1.  What I Would Do With the All-Star Game

If It Was Up To Me

            First of all, I’d cut the rosters to 18 players.   I used to be more radical on this, back when the managers had entirely lost track of trying to win the game and were just trying to make sure that everybody got one at bat.   I couldn’t stand that. .. long strings of at bats the ONLY purpose of which was to enable people to say that “I played in the All Star game”.   When they were doing that I wanted to cut the rosters down to 14.   I have since realized there might be actual problems with playing the game with a 14-man roster, so I’ll agree to 18.

            The composition of the rosters would be largely determined by the fans’ votes, in this way.   Each city would be its own “precinct” for the voting, and each ballpark would vote only on the players who played there.   In other words, the Pirates’ fans would vote on which Pirates’ player they wanted to send to the All-Star game; the Dodger fans would vote on which Dodger to send.   Balloting would be only at the ballpark, and only for the players who played in that park.

            The manager would pick the roster, within these limits: 

            1) That he must pick one player from each team,

            2) That he must pick at least 8 players who finished first in the balloting in their precinct,

            3) That he must pick at least 12 players who finished 1 or 2 in the balloting in their precinct,

            4) That he must pick at least 16 players who finished 1, 2 or 3 in the balloting in their precinct, and

            5)  That he cannot pick more than two players from any team.

Four players from the host team or the closest team in the other league would be designated as injury backups, on hand to insure that the game can be completed.  

            There would never be a situation in which it would be difficult to choose an outstanding team within these rules, absolutely never.   In 2009, the options might be something like this:

            You look at that as a manager, you’ve got to think “Wow.  I get 18 of these guys?  I’m going to be OK.”   You’d start by looking down the list of #1s that you would want on the team. . ..Mauer, Greinke, Halladay, Longoria, Markakis.    At second base I’ve got Pedroia or Kinsler; either one of them is OK, so I’ll choose one of them later.  I don’t have a first baseman who is a “1”, so my options there are Youkilis, Victor Martinez, Morneau, Teixeira.. ..Billy Butler, but I don’t want Billy Butler and Greinke is the only Royal I need, so let’s scratch him.   If I choose Teixeira I don’t have a shortstop, plus I may need Mariano for the bullpen, so let’s skip him, so that makes Jeter the shortstop, and I’ll pick Victor Martinez as a first baseman or DH.    What do I have. .

            C—Joe Mauer

            1B—Victor Martinez

            3B—Evan Longoria

            SS—Derek Jeter

            RF—Nick Markakis

            P—Roy Halladay

            P—Zack Greinke

            I’ve got seven number ones already; I need a second baseman, two outfielders, and several pitchers.   Let’s put in Kinsler as the second baseman, and I don’t need to worry about the eight number ones any more.   Left and center I can use Holliday and Torii Hunter:

            C—Joe Mauer

            1B—Victor Martinez

            2B—Ian Kinsler

            3B—Evan Longoria

            SS—Derek Jeter

            LF—Matt Holliday

            CF—Torii Hunter

            RF—Nick Markakis

            P—Roy Halladay

            P—Zack Greinke

            So at this point I need a DH and four five pitchers, and I’ve got to cover the Red Sox, White Sox, Tigers and Mariners.   Maybe I should have used Granderson in center;

C—Joe Mauer

            1B—Victor Martinez

            2B—Ian Kinsler

            3B—Evan Longoria

            SS—Derek Jeter

            LF—Matt Holliday

            CF—Curtis Granderson

            RF—Nick Markakis

            P—Roy Halladay

            P—Zack Greinke

            OK, now I’ve got to cover the Red Sox, White Sox, Angels and Mariners.   I’ll choose Buehrle and Jered Weaver because that gives me four starting pitchers; I can get nine innings out of them without any problem:

C—Joe Mauer

            1B—Victor Martinez

            2B—Ian Kinsler

            3B—Evan Longoria

            SS—Derek Jeter

            LF—Matt Holliday

            CF—Curtis Granderson

            RF—Nick Markakis

            P—Roy Halladay

            P—Zack Greinke

            P—Mark Buehrle

            P—Jered Weaver

            OK, I’ll take Jason Bay from the Red Sox to DH and Ichiro from the Mariners, and I’ve covered every team:

C—Joe Mauer

            1B—Victor Martinez

            2B—Ian Kinsler

            3B—Evan Longoria

            SS—Derek Jeter

            LF—Matt Holliday

            CF—Curtis Granderson

            RF—Nick Markakis

            OF—Ichiro Suzuki

            DH—Jason Bay

            P—Roy Halladay

            P—Zack Greinke

            P—Mark Buehrle

            P—Jered Weaver

            I’ve got four picks left, and I can pretty much pick whoever I want.  I’ll take Youkilis, because that gives me a first baseman or a third baseman, and that will enable me to use Martinez as a catcher if I need to.  I’ll take Michael Young because he can play second, third or short, and two closers. . .let’s say Mariano and Joe Nathan.  

