collecting. The figures are going up on almost a yearly basis. However, the variation across teams has been jumping especially the last couple of years.

One of the biggest jumps was 2003 when Bill James tried to put into practice his theories of relief pitching that got mislabeled bullpen by committee. That attempt failed, but perhaps the theories did take hold. Maybe teams are using their closers in more appropriate situations rather than just in the color-by-numbers save situations.

I took a look at the men who were second in saves on their teams. Below is the annual average of team saves amassed by the man second saves on his team (for teams with co-leaders in saves, one of the co-leaders was considered the second-place save guy). Note that the numbers are declining steadily throughout the save era:

Note also that the n deviation of the values is at the same time increasing steadily with the most growth in the 2000s. What this tells me is that managers are using their closers more efficiently to collect saves, but they are tinkering with the other situations. Some use their entire bullpen, some use their closer, some use a secondary reliever regularly.

Couple this with the fact that the biggest opportunity for bullpen improvement is after the sixth and seventh innings. The percentage of leads retained after seven and especially six innings are the only figures that have been declining in the post-Eck era.

So what's in the cards for the relieving in the foreseeable future? It looks like the closer will be tweaked but will remain fundamentally unchanged. I expect the next shift to come in the rest of the bullpen.

Previous entries:

The 1870s, '80s, and '90s

The 1900s and '10s

The 1920s, '30s, and '40s

The 1950s

The 1960s

The 1970s

The 1980s

The 1990s and 2000s

20003 Notes: Part I & II

Final Analysis: I, II, III, and IV.

Base Relief I, II, and III

2006 Edition: Part I, II, III, IV, and V

In the final update to this gargantuan series, I would like to take a look at relief pitching and the Hall of Fame.

First, I should mention that whereas the press always say that there are just four relievers in the Hall (Wilhelm, Fingers, Eckersley, and now Sutter), in actuality for most of baseball history (i.e., through the Fifties), teams cannibalized their starters as relievers. All of the Hall of Fame pitchers were used in relief to some extent and many were used quite extensively, including Lefty Grove, Walter Johnson, Pete Alexander, Three-Finger Brown, Ed Walsh, Kid Nichols, Christy Mathewson, and Cy Young. Also, two of the four "relief pitcher" Hall of Famers worked as a starter for a large portion of their careers (i.e., Wilhelm and Eck). So to say that there are four relief pitchers in the Hall is inaccurate as best.

I ran the numbers for total Relief Wins (RWins) and Starter Wins (SWins) for every Hall of Fame pitcher through Sutter. I combined the two and called the results Pitching Wins (PWins) Below are the results. Listed are all of the pitchers who a) went into the Hall as a player (not an executive or manager) and b) primarily were pitchers throughout their careers. I will list the Hall-of-Famers who do not meet these criteria later—because I like listing stuff—but will not use them in the analysis.

The pitchers are sorted by PWins in descending order, and the Pitching Runs (PR), Pitching Win Shares, and various other pitching stats are listed. At the bottom are the averages for each stat along with the standard deviation for each and the min and max values one standard deviation above and below the average:

So the average Hall of Fame PWin total, whether from starting or relieving, is 33.83. 45 of the 60 Hall of Fame pitchers or 75% are within one standard deviation of the average (i.e., 15.98 to 51.68).

Next, we will take a look at all of the relief pitchers in baseball history that are within this range (more than 15.98 PWins). Since it's difficult to determine which pitchers are technically relievers, I will use a somewhat arbitrary cutoff —5 RWins—to cast a wide enough net to get all pitchers who have spent some portion of his career as a reliever.

Here are the pitchers that qualify, including nine current Hall-of-Famers:

Before you shake your head and waggle your finger at me for advocating the likes of Gary Lavelle for the Hall, I am not promoting all, or even most, of these men for the Hall, just pointing out which relievers fall within the sweep of the de facto Hall standards.

There are couple of things that do become clear, however. Three players are above the Hall average: Goose Gossage, Mariano Rivera, and John Smoltz. I am hoping that Gossage finally garners enough votes to get into the Hall next year. Rivera is not yet eligible but should be the strongest candidate of his era, meaning that even the writers couldn't screw his election up. Smoltz is a difficult case because he will probably fall short as both a starter and a reliever in the counting stats that usually get cited for Hall-worthiness, but I think this shows how he compares to a typical HOFer when you take into account his two careers. We'll have to see how well he is received by the writers after he retires.

Consider that there are eight starters that meet this criterion, and almost all have a good shot of getting into the Hall. Only one is eligible (Bert Blyleven (37.18 PWins)), and Blyeleven's candidacy is gaining momentum on the writers' ballot every year. Of the seven that are not yet eligible, Maddux (65.21), Johnson (61.33), Martinez (59.02), and Glavine (36.29) should be locks for the Hall. The rest are going to be hurt by the counting numbers but are solid picks (Kevin Brown (39.79), Mussina (35.92), and Schilling (34.48)).

It is also gratifying that a number of old timers show up in the table. I think the Vets, if they ever get their act together, or perhaps a special committee (though the one on the Negro Leagues might have put the kibosh on future committees) should take a second look at guys like Firpo Marberry, Ellis Kinder, John Hiller, Ron Perranoski, and Dan Quisenberry to name a few because of their importance to this ever-developing role.

