29 February 2012

AL East FIP and Pitching xWARs by Slot

There is a series of articles by Jack Sackman that you can find here.  It is an idea I found interesting an often use when I describe pitchers as a certain type of slot pitcher.  I think in common use a person referring to a guy as a one slot pitcher is more or less actually saying that the guy is a one slot pitcher on a first division team.  In other words, an ace on one of the ten best teams in baseball.  In this series of posts, I plan on going through each division and describing what each slot means and how that relates to teams.
AL East | Central | West
NL East | Central | West
For each team, slots were broken down in 32 starts per slot (for a total of 160 games).  Pitchers were ordered by FIP, lowest to highest.  They were then broken up into slots.  Their FIPs were weighted by inning allotment and a weighted average was calculated for each slot.

An example:
Chris Tillman had an FIP of 3.99 over 11 starts, Zach Britton had an FIP of 4.00 over 28 starts, and Jeremy Guthrie had an FIP of 4.51 over 32 starts.  For the purpose of this study, Tillman's 11 starts all counted for the Orioles slot 1 along with 21 of Britton's 28 starts.  The remaining 7 starts were put into the slot 2 position with 25 of Jeremy Guthrie's starts.  When a pitcher's starts are split between multiple slots, it is assumed that their FIP is equal for every inning thrown and that their IP are equal for every start.

AL East


As you can see above, the Orioles were amazingly bad in comparison to the rest of the AL East.  In terms of FIP, the number 1 slot was worse than all other number 1s in the AL East, worse than the Red Sox, Rays, and Yankees number 2s, and worse than the Rays number 3.

Below is the AL East FIP by Slot Table.

1 2 3 4 5
Orioles 4.00 4.38 4.59 5.01 6.64
Red Sox 3.55 3.78 4.36 4.82 5.33
Rays 3.22 3.41 3.88 4.40 4.90
Yankees 2.88 3.89 4.04 4.33 4.82
Blue Jays 3.66 4.11 4.29 4.82 5.39


The following graph is simply a predicted fWAR value using only FIP and IP as described in this post.  It serves as an approximation of fWAR worth.

Again, all other number 1s generated more worth than the Orioles' number 1, all other number 2s, and three of the number 3s.  In fact, the Orioles number 1 slot was neck and neck with the Rays number 4 slot.  All of this does not even address the wretched state of the team's fifth starters.  If you add in two of Rick VandenHurk's starts, the value would near -5 xWAR.

A basic concept when it comes to WAR is the R.  The replacement level production should be roughly the quality of pitching that is freely available at the AAA level.  In part, you would expect the worst starter on an MLB club to be below replacement level for many clubs because you are likely to be replaced only after you show that you cannot match what a AAA may accomplish.  This idea likely explain why the Red Sox, Blue Jays, and Orioles all have negative values to show for the fifth slot.  However, the Orioles are five times worse than the next closest team's production from the fifth slot.  This speaks to either a paucity of talent in the high minors or a willingness to stick with a pitcher long after they show they are not capable of pitching at an MLB level.

Explaining Duquette Starting Pitcher Fancy
For the Orioles, you can blame the paucity of talent on giving innings to pitchers like Mitch Atkins and Jo-Jo Reyes (we will pretend that Rick VandenHurk's starts never happened...we are better off that way).  Brian Matusz' are more about giving a pitcher chance after chance to show he can do something when it is obvious that he cannot.  With this understanding, you can now see why Dan Duquette has authorized the use of a 40 man roster clown car for the team this year.  The Orioles have over a dozen pitchers on the 40 man roster and several more off the roster who have started in the Majors or are capable of starting in the Majors.  The idea is that by deluging the roster with starting pitching, you could hypothetically win 3-5 more games.