            Now that’s an All-Star team:

C—Joe Mauer

C-1B—Victor Martinez

1B-3B—Kevin Youkilis

            2B—Ian Kinsler

            3B—Evan Longoria

            SS—Derek Jeter

            INF—Michael Young

            LF—Matt Holliday

            CF—Curtis Granderson

            RF—Nick Markakis

            OF—Ichiro Suzuki

            DH—Jason Bay

            P—Roy Halladay

            P—Zack Greinke

            P—Mark Buehrle

            P—Jered Weaver

            P—Mariano Rivera

            P—Joe Nathan

            You’ve got about 12 Hall of Famers there, eleven #1 picks in the voting, three #2 picks and four #3 picks.   I’d pay to watch those guys play a baseball game.    Chone Figgins taking his at bat and getting congratulations for being an All-Star… .I’ve seen enough of that.  No city gets shorted in the voting, nobody gets to go to the All-Star game because his old manager owes him a favor and he’s having a decent year.  It’s all stars.  Almost everybody on the roster is going to get into the game.  St. Louis is the host city, so Kansas City sends four guys to protect the roster in case of injuries. . .we’ll take Soria, obviously, Teahen because he can play infield or outfield, Bloomquist because he can play anywhere, and a catcher (doesn’t matter which one).

            That was kind of fun; let’s do the NL.   The voting might go something like this:

            We’ll start with the most obvious picks—Hanley Ramirez, Adrian Gonzalez and Albert Pujols.   Brian McCann looks like the only catcher who made any list anywhere, so we’ll include him:

            C—Brian McCann

            1B—Albert Pujols

            SS—Hanley Ramirez

            DH—Adrian Gonzalez

            We’ll pick Zambrano and Matt Cain to give us a start on a pitching staff, and Ryan Braun and Justin Upton so that we have a couple of outfielders:

C—Brian McCann

            1B—Albert Pujols

            SS—Hanley Ramirez

            LF—Ryan Braun

            RF—Justin Upton

DH—Adrian Gonzalez

            P—Carlos Zambrano

            P—Matt Cain

            Third base is between Ryan Zimmerman and David Wright, and we’ll choose Zimmerman because I want to get Santana, who I think is actually the league’s best pitcher, on the team if I can.   At second base we’ve got Utley, Hudson and Sanchez.   I’m going to choose Utley because I think he is just the best player, even though it makes the other options more complicated:

C—Brian McCann

            1B—Albert Pujols

            2B—Chase Utley

            3B—Ryan Zimmerman

            SS—Hanley Ramirez

            LF—Ryan Braun

            RF—Justin Upton

DH—Adrian Gonzalez

            P—Carlos Zambrano

            P—Matt Cain

            P—Johan Santana

            OK, we’ve got to cover Cincinnati, Houston, Colorado, LA and Pittsburgh, and we need a center fielder, pitching, and bench players.   We’ll take Francisco Cordero for the bullpen, and Matt Kemp looks like the best option in center field:

C—Brian McCann

            1B—Albert Pujols

            2B—Chase Utley

            3B—Ryan Zimmerman

            SS—Hanley Ramirez

            LF—Ryan Braun

            CF—Matt Kemp

            RF—Justin Upton

DH—Adrian Gonzalez

            P—Carlos Zambrano

            P—Matt Cain

            P—Johan Santana

            P—Francisco Cordero

            We’ll take Brad Hawpe from Colorado and Freddy Sanchez from Pittsburgh, because those are probably the best players on those teams, and that gives us an infielder and an outfielder; Sanchez can play second or third, and Miguel Tejada from Houston gives us two shortstops:

C—Brian McCann

            1B—Albert Pujols

            2B—Chase Utley

            3B—Ryan Zimmerman

            SS—Hanley Ramirez

            SS—Miguel Tejada

            INF—Freddy Sanchez

            LF—Ryan Braun

            CF—Matt Kemp

            RF—Justin Upton

            OF—Brad Hawpe

DH—Adrian Gonzalez

            P—Carlos Zambrano

            P—Matt Cain

            P—Johan Santana

            P—Francisco Cordero

            What do we need here. . .we need a catcher and a pitcher.  I’ll take Yadier Molina of St. Louis and maybe Lincecum.  That gives us:

C—Brian McCann

C—Yadier Molina

            1B—Albert Pujols

            2B—Chase Utley

            3B—Ryan Zimmerman

            SS—Hanley Ramirez

            SS—Miguel Tejada

            INF—Freddy Sanchez

            LF—Ryan Braun

            CF—Matt Kemp

            RF—Justin Upton

            OF—Brad Hawpe

DH—Adrian Gonzalez

            P—Carlos Zambrano

            P—Matt Cain

            P—Johan Santana

            P—Francisco Cordero

            P—Tim Lincecum

            For injury backups we’ve got the Cardinals; I’ll take Schumaker because he is an infielder/outfielder and having a good year, Colby Rasmus because we really need help in center field, Ryan Franklin to protect the bullpen and probably Wainwright as well.

