Bullpen Report: September 20, 2014

by
September 21, 2014

The closer grid is updated below, but first up is your current rankings for 5×5 (W,SO,ERA,WHIP,SV), 6×6 (+HLD) and rate-stat (K/9, BB/9, HR/9, H/9, ERA) leagues using z-scores (rate stats adjusted by IP) for each category. It’s currently sorted by the standard 5×5 format. Scroll to the right for all of the individual z-scores in gray. I highlighted the relievers in green:

Here is yesterday morning’s review of Friday’s games including a grundle-grab reference, Edward Mujica probably sticking at closer, Drew Storen definitely sticking at closer, a White Sox closer swap rewarded with shakiness, and an update on Glen Glen Glen!

Here are tonight’s decisions and notes:

  • Mets/Braves: Jenrry Mejia earned his 27th save today against the Braves by continuing to elevate his elite career groundball rate. B.J. Upton flew out, but Simmons, Doumit and Pena all hit grounders (out, out and a single to center respectively). Of the 31 relievers with more than 10 saves, Mejia has the 16th best contact rate (better than McGee, Street, Papelbon, Rodney, Cishek and Perkins) and 8th best grounder rate. The 3.81 ERA/1.52 WHIP is uggly, but the .341 BABIP will certainly regress. The big complaint is the command, but he’s only 24 (the youngest of these 31 RP). I can see his swinging-strike rate and K% jumping about 2% next year, which would put him in Fernando Rodney surface stats territory (2.70 ERA/1.34 WHIP), but BABIP regression and a great groundball rate could keep the WHIP down.  I think he’s earned the role going into next year and the Mets probably won’t toy with him starting again based on their all-of-a-sudden depth assuming health.
  • Tigers/Royals: It took a Raul Ibanez appearance to ensure a Joe Nathan save. If it was earlier in the season, I would 100% urge you (and then some) to pick up Joakim Soria. Joe Nathan gave up 2 hits, but backing up the grossness, off of his 21 pitches, the Royals swung 16 times. They whiffed only once. Wade Davis pitched a scoreless 9th and struck out two. He’s been unbelievable this year. In standard 5×5 leagues, where he’s not racking up saves for you, he is still a top 45 pitcher based on the lucky wins total and sub 0.95 ERA and WHIP. In 6×6 leagues with holds, Wade Davis is a top 10 pitcher. He’s #11 in if you’re using all rate stats (K/9, BB/9, HR/9, H/9 and ERA).
  • Blue Jays/Yankees:I know there was some excitement about Aaron Sanchez consuming high leverage situations, but it was Casey Janssen again for the 2-out 3-run save after a Jeter RBI double off of Brandon Morrow.
  • Brewers/Pirates: Andrew McCutchen got beaned by pitches more tonight than I had alcoholic drinks at the Twins game tonight – it was too cold, so I transitioned to coffee. Francisco Rodriguez earned (clean) save number 43 tonight. #2008? That’s a hashtag by the way and not a number/pound sign.
  • Phillies/Athletics: Ken Giles finally gets a save chance. Ken Giles averaged 100+ MPH on his fastball. Ken Giles K. Ken Giles Save.

Regarding the grid, I looked up the last three Steve Cishek saves and it basically went Mike Dunn the earliest, followed by A.J. Ramos and then Chris Hatcher prior to Cishek so i’ll keep the MIA line item as such with Dunn omitted.

I will not change the Padres line item nor the D-backs at this time. D-backs situation may be one of situation. Ziegler, Perez and Marshall all make sense. I’ll keep Ziegler depicted in the matrix even though he’s not hurt.

Closer Grid:

Closer First Second DL/Minors
Arizona Addison Reed Oliver Perez Evan Marshall Brad Ziegler
Atlanta Craig Kimbrel Jordan Walden David Carpenter Shae Simmons
Baltimore Zach Britton Tommy Hunter Andrew Miller
Boston Edward Mujica Koji Uehara Junichi Tazawa
CHI (NL) Hector Rondon Neil Ramirez Kyuji Fujikawa
CHI (AL) Jacob Petricka Zach Putnam Matt Lindstom
Cincy Aroldis Chapman Sam LeCure Manny Parra Sean Marshall
Cleveland Cody Allen Bryan Shaw Scott Atchison
Colorado LaTroy Hawkins Adam Ottavino Rex Brothers
Detroit Joe Nathan Joakim Soria Joba Chamberlain
Houston Chad Qualls Tony Sipp Josh Fields
KC Greg Holland Wade Davis Jason Frasor
LAA Huston Street Joe Smith Jason Grilli
LAD Kenley Jansen Brian Wilson Brandon League
Miami Steve Cishek Chris Hatcher A.J. Ramos
Milwaukee Francisco Rodriguez Jonathan Broxton Zach Duke Jim Henderson
Minnesota Jared Burton Michael Tonkin Glen Perkins
NY (NL) Jenrry Mejia Jeurys Familia Buddy Carlyle Bobby Parnell
NY (AL) David Robertson Dellin Betances Adam Warren
Oakland Sean Doolittle Luke Gregerson Eric O’Flaherty
Philly Ken Giles Jake Diekman Jonathan Papelbon (SUSP) Mike Adams
Pittsburgh Mark Melancon Tony Watson Jared Hughes
St. Louis Trevor Rosenthal Pat Neshek Seth Maness Jason Motte
SD Kevin Quackenbush Dale Thayer Nick Vincent Joaquin Benoit
SF Santiago Casilla Sergio Romo Jeremy Affeldt
Seattle Fernando Rodney Danny Farquhar Dominic Leone
TB Jake McGee Brad Boxberger Grant Balfour
Texas Neftali Feliz Neal Cotts Shawn Tolleson
Toronto Casey Janssen Aaron Sanchez Brett Cecil
Wash. Drew Storen Tyler Clippard Rafael Soriano

[Green light, yellow light, red light: the colors represent the volatility of the bullpen order.]





 

Daniel Schwartz contributes for RotoGraphs when he's not selling industry leading thermal packaging. You can follow him on twitter @RotoBanter

6 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
buddyglass
9 years ago

Ranking SPs and RPs together is interesting. I struggle with how to value closers relative to starters. For the past few years in my H2H league I’ve done a linear regression of “wins in a category” vs. “regular season stats in that category” in an attempt to come up with a formula for each category that would allow me to determine how many “points” a given player is likely to get me. So maybe a single win is worth 0.05 “points”, a strikeout worth 0.005, etc.

I scale the point totals for relievers’ ERA and WHIP since they pitch fewer innings over the course of the year and, as such, have a reduced influence on my team’s overall ERA/WHIP.

The formula works pretty well for comparing starters vs. starters and relievers vs. relievers, but when I compare starters to relievers the relievers end up way, way overvalued.

The reason: a single top-shelf closer can knock out about 30-40% of the saves you hope to get over the course of an entire season, so that one guy gets “credit” for 30-40% of all the points that will be awarded for the saves category. Which makes him look really valuable compared to a starter, who is going to be one of 6-7 similar players on your team. So far I haven’t been able to think of a non-arbitrary way to “scale down” closers’ scores so that they aren’t ridiculously overvalued relative to starters.

Any ideas?

The league has 9 general-purpose pitching slots, 3 bench slots and an 80 IP/week cap that can be exceeded on the last day of a scoring period.

0
Daniel Schwartz
9 years ago
Reply to  buddyglass

Hardcore. These rankings would be relevant for you since you have 9 general P slots. For leagues that break up sp and RP would have a small Adjustment where I chose a replacement value for each pitcher type and then adjust all values by those replacement values. Read FAVRz by zach sanders. This approach is summing the categorical values relative to all others. If you know your historical results or your format matches standard formats, just check out SGP approach to value (standard gain points)…this is more like what you want to do ie how to value certain categories.

0
buddyglass
9 years ago
Reply to  Daniel Schwartz

Thanks. Will look into those. I toyed around with the idea of doing an “above replacement” adjustment where I subtract out the point total that a replacement level guy would generate at that position. It’s hard to determine what a “replacement level” closer is, though, since guys move in and out of the job so frequently.

0
Daniel Schwartz
9 years ago
Reply to  buddyglass

It’s relatively easy with a wee bit of subjectivity. Order by zsum and then in that format (9 general P and then maybe 1-2 more P per team’s bench in a 10 team league let’s say is 150 P. So go to the 150th P and subtract/add that zsum by itself so it’s 0 and he’s the replacement. Subtract/add that same value from/to every other pitcher. You have to do this prior to integration with hitters. You would have to do those for both rp and sp separately if you had both sp and rp slots vs general P slots.

0
mattmaison
9 years ago
Reply to  buddyglass

Why not use ottoneu points system? (based on linear weights)

http://ottoneu.fangraphs.com/support

0