Below is the xWAR AL East  by Slot Table

1 2 3 4 5
Orioles 2.50 2.14 1.62 0.17 -4.04
Red Sox 4.01 3.60 2.00 0.83 -0.51
Rays 5.09 5.18 3.29 2.26 0.99
Yankees 6.27 3.19 2.80 1.90 0.86
Blue Jays 3.75 2.86 2.60 0.89 -0.76

Baltimore Orioles by the Numbers
Slot 1 (166.2 IP, 4.00 FIP)
Chris Tillman 13g 62IP 3.99 FIP
Zach Britton 19g 104.2IP 4.00 FIP
Slot 2 (196 IP, 4.38)
Zach Britton 9g 49.2IP 4.00 FIP
Jeremy Guthrie 23g 146.1IP 4.51 FIP
Slot 3 (195 IP, 4.59 FIP)
Jeremy Guthrie 9g 57.1IP 4.51 FIP
Alfredo Simon 16g 94.1IP 4.59 FIP
Tommy Hunter 7g 43.1IP 4.71 FIP
Slot 4 (169.2 IP, 5.01 FIP)
Tommy Hunter 4g 25IP 4.71 FIP
Brad Bergesen 12g 62.1IP 4.84 FIP
Chris Jakubauskus 6g 27.1IP 5.00 FIP
Jake Arrieta 10g 54IP 5.34 FIP
Slot 5 (149 IP, 6.65 FIP)
Jake Arrieta 12g, 65IP, 5.34 FIP
Jo-Jo Reyes 5g, 23.2IP, 7.12 FIP
Brian Matusz 12g, 49.2IP, 7.66 FIP
Mitch Atkins 3g, 10.2IP, 8.93 FIP

28 February 2012

Eyes on Spring Training: 2B, Andino and Antonelli

Eddie Stanky
With Brian Roberts out for most of 2011, Robert Andino was unexpectedly elevated from the utility infielder spot to take an everyday line-up spot (mostly at second, though he filled in at third and short on occasion). Overall, he was adequate if unspectacular, posting a 1.8 fWAR season while upping his walk rate to 8%. Coming into 2012 it seems like Andino would be the default option at second, but with Matt Antonelli being signed to a major league contract, it's possible that that isn't the case.

I certainly like the on-base skills Antonelli showed at the minor league level, but to expect him to get his first real extended amount of playing time in the majors for the first time at age 27 and out-produce Andino seems... doubtful. Here is the whole list of second and third base-man to get 300 plate appearances in their first season at ages 27-28 who beat Andino's 2.0 rWAR from last season:
Ron Theobald (1971), 452 PA, 2.4 rWAR
Spider Jorgensen (1947), 506 PA, 2.4 rWAR
Coco Laboy (1969), 616 PA, 2.3 rWAR
Akinori Iwamura (2007), 559 PA, 2.2 rWAR
Eddie Stanky (1943), 616 PA, 2.2 rWAR
That's it - five guys. And one could argue that Iwamura shouldn't count, given his time in Japan. It's certainly possible that Andino won't repeat his 2011, but it seems more likely that he'll do so than that Antonelli is good enough to do what very few other players in baseball history have been able to.

In his time at Triple-A, Antonelli hit .237/.347/.369. That doesn't inspire a great deal of confidence, though part of that was a low BABIP (~.275). Maybe that will carry forward, but in any case, he'll need to continue taking walks (13% rate at Triple-A) to produce enough with the bat to make up for what sounds like a below average glove. How many of those older first-year infielders posted even a 10% walk rate? Six:
Eddie Stanky (1943) - 14.9%
Spider Jorgensen (1947) - 11.4%
Ike Rockenfield (1905) - 11.4%
Al Rubeling (1940) - 11.2%
Spook Jacobs (1954) - 10.4%
Akinori Iwamura (2007) - 10.4%
The thing is, though, these types of player don't tend to get chances at all. There are only 23 second/third-basemen who had their first season at ages 27-28 and got any significant playing-time (because players who are good enough to start in the majors are usually good enough to show up earlier). Just getting 300+ plate appearances would be something for Antonelli - if he does, it's probably a clue that he's playing pretty well (or there have been some injuries).

It's good that the Orioles were willing to take a chance on a potentially undervalued player. Relying on him to reach his ceiling as part of Plan A would be slightly less good. We'll see how it plays out.

27 February 2012

Arrivals and Departures: February 27, 2012

Sorry, this has been overdue for a while.  Here is an update of the Orioles' 40 man roster and options.
  • January 24, 2012 - Jason Hammel and Matt Lindstrom were added to the team when they came over from the Colorado Rockies for Jeremy Guthrie.  Clay Rapada was DFA'd.  He later signed a minor league deal with the Yankees.  Neither Hammel or Lindstrom have options for use.
  • February 6, 2012 - Wilson Betemit was signed and Rick VandenHurk was designated for release.  VandenHurk is now a Blue Jay on a minor league split deal.  Options cannot be used for Betemit.
  • February 10, 2012 - Luis Ayala was signed on a 1+1 deal.  He has no available options.  Matt Angle was DFA'd, passed by every AL team, and was claimed by the LA Dodgers.