            I think it’s a better system; I think it makes a better game with stronger rosters, more meaningful participation from the fans, and it makes selection to the game a real honor.    New York and Boston and LA fans can’t swamp the voting because there are more of them; they can only vote for their own guys.  Fans are not asked (or not allowed) to vote on hundreds of players, many of whom they probably haven’t thought about all season.   They’re asked to sort out the players on their home team.  Here’s a few other rules I’d put in place to make it work:

            1)  Home-town injury backups can only be added to the active roster by the consent of the commissioner.

            2)  Starting pitchers are placed on the inactive list BY THE LEAGUE for the Saturday and Sunday before the All-Star game.  They can be replaced on the active roster for those games by minor league callups, without placing those players on the 40-man roster in the normal way.   In other words, it’s an “exceptional callup”—an exception to the normal rule that a player called to the majors must be placed on the 40-man roster.   San Francisco loses two starting pitchers for Saturday and Sunday, but they can call up Kevin Pucetas and Ryan Sadowski for those two games if they want to.  

            3)  Relief pitchers are placed on the inactive list by the league for the Sunday before the All-Star game, and can also be replaced on the roster for that day.

            4)  Players who are selected for the All-Star game and choose not to participate are replaced on the All-Star roster by hometown alternates, and are automatically ineligible to play for seven games following the All-Star break.  

            5)  If the manager feels that he is unable to form a satisfactory roster from the players selected by the fans as the top three options on the team, he can petition the commissioner for an exemption from the rules requiring eight #1 fan selections, twelve one or two selections, and sixteen one-three selections.

            It’s not going to happen; there’s never going to be a situation where you can’t choose a representative roster from the top three players on each team. But just remembering Murphy’s Law, you would need a plan in place in case it does. . .in case, for example, the top three players listed by American League fans include no shortstops and no catchers.    Or in case the four first basemen available to the American League all pull up with actual injuries in the last week.   But remember—if you don’t play in the All-Star game, you don’t play in the seven games following the All-Star game.  That’s just the rule.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

2.  Percentage of Full Career

            Whenever I look at Willie Mays’ record, I am always astonished by how close he came to realizing 100% of his career potential.   He was in the majors when he was 20; he was there when he was 42.   He played something very close to the full schedule every year until he was 35 years old, and he had good years every year; he never took a year off and hit .255 with 18 homers, the way almost everybody else did.   The only part of his career that he is missing is the 1952-1953 portion when he was in the Army.   That took about 1.8 years out of his 20-year career, so that’s about a 9% gap, but other than that, he was astonishingly close to having a perfect career.

            Mickey Mantle, you wouldn’t say that; Mantle was as good or better than Mays at his best, but he had some injuries and was basically finished at 32.   The players in the current generation who are (so far) having near-perfect careers are Albert Pujols and Carlos Beltran, and on the other end there is Milton Bradley and Carlos Pena; if it’s not one damn thing with those guys, it’s another.   Ted Williams has major gaps in his career, when he was out of action with military obligations; Pete Rose is free of major gaps. 

Norm Cash played for a long time, but he had a lot of years when he just really was not that good.  Is there some way, I wondered, to estimate what percentage of a player’s “career potential” the player has been able to achieve?

            I decided to approach it this way.   We establish the player’s career potential by his three best consecutive years.   In other words, since Willie Mays’ three best consecutive seasons are 1963-64-65 (a total of 119 Win Shares), his “peak value”, which establishes his potential, is 39.7.    Since Mickey Mantle’s three best consecutive seasons are 1955-56-57 (a total of 141 Win Shares), his “peak value” is 47.0.   Since Keith Hernandez three best consecutive seasons are 1984-1986, a total of 89 Win Shares, his peak value is 29.7.  Since Don Mattingly’s three best seasons are 1984-1986, a total of 95 Win Shares, his peak value is 31.7.

            Not quite; I had to make another adjustment later on.   I am getting this out of sequence, but when I got some data later on, one of the problems with it was that I had too-high percentages for some players—like Mattingly—who reached very high peaks early on in their careers, but then were prevented by injuries from moving on to another level in what should have been their prime seasons.   To adjust for that, I figured two “peaks” for each player—his three best consecutive seasons, and his three best consecutive seasons ending no later than age 25.   The second figure, his “young peak”, I increased by 10%.   If the “young peak”, increased by 10%, was greater than the other peak, then I used the young peak.   Mattingly’s peak three-year period was increased from 95 to 104.5, because he was young when he had those years, and thus his peak value was increased to 34.83. 

            This peak value, to get the career potential value, was multiplied by a number of years representing a full career.   A full career is longer for a superior player than for a weaker player, since the best players can stay around longer.   After experimenting with different formulas, I assigned each player a potential number of years, which was:

            11.2 years (every player has a potential of at least 11.2 years in his career),

            Plus his three-year peak, to the power 0.400. 

            Mattingly’s three-year peak (adjusted) is 104.5.   104.5 to the power 0.400 is 6.42.  6.42 plus 11.2 is 17.62, so Mattingly has a potential of 17.62 seasons.   17.62—his potential seasons—times 34.833—his peak value—is 614 Win Shares.   Don Mattingly has a potential value of 614 Win Shares.  

