Saturday, November 6, 2010

Posnanski's Every Run Average

This past week, our favorite writer, Joe Posnanski had a great run.  He produced more than a handful or columns, including four that really stood out.  Here is Joe's Sparky Anderson eulogy, a ranking of the weakest World Series winners since WWII, a look into the new Bill James Handbook and Active Certain Hall of Famers, and this piece about baseball stats:
For some of us, it will always be fun to explore the new numbers, to try and separate what a pitcher does from what the defense does, to break down a hitter by what he contributes to winning and losing rather than by how many hits he gets per at-bat, to judge a player's defense beyond the occasional diving catch and by how often the ball boinks off his glove. As I've always said, there are a lot of ways to enjoy baseball. And there is no wrong way.
So with that as my preamble, I've been thinking about the most basic baseball stats -- and five very simple ways I would improve them. And let me say up front that my suggestions would not make the basic stats more advanced -- the opposite. They would make the basic stats even more basic, which I think would be good.
The entire post is worth the read, of course, but essentially Poz is showing how he would change five basic statistics.  They are:

1. Pitcher Wins
2. Sacrifices
3. RBIs
4. Earned Run Average
5. Middle Reliever Stats

Instead of boring you to smithereens with rambling tangents on each of these ideas(for once), let's just take a quick look at what Joe writes about ERA:
The funny thing about xFIP and how much some people despise it is that it's hardly a new effort. People have been trying to pinpoint and separate a pitchers individual ability from the team's defense for 100 years and more. That's the whole concept behind the earned run. The idea is that if a fielder makes an error, well, that's NOT THE PITCHER'S FAULT. And if it's not the pitcher's fault, then why should you count it against his statistics?
This, of course, leads to all sorts of ridiculousness. My mother really was right. For one thing, we don't add runs to the pitcher's "earned run" total when the fielder makes a spectacular run-saving catch. We don't add a home run to the pitcher's home runs allowed total if an outfielder leaps at the wall and brings a home run back. In those cases, the pitcher and the fielders are all in it together. So why discount the pitcher's ERA because of errors? Why mess with reality?
Second, you do know how unearned runs are figured, right? The official scorer goes through the inning and attempts to RECREATE the inning without the error. That is to say, a third baseman boots an easy ground ball with two outs, the official scorer makes the determination that the inning SHOULD be over. That's why every run scored after that error is called "unearned." Sometimes, believe me, this sort of recreation can go beyond absurdity. Let's say a guy is on second with one out. A ground ball is to short. The shortstop throws the ball away, and the batter goes to second. Well, at the moment, that's an unearned run because the guy would not have scored. But if the NEXT GUY hits a single, then it becomes an earned run because now it's assumed the guy would have scored. There are a lot of assumptions like that.
Third, of course, an error is in the eye of the beholder. It's a moving target. An error in Cincinnati isn't necessarily an error in Baltimore. Baseball stats should not change and shape-shift at the whim of some official scorer. Make it ERA -- Every Run Allowed -- after all, it's a pitcher's job to work around errors, to make the best of any and every situation.

2010 Leaders in Every Run Average
1. Josh Johnson, 2.50
2. Roy Halladay, 2.66
3. Adam Wainwright, 2.66
4. Clay Buchholz, 2.85
5. Felix Hernandez, 2.88
6. Tim Hudson, 2.91
7. Ubaldo Jimenez, 2.96
8. Roy Oswalt, 2.98
9. Johan Santana, 3.03
10. David Price, 3.06
Isn't the 2010 EvRA leaderboard fascinating?  (Let's abbreviate Posnanski's Every Run Average as EvRA.)  Josh Johnson lead the National League with a 2.30 ERA, while King Felix lead all of baseball with a 2.27 ERA.

I wondered how good of an EvRA, historically, is Johnson's 2.50?  How good is Santana's 3.03 EvRA compared to his career norms?

Mets fans need Johan's
shoulder to be A-Ok.
The latter question, here, is easier to answer.  With some quick math, Johan Santana's career EvRA is 3.34.  The Mets didn't put Johan Santana's efforts to good use, but he had another very good season in 2010.