An option (optional assignment) allows a club to move a player on its 40-man roster to and from the minor-leagues without exposing him to the other 29 teams.

After 4 or 5 years as a professional, a player must be added to his club's 40-man roster or exposed to the 29 other clubs in the Rule 5 draft. (A club has 5 years to evaluate a player who signs his first pro contract at 18 years old or younger, but only 4 years to decide on a player who signs at age 19.) For purposes of calculating years as a pro, the counting begins the day a player signs his first pro contract, not the season he begins to play.

When a player is added to the 40-man roster, his club has three options, or three separate seasons during which the club may to move him to and from the minor leagues without exposing him to other clubs. A player on the 40-man roster playing in the minors is on optional assignment, and within an option season, there is no limit on the number of times a club may demote and recall a player. However, a player optioned to the minor leagues may not be recalled for at least 10 days, unless the club places a Major League player on the disabled list during the 10-day window.

After three options are exhausted, the player is out of options. Beginning with the next season, he must clear waivers before he may be sent to the minors again. See Waivers. Additionally, a player with 5 years of Major League service may not be sent to the minor leagues on an optional assignment without his consent.

Counting option years
- If a player is not sent to the minors during a year, an option is not used.
- If a player is on the 40-man roster in spring training but optioned to the minors before the season begins, an option is used.
- If a player's optional assignment(s) to the minors total less than 20 days in one season, an option is not used.
- A player may be eligible for a fourth option year if he has been optioned in three seasons but does not yet have five full seasons of professional experience. A full season is defined as being on an active pro roster for at least 90 days in a season. (If a player is put on the disabled list after earning 60 or more days of service in a single season, his time on the DL is counted.) The 90-day requirement means short-season leagues (New-York Penn, Northwest, Pioneer, Appalachian, Gulf Coast, Arizona Rookie, Dominican and Venezuelan Summer Leagues) do not count as full seasons for the purposes of determining eligibility for a fourth option.

The following is the 40 man roster as it stands.  Again, this list is largely informed by the work mentioned above.

Adams, Ryan
Options Remaining: 2/3
Reason: Adams was brought up to replace Brian Roberts in May 2011. He was optioned after only getting a handful of starts.

Andino, Robert
Options Remaining: 0/3
Reason: The Marlins purchased his contract on 9/2/2005. He was optioned for the first time on 3/25/2006. He was optioned for a second time on 3/23/2007. He was optioned for a third and final time on 5/25/2008. Andino was sent outright to AAA Norfolk at the end of Spring Training 2010 with the acquisition of Julio Lugo from the St. Louis Cardinals. He was later added back to the 40 man roster in September 2010 and remained on the roster over the winter.

Antonelli, Matt
Options Remaining: 1/3
Reason: Contract purchased 9/1/2008 and optioned on 3/23/2009 and 3/28/2010.

Arrieta, Jake
Options Remaining: 3/3
Reason: Arrieta's contract was purchased on June 11th 2010 to make his major debut against the New York Yankees.
Ayala, Luis
Options Remaining: 0/3
Reason: Ayala's last option was used in 2009.

Bell, Josh
Options Remaining: 1/3
Reason: Bell had his contract purchased in November 2009. He was optioned for the first time following Spring Training 2010. He was optioned for the second time following Spring Training 2011.

Bergesen, Brad
Options Remaining: 0/3
Reason: The Orioles purchased his contract on 11/18/2008 to protect him from the Rule 5 draft. He was optioned to the minors for the first time on 3/21/2009. Bergesen was sent to the minors on 4/20/2010 but was recalled on 5/1/2010. Bergesen was optioned to the minors again on 6/14/2010 and remained in the minors for more than 20 days throughout the season. Bergesen was optioned to the minors for four days in 2011, from 4/9 to 4/13 before being recalled for an injured starter. He was later optioned on 5/29/11.

Berken, Jason
Options Remaining: 2/3
Reason: The Orioles purchased his contract on May 26th 2009. Berken was optioned to Norfolk on 5/26/2011 to help him work on his command.