            Mattingly actually earned 263 Win Shares in his career.   Thus, what Mattingly was actually able to do in his career was 47% of his career potential—263, divided by 614.  

            I should be done explaining the method here, but I had to make one more stupid adjustment.   I had very high values, at this point, for those few players—almost all of them pitchers—who were able to hang around to an advanced age.   Tommy John, for example, now shows at 289 over 290.   He shows as having essentially a 100% full career.  

            Tommy John, as we know, did not have a 100% full career; he took off a couple of seasons in mid-career to invent Tommy John surgery.   But he shows as having a full career because:

            1)  His peak value is not terribly high, and

            2)  He pitched until he was 46.

            He won 13 games when he was 44 years old, 9 when he was 45.  

            I needed another adjustment.   To avoid guys like John (and Moyer) showing up as having full careers when we know they didn’t, I discounted their Win Shares, after the age of 42, by two-thirds.   This only effects a handful of players in major league history, since there are only a handful of players who have been effective in the majors past the age of 42—and almost all of those players, as we will see later, still show up with a very higher percentage of their potential value achieved.

            Let’s look at some players, chosen kind of at random, but establishing a pattern which you will see shortly:

            That’s about the lowest it can go, just in the way that we’re measuring it.   Players do, of course, show potential in less than three years, but we’re measuring potential over a three-year period, which makes it difficult for our measurements to go much lower than 20%. 

            Let’s go on to some lists now.   The first thing we have to do, to make our lists work, is to get rid of the 19^th-century pitchers.  If we don’t get rid of the 19^th-century pitchers, most of our lists will be completely dominated by 19^th century pitchers, many of whom were spectacularly good for one or two years—often when they were 19 and 20 years old—but many of whom had very short careers.   Nineteenth century baseball really is not major league baseball, regardless of what people might tell you.  

            OK, 19^th century pitchers are gone.   There are basically four questions that we need to ask:

1)  What players came closest to achieving 100% of their full career potential?

2)  What players had the lowest percentage of their potential achieved?

3)  What players had the great potential careers ever?

4)  What are the normal achievement percentages?

            With variations on that for Hall of Fame and non-Hall of Fame and pitcher and non-pitcher categories:    

1)  What players came closest to achieving 100% of their full career potential?

            By my math, the number one player of all time, in terms of achieving his full career potential, was Jake Beckley, a turn-of-the-last-century first baseman who had 2,930 career hits and is in the Hall of Fame.  Beckley, who hailed from the home town of Mark Twain and Joe Hardy, was never a great player; he was just a good player for a really long time:

            As you can see, most—but not all—of the players that we recognize as achieving a very high percentage of their potential career are outstanding players, and Hall of Fame players.   Perhaps the most surprising name on the list is Chili Davis.   When Chili Davis came to the majors in 1982, he was regarded as having a very high upside.   He had knee troubles early in his career, and for much of his career he was regarded as a player who wasn’t what he should have been, because of his early loss of speed.

            Our system doesn’t see it that way.   The way our system sees it is, here is a player who hit just .233 with 11 homers in his second major league season, and who up to the age of 32 had never driven in 100 runs and had hit .300 just once.   He stayed in the lineup and continued to progress as a hitter until he was in his mid-30s.   He played 154 games when he was 22 years old; he played 146 games when he was 39.   Although he was never a great player, he was healthy and productive throughout that long span.   He had a full career. 

2)  Among players of modest skills, who came closest to having a full career? 

            That list would be dominated by pitchers.   My top 25 are Danny Darwin (93%), Bob McClure, Don McMahon (86%), Clarence Mitchell, Jim Dwyer, Mike Stanton, Ray Sadecki (80%), Woodie Fryman, Johnny Klippstein, Tom Gordon, John Burkett, Jeff Reardon, Mike Timlin, Kent Mercker (75%), Rick Honeycutt, Royce Clayton, Grant Jackson, Bob Forsch, Stan Javier, Todd Jones, Dennis Cook, Rick Aguilera, Don Slaught, Alan Ashby and Rube Benton (70%). 

3)  How about non-pitchers?

            Clarence Mitchell was a half-and-half, pitching in 390 games but getting into another 259 as first baseman, outfielder or pinch hitter.   Setting those aside, I have Jim Dwyer (81%), Royce Clayton (75%), Stan Javier, Don Slaught, Alan Ashby (70%), Heinie Peitz, Wilbert Robinson, Sandy Alomar Jr., Todd Pratt, Greg Myers, Rollie Hemsley, Charlie Moore, Dave Martinez, Otto Miller, Howard Shanks, Malachi Kittredge, Charlie Ganzel, Jimmy Austin (65%), Tom Brookens, Pat Kelly, Walt Weiss, Joe Quinn, Ed Kranepool, Joe Girardi and Blondie Purcell (64%).   Blondie is another half-and-half, so I owe you one.   Clyde McCullough.  

4)  Who had the lowest career percentages?