Finding out how good Josh Johnson's 2.50 EvRA is historically appears to be much more difficult.  Finding seasons of the Best EvRA seems like it could be educational, fulfilling, but unfortunately way too time consuming.  I can't find a way to have the computers do the work on baseball-reference or fangraphs.  Manually checking all the leaderboards could ruin my marriage, career, and memories of a social life.  If you're good with computers and know how I could find a list of players by EvRA, please let me know.

We can have some fun just looking at ERA leaderboards.  Then, if there's time, we'll dabble in unofficial looks into EvRA.

First, we'll start with the leaderboard of Best ERA seasons in the history of the game.  Hmmmm.  Only Bob Gibson, Dwight Gooden, and Greg Maddux are in the Top 50 for single-season ERA.  In other words, 47 of the top 50 ERA seasons were before 1920.  That stinks.  After arguing for Joe Cronin and Lou Boudreau to @MLBNetwork last week, I'm ready to look into a lot of those guys, but figure no-one wants to read about them.*

Old Hoss Radbourn
research would be
time well spent.
*We may get posts about some or all of these guys down the road, anyway.  But, Please let me know if you wan to hear about Mordecai Brown, Orvall Overall, Dolf Luque, Old Hoss Radbourn, or Tim Keefe.

The deadball era ended around 1920, and the spitball was outlawed in 1921, so let's draw a line there to take a closer look at ERA while removing those 47 deadball stars and looking at the Top 25.

Best ERA since 1921
1.) Bob Gibson 1.12, 1968
2.) Dwight Gooden 1.53, 1985
3.) Greg Maddux 1.56, 1994
Gibby's 1.12 is proving hard to reach.
4.) Luis Tiant 1.60, 1968
5.) Greg Maddux 1.63, 1995
6.) Spud Chandler 1.64, 1943
7.) Dean Chance 1.65, 1964
8.) Carl Hubbell 1.66, 1933
9.) Nolan Ryan 1.69, 1981
10.) Sandy Koufax 1.73, 1966
11.) Sandy Koufax 1.74, 1964
Pedro Martinez 1.74, 2000
Ron Guidry 1.74, 1978
14.) Tom Seaver 1.76, 1971
15.) Mort Cooper 1.78, 1942
16.) Sam McDowell 1.81, 1968
Hal Newhouser 1.81, 1945
18.) Vida Blue 1.82, 1971
19.) Phil Niekro 1.87, 1967
Roger Clemens 1.87, 2005
21.) Joe Horlen 1.88, 1964
Sandy Koufax 1.88, 1963
23.) Kevin Brown 1.89, 1996
24.) Max Lanier 1.90, 1943
Pedro Martinez 1.90, 1997

Interesting how different it is when only considering time after they lowered the mound:

Best ERA since 1969
1.) Dwight Gooden 1.53, 1985
2.) Greg Maddux 1.56, 1994
3.) Greg Maddux 1.63, 1995
4.) Nolan Ryan 1.69, 1981
5.) Pedro Martinez 1.74, 2000
Ron Guidry 1.74, 1978
7.) Tom Seaver 1.76, 1971
8.) Vida Blue 1.82, 1971
9.) Roger Clemens 1.87, 2005
10.) Kevin Brown 1.89, 1996
11.) Pedro Martinez 1.90, 1997
12.) Luis Tiant 1.91, 1972
Wilbur Wood 1.91, 1971
14.) Gaylord Perry 1.92, 1972
15.) John Tudor 1.93, 1985
Roger Clemens 1.93, 1990
17.) Steve Carlton 1.98, 1972
18.) Gary Nolan 1.99, 1972
19.) Orel Hershiser 2.03, 1985
20.) Catfish Hunter 2.04, 1972
21.) Roger Clemens 2.05, 1997
Dave Righetti 2.05, 1981
23.) Jim Palmer 2.07, 1972
Pedro Martinez 2.07, 1999
25.) Bill Swift 2.08, 1992
Roger Nelson 2.08, 1972
Don Sutton 2.08, 1972
Tom Seaver 2.08, 1973
Jerry Koosman 2.08, 1968

Ignoring time before 1969 would cost us Gibson and Koufax.  They would have still pitched better than everyone else at the new mound level.  It's all about timing, with people's careers.