Betemit, Wilson
Options Remaining: 0/3
Reason: Betemit long ago had his options exhausted.

Britton, Zach
Options Remaining: 3/3
Reason: Britton was added to the 40 man roster in November 2010. Britton was optioned on 3/29/2011 but was recalled on 4/3/2011 so an option year was not used.

Bundy, Dylan
Options Remaining: 4/4
Reason: Bundy was added to the 40 man roster upon signing in August 2011.

Chavez, Endy
Options Remaining: 0/3
Reason: Chavez no longer qualifies for options.

Chen, Wei-Yin
Options Remaining: 3/3
Reason: Signed by Baltimore as a Free Agent on 1/10/2012.

Davis, Chris
Options Remaining: 0/3
Reason: Contract purchased on 6/26/2008 and optioned on 7/6/2009, 4/23/2010, and 3/29/2011.

Drake, Oliver
Options Remaining: 3/3
Reason: Drake was added to the 40 man roster in November 2011. 

Eveland, Dana
Options Remaining: 0/3
Reason: Eveland was acquired via trade in December 2011 and is without options.

Gregg, Kevin
Options Remaining: 0/3
Reason: Gregg is in the last year of his free agent contract.

Flaherty, Ryan
Options Remaining: 3/3 (Rule 5 draftee, cannot use options in 2012)
Reason: Flaherty was drafted in Rule 5 draft in December 2011.

Hammel, Jason
Options Remaining: 0/3
Reason: Last option used with Tampa.

Hardy, J.J.
Options Remaining: 0/3
Reason: Hardy has signed an extension and no longer qualifies for options.

Hunter, Tommy
Options Remaining: 0/3
Reason: Rangers purchased his contract in 2008 and was optioned in each season subsequently.

Johnson, Jim
Options Remaining: 0/3
Reason: The Orioles purchased Johnson's contract on 11/18/2005. The Orioles optioned Johnson to the minors for the first time on 3/16/2006. On 3/12/2007, the Orioles optioned Johnson to the minors for a second time. The Orioles optioned Johnson to the minors in March of 2008, but he spent less than 20 days in the minors so his optional assignment is withdrawn. Johnson was optioned to Norfolk on 5/1/2010 to make room for the returning Brad Bergesen on the major league roster. Johnson was recalled on 5/28/2010 and placed on the major league DL, unfortunately, the final option was used.

Jones, Adam
Options Remaining: 1/3
Reason: Contract was purchased on 7/14/2006 and optioned on 8/22/2006 and 4/1/2007.

Lindstrom, Matt
Options Remaining: 0/3
Reason: Houston used Lindstrom's last option.

Mahoney, Joe
Options Remaining: 2/3
Reason: Mahoney was added to the 40 man roster in November 2010. Mahoney was optioned to the minors for the first time during Spring Training 2011.

Markakis, Nick
Options Remaining: 0/3
Reason: No longer qualifies.

Matusz, Brian
Options Remaining: 2/4
Reason: Matusz signed a MLB out of the draft with options used on 3/14/2009 and 6/30/2011.

Miller, Jai
Options Remaining: 0/3
Reason: Contract purchased on 11/20/2007 and optioned on 3/10/2008, 3/13/2009, and 4/8/2010.

O'Day, Darren
Options Remaining: 1/3
Reason: Contract was purchased on 3/29/2008.  Options were used on 5/13/2008 and on 7/14/2011.

Patton, Troy
Options Remaining: 0/3
Reason: On 8/25/2007, Patton had his contract purchased by the Astros after completing his fourth season in the minors. Patton remained with the Astros throughout September of 2007. Patton was optioned in Spring Training 2009 and 2010. Patton was optioned for a final time during Spring Training 2011.

Phillips, Zach
Options Remaining: 1/3
Reason: Contract was purchased on 11/19/2009 with options executed on 3/17/2010 and 3/12/2011.

Reimold, Nolan
Options Remaining: 0/3
Reason: Reimold had his contract purchased on 11/18/2008 to protect him from the Rule 5 draft. He was optioned for the first time during Spring Training 2009. After opening up with a horrendous beginning to his 2010 season, Reimold was sent to Norfolk to work out some of his issues in May and has remained in Norfolk. Reimold was optioned for a final time at the end of ST 2011.

Reynolds, Mark
Options Remaining: 0/3
Reason: No longer qualifies for options.