            Jim Viox ranks last in my study, not because he probably deserves to, but because he perfectly fits the criteria of the study:  He played regularly for three years, played very well, but then disappeared quickly from the major leagues.  I didn’t include anyone in my study who didn’t earn at least 50 Career Win Shares; Viox had 61.  He was the second baseman for the Pirates at the tag end of the Honus Wagner era, had an excellent .361 career on-base percentage in a very short career.   I have Viox as achieving 18% of his career potential. 

            Behind him I have a pitcher named Fred Blanding, who pitched very well for the Cleveland Indians in 1912-1913, disappeared from the majors in 1914, presumably because of an arm injury.  

            Then we have what is at least a recognizable name, Sam Jethroe.  Sam Jethroe was a Negro League star who was in his thirties by the time the color line broke, a good center fielder who scored 100 runs for the Braves in 1950 and 1951.   Then we have Boo Ferriss—a 25-game winner with the 1946 Red Sox—and Milt Byrnes, a war-time third baseman who helped the Browns win the pennant in 1944, never got into a game before or after the war despite a nice .373 on base percentage during the war .  Byrnes is at 19%, then Dickie Kerr—kicked out of baseball by Landis for playing exhibition games against the Black Sox—then a couple of Federal League players, Duke Kenworthy and Dutch Zwilling.   That’s eight.  Rounding out the bottom ten are Lou Fette, who won 20 games as a rookie for the Braves in 1937, and Gerald Young, who hit .321 in a half-season as a rookie for the Astros in 1987, stole 65 bases in 1988, then totally stopped hitting.

            The next 15:   Bob Lee (1960s reliever, dominant for a couple of years), Joe Connolly (outfielder who was part of the Miracle Braves in 1914), Alex Metzler (hit .319 for the White Sox in 1927), Steve Busby, Hod Eller, Tom Bradley, Herm McFarland, Rube Ellis, Ed Summers (one of the inventors of the knuckle ball), Paul Dean, Josh Devore, Lyman Bostock, Buster Adams, Luke Easter, and Herb Score.  

5)  What is the historic norm?

            The norm for all players, not including 19^th century pitchers, is 46%.  Most players are able to achieve a little less than one-half of what they reasonably might, in a full career.

6)  What is the norm for pitchers?

            The historic norm for pitchers is 45%; for non-pitchers, 47%.

7)  What is the norm for Hall of Famers?

            The norm for Hall of Famers, still excluding 19^th century pitchers, is 64%.   Hall of Fame players typically achieve about 64% of what they might have achieved without career interruptions of any kind.   I listed eleven Hall of Famers earlier—Beckley, Aaron, Kaline, Spahn, Mays, Aparicio, Frank Robinson, Don Sutton, Steve Carlton, Winfield and Musial.   Behind them are Babe Ruth (83%), Bid McPhee (82%), Gabby Hartnett, George Davis, Zack Wheat, Cap Anson (81%), Phil Niekro (80%), Sam Rice, Nap Lajoie (79%), Fred Clarke, Mel Ott, Carlton Fisk, Tris Speaker (78%) and George Brett.

8)  And the lowest percentages for Hall of Famers?

            Dizzy Dean, 31%.  Mostly—not entirely—a list of pitchers who had short but brilliant careers:

            It seems remarkable that there are that many Hall of Famers who achieved less than one-half of what they might have achieved, but I think it stands up to skepticism.   Look at Chick Hafey.  He had 1.466 hits in his career.  All we’re really saying is that, given a longer run, given that his eyesight doesn’t fail him in mid-career, given a chance to play earlier in his career, rather than starting out as a minor league pitcher, he might have had 3,000 career hits.   I don’t think that’s unreasonable.  

9)  Who does our system see as having the greatest upside potential of all time?

            We should acknowledge here that we’re departing into potentially controversial claims, with a speculative method that is clearly in need of more testing and refinement.  In other words, I’ll give you a list if you don’t take it too seriously, OK?   I don’t want to be hearing about this ten years from now.   I don’t want it in the Gold Mine next year, because I don’t want to be hearing about it twenty years from now. 

            In the view of our system, the potentially greatest player of all time, if you mark off his high spots and project from there, is Mickey Mantle.   These are what we would see as the 25 most promising careers of all time, and how they worked out.  I’ll mark with an asterisk those whose Peak Values were adjusted upward because of outstanding seasons as young players:

            Monte Ward is left off of a couple of other lists as a 19^th century pitcher; I decided to go ahead and include him here.   So I still owe you one:

            All of the potentially greatest players of all time are in the Hall of Fame except Barry Bonds and Joe Jackson, who are not eligible, and Will Clark.   Among players with very, very high peaks who ultimately didn’t (or haven’t yet) made the Hall of Fame:   Tim Raines, Don Mattingly, Dale Murphy, Mark McGwire, Pedro Guerrero.   Mostly 1980s players.   All 1980s players.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

3.  The Pitcher’s Area of Responsibility

            I have been trying to explain a little bit each week about how we figure Win Shares and Loss Shares.  In this effort, I’ll try to explain how we assign the “Area of Responsibility” to a pitcher.   The Area of Responsibility for a pitcher is not his Win Shares or his Loss Shares; it is the total of the two.  Game Shares. To assign Win Shares and Loss Shares to a pitcher, we first assign him Game Shares, and then divide those into Wins and Losses.   Wins and Losses, I’ll get to another time; this is just Area of Responsibility, or Game Shares.