Love Maddux.  Pedro makes himself look really good.  So does Seaver, Carlton, Clemens...  It was nice nice to see Luis Tiant have a 1.91 ERA in 1972, therefore making both lists.  Don Sutton supporters can counter the "longevity" argument against Sutton's Hall of Fame worthiness can point to 1972 as the 25th best ERA since 1969.  Personal note: One of our favorite player's of all time, Dolf Luque, had a 1.93 ERA in 1923.

Zack entered RARE
company in 2009.
The Top 10 ERA seasons since the strike and steroid super explosion is an exclusive club, with its membership limited to five players.

Top 10 ERA since 1995
1.) Greg Maddux 1.63, 1995
2.) Pedro Martinez 1.74, 2000
3.) Roger Clemens 1.87, 2005
4.) Kevin Brown 1.89, 1996
5.) Pedro Martinez 1.90, 1997
6.) Roger Clemens 2.05, 1997
7.) Pedro Martinez 2.07, 1999
8.) Zack Greinke 2.16, 2009
9.) Greg Maddux 2.21, 1997
10.) Greg Maddux 2.22, 1998

Well, we're here, so why not?  Let's give this EvRA thing a shot.  This is the most Unofficial of Unofficial lists out there.  I'm sure that I've missed someone.  If you know of anyone, please alert us in the comment section.

Best EvRA since 1995
Maddux could do it all.
1.) Greg Maddux 1.68, 1995
2.) Pedro Martinez 1.82, 2000
3.) Roger Clemens 2.17, 2005
4.) Roger Clemens 2.22, 1997
5.) Greg Maddux 2.24, 1997
6.) Chris Carpenter 2.29, 2009
7.) Kevin Brown 2.32, 1996
8.) Pedro Martinez 2.36, 1999
9.) Pedro Martinez 2.42, 1997
10.) Jason Schmidt 2.43, 2003
11.) Josh Johnson 2.50, 2010
12.) Pedro Martinez 2.51, 2003
Zack Greinke 2.51, 2009
14.) Randy Johnson 2.54, 1997
Derek Lowe 2.54, 2002
16.) Al Leiter 2.56, 1998
17.) Tom Glavine 2.63, 1998
18.) Jake Peavy 2.65, 2004
19.) Adam Wainright 2.66, 2010
Roy Halladay 2.66, 2010
21.) Randy Johnson 2.67, 2001
Andy Pettitte 2.67, 2005
23.) Greg Maddux 2.69, 1998
24.) Kevin Brown 2.70, 1998
Randy Johnson 2.70, 2002
Jake Peavy 2.70, 2007
27.) Randy Johnson 2.73, 1995
28.) Cliff Lee 2.74, 2008
29.) Tim Lincecum 2.76, 2009
Johan Santana 2.76, 2004
31.) Pedro Martinez 2.80, 2002
32.) Johan Santana 2.84, 2008
33.) Randy Johnson 2.85, 1999
Clay Buchholz 2.85, 2010
Mark Prior 2.85, 2003
Tim Lincecum 2.85, 2008
37.) Kevin Brown 2.86, 2003
Pedro Martinez 2.86, 2005
39.) Felix Hernandez 2.88, 2010
40.) Clayton Kershaw 2.89, 2009
41.) Adam Wainright 2.90, 2009
42.) Tim Hudson 2.91, 2010
43.) Kevin Brown 2.92, 1997
44.) Jake Peavy 2.96, 2008
Hideo Nomo 2.96, 1995
Ubaldo Jimenez 2.96, 2010
47.) Jair Jurjens 2.97, 2009
Kevin Brown 2.97, 2000
49.) Roy Oswalt 2.98, 2010
JA Happ 2.98, 2009
Esteban Loaiza 2.98, 2003
52.) Johan Santana 2.99, 2005
53.) Dontrelle Willis 3.01, 2005

How about that JA Happ in 2009?  Who knew?!  Also, Al Leiter at 16th on this list is mighty impressive.  He had a great career, which he is currently extending as one of the better play-by-play men in the game.  Kevin Brown also makes another Hall of Fame case, if not for that whole Mitchell Report problem.  Maddux, of course, looks great at the top of the list, and Pedro shows up three times in the Top 9.  Amazing.

At least we found out that, according to our informal study, Josh Johnson had the 11th best EvRA season since the 1994 player's strike.  Comments below are more than encouraged; in fact, we insist.  Thanks again for stopping by.

No comments:

Post a Comment