Finally, here are the Hall of Famers, as promised, who did not meet the criteria I outlined above. Obviously, a few would have gotten in as a pitcher:

Previous entries:

The 1870s, '80s, and '90s

The 1900s and '10s

The 1920s, '30s, and '40s

The 1950s

The 1960s

The 1970s

The 1980s

The 1990s and 2000s

20003 Notes: Part I & II

Final Analysis: I, II, III, and IV.

Base Relief I, II, and III

2006 Edition: Part I, II, III

I'd like to turn to ranking teams based on Relief Wins. First, here are the best by total relief wins:

That seems to favor more recent teams who tended to give more innings to their bullpens. Next, I adjusted the numbers by the average bullpen totals for that season.

First, however, I should explain that a bullpen's worth has been exploding. Take a look at the average per decade:

OK, so here are the top twenty-five adjusted for era:

The Yankees and Indians faced off tonight with dueling Johnsons on the mound, Jason for Cleveland and Randy for New York. Unfortunately, the pairing ended up a bit lopsided with the Yankees prevailing 6-1 even though their Johnson was ejected in the process.

It was the first matchup between two starting pitchers named Johnson in over 74 years and just the 14th recorded faceoff in baseball history. The last double-Johnson pairing was on May 2, 1932, as Syl Johnson and the Cardinals edged Si and the Reds, 2-1, at home. (Syl? Si? I bet he had a sister named Sue. Thank you, Jack Benny.)

Here are all the games between starting pitchers named Johnson in baseball history. As you can imagine there a number of Walter Johnson starts in there, But just one against Walt Johnson:

The fourteen Johnson pair-ups are second all time to—you guessed it—the ones by the Smiths. No, not that annoying and incredibly overrated Morrissey band—"Bigmouth Strikes Again" was inexcusable—I mean the last name Smith. Here are the all-time rankings:

By the way, the Walt/Walter Johnson game made me wonder how often two pitchers with the same first and last name faced off. The answer is just three times, with the Bobby J. vs. Bobby M. Jones matchup from a few years back being one. The other two are from one century earlier:

Previous entries:

The 1870s, '80s, and '90s

The 1900s and '10s

The 1920s, '30s, and '40s

The 1950s

The 1960s

The 1970s

The 1980s

The 1990s and 2000s

20003 Notes: Part I & II

Final Analysis: I, II, III, and IV.

Base Relief I, II, and III

2006 Edition: Part I, II

I've rerun all the numbers and here are the most valuable of all time. First, the single-season best.

Keep in mind that the reliever types are as follows:

OK, here are the most valuable reliever years by Relief Wins:

Now, here are the worst (or most damaging) seasons:

Now, here's the kicker. Here are the most valuable relievers of all time:

Now, the best in Relief Wins per year:

You may notice that if Mariano Rivera has an average (for him) season this year, he will pass Hoyt Wilhelm as the most valuable reliever of all time. But even so, it is apparent that the best three relievers by a long shot are Wilhelm, Rivera, and Goose Gossage. Hopefully, Goose will get the call to the Hall next year.

Now, here are the worst career relievers:

Previous entries:

The 1870s, '80s, and '90s

The 1900s and '10s

The 1920s, '30s, and '40s

The 1950s

The 1960s

The 1970s

The 1980s

The 1990s and 2000s

20003 Notes: Part I & II

Final Analysis: I, II, III, and IV.

Base Relief I, II, and III

2006 Edition: Part I

As promised, here are the updated middle relief leaders for Relief Wins based on the updated runs saved-to-relief wins matrix. First, here are the all-time best:

Now here are the worst all time:

Next, I will meld the middle relief data in with the rest of the relief pitcher classifications and will present the all-time best for a single season as well as per career.

Previous entries:

The 1870s, '80s, and '90s

The 1900s and '10s

The 1920s, '30s, and '40s

The 1950s

The 1960s

The 1970s

The 1980s

The 1990s and 2000s

20003 Notes: Part I & II

Final Analysis: I, II, III, and IV.

Base Relief I, II, and III

In order to present my old study on relief pitching at SABR 36, I decided to update the stats and to also resolve an issue that cropped up in the original study. The problem is that I jury-rigged the stats for middle relievers by extrapolating between starters and old-tyme closers. The results were that, I felt, the evaluations for the best middle relievers were way out of skew.

Now that I have loaded Retrosheet's game log data, I can derive how well saving a run led to wins in different scenarios for different eras. I used the data for starting pitchers and closers to determine the scenarios that would apply to middle relief and to weight the various scenarios.

Here are the results for middle relievers by era. Note that they have not had much impact on the game until the late Seventies:

I also listed the average appearance length for a middle reliever and for the starting pitcher for each era. Now, we can plug these numbers into my formulae for middle relievers, stir, and then we will have updated middle relief data.

I'll update all the stats to include the new middle relief numbers as well as all data through 2005 tomorrow.

I know that I said I would look at the overall franchise-versus-franchise records, but I want to take one more look at the season-series records more closely first. I thought that looking at just those teams that competed for a title would be a bit more interesting and more to the point.

So here are the closest (in terms of average margin of victory) season series of all time involving two teams including at least one first-place or playoff team. I limited the games apart between the two to less than six games.