            The first thing we have to determine here is the TEAM’S area of responsibility assigned to pitchers.   This is a number that, in modern baseball, averages about 185.  A team playing a 162-game schedule has 486 Game Shares in a season.  Of those 486, about 185 will be assigned to pitchers based on their pitching (not including their work as hitters.)    This number is higher now than it has been historically.   In the 1930s, when teams played 154 games a season, there were an average of 146 Game Shares assigned to pitchers.  In the 1950s, still with a 154-game schedule, there were an average of 161 Wins and Losses assigned to pitchers.   Now, with more strikeouts, walks and home runs, the share assigned to pitchers is larger. 

            We start by figuring this number for each team (which is actually the fielder’s responsibility for the success of the team)

            38.8 Times (Wins + Losses)

            Minus Strikeouts

            Minus (Walks + Hit Batsmen) * 27/14

            Minus Home Runs * 9

            Plus Errors * 27/14

            Plus 3 Times Double Plays

            Let us contrast the 2005 Minnesota Twins with the 2006 Chicago Cubs.   Both teams had 162 decisions; we multiply that by 38.8, and that gives us 6285.6 in both cases.  

            It’s an estimate of about how many batters the team would face in a season.   The actual number was 6072 for the Twins and 6366 for the Cubs, but for some reason that I forget right now, I decided to use a standard starting point for all teams—38.8 times games played.   From this we subtract strikeouts.  The Cubs had 1,250 strikeouts; the Twins had 965:

            From this we subtract walks and hit batsmen, times 27/14.  The Twins had 348 walks and 43 hit batsmen; the Cubs had 687 and 67.   That makes a “subtraction figure” of 754.07 for the Twins, and 1454.14 for the Cubs:

            You will no doubt want to know why we multiply these by 27/14.  I don’t exactly remember.  All of this was created maybe seven or eight years ago, and I don’t precisely remember all the ins and outs of it. . .an unfortunate reality, but that’s the way it is.   I know that those numbers came at the end of a question that began “How much more important is one walk than one strikeout?”   Obviously, one walk does more to create a run than one strikeout does to prevent a run, but how much more?   I don’t remember how I approached that problem, but I remember that the answer I came up with was 27/14.  A walk is almost, but not quite, twice as important as a strikeout. 

            From this we subtract Home Runs, times 9.  Same logic, whatever it was.   The Twins allowed 169 home runs; the Cubs 210:

            To this we ADD the number of errors committed by the team, times 27/14.   The Twins committed 102 errors; the Cubs committed 105.  

            And to this we add the team double plays, times 3.  The Twins turned 171 double plays; the Cubs, 122:

            This represents the volume of work NOT done by the pitchers; done by the fielders.   This number we divide by 54:

            This number represents the number of Game Shares to be assigned to the team’s FIELDERS.   The remainder will be assigned to the pitchers.   Each team has 243 Game Shares to be credited to pitchers and fielders combined.   The share of those going to PITCHERS, which is what we’re trying to get to, is:

            The 2006 Cubs have 201.15 Win Shares and Loss Shares assigned to pitchers; the 2005 Twins have 173.46. These numbers represent the highest and lowest figures in modern baseball.   The 201.15 figure for the 2006 Cubs is the highest figure of all time, except for the 1996 Detroit Tigers, who walked 784 batters and gave up 241 home runs.  

            Why 54?   I forget that, too. . .let me see if I can reconstruct that.   I think my logic was:

            1)  The figure above (3745.24 for the Twins, 2259.96). . .for a typical team in baseball history, that will work out to about 27 * games played.  

            2)  A team has 1.5 Win Shares and Loss Shares per game played to be assigned to pitching and defense.  

            3)  I wanted about two-thirds of those, in a typical case, to be assigned to pitching, and about one-third to fielding. 

            4)  That’s .50 Game Shares per game played to be assigned to fielding. 

            5)  To get to .50, we divide the 27 by 54.  

            OK, all of this, in a sense, just gets us to the starting line.   Now we need to move these “Game Shares” from the team to the individual pitcher.   Here are some pitchers from these two teams:

            We move these Games Shares from the pitching staff to the individual pitcher by this process.   First, we figure for each pitcher his share of the team’s Innings Pitched and Saves (actually, IP + 2 * Saves).   Santana, Radke, Silva, Zambrano, Maddux and Marshall have no saves:

            This we divide by the team total of Innings Pitched Plus (2 * Saves)—1552.33 for the Twins, 1526 for the Cubs:

            And multiply by the team’s Game Shares for pitchers, which we figured earlier:

            At this point in our analysis Santana is assigned responsibility for 25.89 Game Shares, Brad Radke 22.42, etc., but we’re not done.    At this point these numbers are apportioned to the pitcher only on the basis of innings pitched and Saves, without regard to the extent to which the pitcher has asserted authority over those innings by strikeouts, walks and big flies.   We have to adjust for the pitcher’s individual contribution to the “power numbers”.   For some reason—probably a lack of conceptual clarity about the overall product, but it’s always possible that I had a legitimate reason—I used a different formula here than I did earlier.   The formula here is:

            (Walks + Hit Batsmen + Strikeouts + 4 * Home Runs) / IP

            For Johan Santana this is:

            (45 + 1 + 238 + 4 * 22) / 231.67 =  1.606

            We figure this by the PITCHER and also the TEAM.   We multiply by the pitcher, and divide by the team, so that the Game Shares go up for power pitchers, and down for the finesse pitchers.   The figure for Minnesota, 2005, is:

            (348 + 43 + 965 + 4 * 169) / 1464.33  =   1.388

            Except that we don’t move people up and down on a straight-line basis, or we’d get extraordinarily high numbers for power pitchers.  Strikeouts, walks, and home runs aren’t the ONLY things the pitcher does; he also pitches to a lot of other batters, getting a neutral result.   So actually, we multiply by the pitcher’s rate, plus two (to represent the neutral results), and divide by the team rate, plus two—for Santana, 3.606/3.388 rather than 1.606/1.388.

            So Joe Nathan, pitching 70 innings, is assigned a larger number of Game Shares than Carlos Silva, pitching 188 and a third, and Ryan Dempster, pitching 75 innings, is assigned more Game Shares than Greg Maddux, pitching 136 and a third.   This happens for two reasons:   Leveraged innings, as reflected in Saves, and power pitching.  

            So Johan Santana, 16-7, will be assigned 28 Game Shares, and Carlos Zambrano, also 16-7, will also be assigned 28, although Zambrano actually has 0.90 more.   Santana’s record, in the next stage of this analysis, will come in at 23-5, and Zambrano will come in at 21-7, plus his hitting (and he hit six bombs that year.)  They are both outstanding pitchers, but we’re getting ahead of ourselves.     Our goal in this section was only to get to the numbers above—the total of Win Shares plus Loss Shares to be awarded to each pitcher. 

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

4.  Correcting the Record

            Just keeping the record clean. . .in the comments following last week’s blog there was a comment from a reader assuming that, in the Win Shares/Loss Shares system, a Designated Hitter gets a free pass on fielding because he doesn’t play in the field, and then there were comments from other readers also assuming that this was true.   It isn’t true.   The operating assumption regarding Designated Hitters’ fielding is that, as the team has a responsibility to play defense, every individual on the team shares in that responsibility, proportional to the number of outs that he makes as a hitter (since it is making outs that forces you to play defense.)   A Designated Hitter, making no defensive contribution to offset his defensive responsibility, is shown with a defensive won-lost record typically of 0-3 or 0-4. 

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - -

5.  Bruce Sutter Vs. Lee Smith

            Bruce Sutter was the All-Star reliever of the Cubs in the late 1970s, the Cy Young Award winner in 2006, and elected to the Hall of Fame in 2006.    Lee Smith was the All-Star reliever for the Cubs in the early 1980s, held the record for career Saves for a good many years, never won a Cy Young Award, and hasn’t been elected to the Hall of Fame, although he has his advocates.    

            I read somewhere recently that the Cubs were able to trade Sutter in the winter of 1980-1981 because they had Smith on hand.   I don’t think that’s strictly true.  In 1980 the Cubs still thought Lee Smith was going to be a starting pitcher.  The Cubs traded Sutter in the winter of 1980-1981 because that’s what they did with star players, in those years; they cashed them in while their value was high.   They traded Sutter for Ken Reitz and Leon Durham in the same way, a few years earlier, they had traded Ferguson Jenkins for Bill Madlock and Vic Harris.  

            Anyway, Sutter and Smith both really began their major league careers at age 23, although Smith pitched a few innings at age 22—the year he and Sutter were teammates for a while:

            Sutter had a much better year than Smith at age 23, and a much better year at age 24.  Sutter at age 24 was one of the greatest relievers of all time.   Signed as a non-drafted free agent in 1971, Sutter had been taught to throw a forkball in the minor leagues.   The forkball was basically a dead pitch at that time; nobody had thrown it with notable success in the majors for several years, and Sutter’s forkball was something special.   The thing would get about 12-15 feet in front of home plate, belt high, and then just dive.  Sometimes it would hit the ground.  You couldn’t take it, because if it was the fastball you were dead, and you couldn’t hit it because when you swung at it it wasn’t there.  

By the end of his age-24 season Lee Smith was the Cubs’ closer and a good one, but he was not—and would never be—at the level where Sutter was in 1977.

            At age 25 Smith had the better year:

            But at age 26 Sutter won the Cy Young Award:

            Sutter by age 26 had a career won-lost contribution of 62-12; Lee Smith, of 46-22.  Smith’s record was very, very good—but Bruce Sutter was 13 games ahead of him.