Roberts, Brian
Options Remaining: 0/3
Reason: No longer qualifies for options.

Simon, Alfredo
Options Remaining: 0/3
Story: Had his contract purchased by the Phillies on 11/19/2003. He was optioned for the first time on 3/13/2004 by the Phillies. Upon being traded to the Giants during the 2004 season, he was optioned for a 2nd time on 3/14/2005. He was optioned for a 3rd and final time on 3/13/2006. He was sent outright to the minors on 7/29/2006.

Strop, Pedro
Options Remaining: 0/3
Reason: Last options used on 5/4/2011.

Teagarden, Taylor
Options Remaining: 0/3
Reason: Contract purchased 7/18/2008 and optioned on 7/21/2008, 4/27/2010, and 3/29/2011.

Tillman, Chris
Options Remaining: 1/3
Reason: The Orioles purchased his contract on 7/29/2009 to make his major league debut against Kansas City. Tillman was optioned to the minors for the first time during Spring Training 2010. Tillman was optioned to the minors on 5/29/2011.

Wada, Tsuyoshi
Options Remaining: 3/3
Reason: Wada was signed as a free agent on 12/14/2011.

Wieters, Matt
Options Remaining: 3/3
Reason: Contract purchased on 5/29/2009.

Four Options Remaining:

Three Options Remaining:
J. Arrieta, Z. Britton, O. Drake, M. Wieters

Two Options Remaining:
R. Adams, M. Angle, B. Matusz, J. Mahoney 

One Option Remaining:  
J. Bell, A. Jones, D. O'Day, Z. Phillips, P. Strop, C. Tillman

Zero Options Remaining:
R. Andino, L. Ayala, B. Bergesen, E. Chavez, W. Chen, C. Davis, D. Eveland, K. Gregg, J. Hammel, R. Flaherty, J. Hardy, T. Hunter, J. Johnson, M. Lindstrom, N. Markakis, J. Miller, T. Patton, N. Reimold, M. Reynolds, B. Roberts, A. Simon, T. Teagarden, T. Wada

We will be following the option year process with each player throughout the season.

23 February 2012

With Braun, MLB and Media Make Mistakes

Ryan Braun won his appeal on a 2-1 vote.  The only vote going against him was made by MLB.  MLBPA and the important one, the arbiter, ruled for the appeal.  It has been explained that the appeal was based on an improper chain of custody.

Yahoo Sports Writer Jeff Passan tweeted that Braun escaped due to using a loophole.  This is lazy reporting and should really be an embarrassment.  A loop hole is an ambiguity in the rule of law that can be exploited to avoid the intent of that law.  In no way could this be considered a loophole.  Braun was in no way responsible for the broken chain of custody.  In order for a loophole argument to be made, Braun would have held some responsibility.  Passan decided to go further and second a statement that Braun won on a technicality as if there was no reason why a chain of custody exists.

Second, his tweet acts like a chain of custody is not important.  I think it is rather safe to say that if a piece of evidence could effectively damage someone's reputation and affect his future earning potential on the field and through sponsorships off the field, the integrity of that biological sample is quite important.  A chain of custody is a guarantee of the identity and integrity of the sample.  This means that not only was the sample not tampered with, but that the sample is handled in a way that would not affect the protocols for testing the sample.

I typically like the work Passan does, but I find his disregard for the scientific integrity of a sample to be rather distasteful.

There are reasons why these rules exist.

The real story here is why was MLB so hell bent to use bad evidence to go after Ryan Braun in what was surely to be a rather public case.

22 February 2012

Predicting WAR from FIP and IP

This is more or less a somewhat uninteresting way to determine fWAR in a messy shorthand way.  I ran a regression with FIP and IP as my variables and WAR as the y.  For those who do not often pay attention here, the following definition might be useful:
FIP - Fielding Independent Pitching is a metric that tries to isolate aspects of a stat sheet that the pitcher directly affects. It is calculated using this formula:

        FIP = coefficient + (HR*13 + (BB + HBP - IBB)*3 - K*2) / IP.
The objective of this post is merely to show how FIP and IP relate to fWAR.  I will be using this messy number from time to time over the next couple weeks.  It is also important to recognize that the numbers are league dependent, so an AL and NL equation is needed.

Again...this is just a fast an easy way for me to generate rough fWARs.  You should use the real thing.