            Sutter by 1979 was calling his pitch the “split-fingered fastball”, which sounded cooler than a forkball although it amounted to about the same thing.   The “forkball” in the 1960s, early 1970s, was an old man’s pitch, almost a changeup.  The split-fingered FASTBALL had charisma.   It wasn’t exactly the same pitch; the splitter is jammed down into the fingers more, thrown harder, spins more, dives more.  The Splitter—now one of the most common pitches in baseball—came into baseball almost directly from the Bruce Sutter experience, although the wheel has come full circle, and there are now many pitchers who use the splitter as a change. 

            And also. . .I have told this story several times already, but it is integral to understanding Bruce Sutter.   In both 1977 and 1978, Bruce Sutter was fantastic the first half of the season, and in the All-Star game.   In both seasons, however, he was worked very hard in the first half of the season.   In those days there was no such thing as a “closer” role; there was a “relief ace”.   The relief ace pitched in the late innings of a close game—tied, up a run, down a run, didn’t matter.  The relief ace often pitched two and even three innings at a time, and a good relief ace might pitch 80 games and 140 innings in a season.  

            Sutter was a fantastic relief ace, but the job description was undisciplined, and Sutter was over-worked in both seasons, 1977 and 1978.  He broke down the second half of the season both years.    His manager, Herman Franks, announced before the 1979 season that, in the future, he would only use Sutter in “Save” situations.   Herman Franks was fired before the season was over, but Bruce Sutter won the Cy Young Award, and this idea, to hold back your relief ace for Save situations, caught on like a house afire.  Within three or four years after that, that was just the way it was done.   Relief Aces became Closers.   They only pitched in Save Situations, and, of course, in emergencies and to get some work, but basically, only in Save Situations.   By the early 1980s, the way that Bruce Sutter was used had become the way that Relief Aces were used.

            So Sutter had a huge impact on baseball history—first, by popularizing the “Splitter”, and second, by playing a critical role in the evolution of the modern bullpen.  Lee Smith is lacking this cachet.   From ages 27 to 29, however, both Smith and Sutter were highly effective closers, not quite at a Mariano Rivera level:

            Lee Smith, I think it is fair to say, was not exceptionally well liked.  He was a great big dude—6-foot-5 and powerfully built—and he did not wear an inviting smile.   He wasn’t reporter-friendly.  He lived off of high heat, and later, off a combination of high heat and pinpoint control.   He was much more like Goose Gossage than he was like Sutter.  He was a great pitcher and well respected, but not universally beloved.

            By age 29, however, Smith was catching up to Sutter.   Sutter’s career record was now 98-28, still a fabulous .777 winning percentage, but Smith was now 90-38, a .706 percentage, and only nine games behind him.  Smith was traded to the Red Sox in 1988, and gained ground in the competition at age 30, as Sutter had his first poor season:

            But Sutter stormed back in 1984 with 45 Saves and a 1.54 ERA in 123 innings:

            Sutter was sixth in the National League MVP voting in 1984, third in the Cy Young voting.   This made Sutter’s career record 124-40, still eight and a half games ahead of Smith at 115-48.  

            But that was Bruce Sutter’s last good year.    He was a free agent that winter, signed with the Braves, and in all candor he never earned his money with the Braves.  

            Smith, on the other hand, had a couple of good years with the Red Sox, went to the Cardinals and was outstanding with the Cardinals in 1990-1991.   By the end of the age 33 season, Smith’s career Win Shares and Loss Shares were 146-54 (.731); Sutter’s were 134-50 (.729). 

            Sutter was done.  He didn’t pitch at all in 1987:

            Lee Smith was the first man to get 400 career Saves, and held the career Saves record, with 478, until Trevor Hoffman passed him in 2006.   He had held the Saves record for about 15 years. 

            Sutter was done at age 35, but Smith was beginning to fade, too.  Being a closer is different than any other job in baseball, in that the margin is higher.  You can stay in the league as a .450 second baseman, a .450 center fielder.   A closer—if you’re not good, you’re not the closer.   Smith after age 34 was sporadically but not consistently good, although he continued to rack up Saves until age 37:

            Lee Smith aged much better than did Bruce Sutter; of that there is no doubt.   Smith made it to 1,000 games pitched in his career; Sutter petered out at 661.

Lee Smith’s final career Win Shares and Loss Shares are 192-85, a .692 percentage.   Bruce Sutter’s are 137-55, a .713 percentage.  These records include the men’s performance as hitters—2 and 4 career for Sutter, 1 and 3 for Smith.  Sutter had nine hits in his career; Smith had only 3, but one of them was a tater. 

            There may be three legitimate reasons why Bruce Sutter is in the Hall of Fame, and Lee Smith is not:

            1)  Sutter’s peak period was more dominant,

            2)  Sutter introduced baseball to the power of the Splitter, and

            3)  Sutter played a critical role in defining the usage patterns of the modern bullpen.

            Still, if it was up to me, I guess I would have voted for Lee Smith to go into the Hall of Fame before Bruce Sutter.   I think, in the final analysis, that Smith had the better career.