Lookout Landing: An SB Nation Community

Navigation: Jump to content areas:


Sports blogs for fans, by fans.
New Blog: RSL Soapbox for Real Salt Lake Fans!

Best Pitching Units by tRA, 2007

Now that you are all up to speed on tRA (see related link) (and if you're not, feel free to keep asking questions, no matter how dumb you think they might be.) it's time for some more leaderboards. You've already seen some of the individual boards, so this time, let's look at pitching units (defined as starters or relievers) instead.

First off, some league wide numbers to provide that juicy, vine-ripened, context.

EntryO-xOxOxRtRA
NL - SP 44 44536 8315 5.04
NL - RP -21 25013 4165 4.50
AL - SP 28 39826 7323 4.96
AL - RP -42 20584 3522 4.62
Table 1: 2007 tRA Figures

O-xO = Outs - eXpected Outs: A positive number means more outs were record than the underlying numbers would suggest (good luck). A negative number means the opposite.

You see the expected difference between relievers and starters here; a smaller gap in the AL than in the NL last year, but not a significant difference. On to exploring some of the best and worst units from 2007.

BEST ROTATIONS IN 2007

  1. San Diego Padres, 4.35: Jake Peavy was awesome as you all know but behind him, Greg Maddux had a fabulous year at preventing baserunners from getting on via the walk at just and he kept the ball on the ground often enough to post an xFIP just above 4. Chris Young was a watered down version of Jake Peavy with loads of Ks, but more walks and yet despite a disgusting sub-30% GB rate, a lower overall HR rate than Peavy and his 44% GB rate. It was incredibly fluky, but tRA doesn't care (tRA* would).
  2. Anaheim Angels, 4.54: The Angels weren't so much lead by greatness as deep in above average as John Lackey, Kelvim Escobar and Jered Weaver all had tRAs just above 4 (again with HR/FB playing a big role in this) which helped cancel out the crappiness of Ervin Santana, Bartolo Colon and Joe Saunders.
  3. Los Angeles Dodgers, 4.55: The Dodgers had just a ridiculous pitching staff top to bottom fronted by three main prongs. Brad Penny who posted a 3.03 ERA managed it thanks largely to a (you guessed it) flukey low HR/FB ratio of just 4.7%. Make no mistake though, Penny was good regardless with a 4.38 xFIP. Derek Lowe saw a staggering 16.5% of FB leave for HRs, but thanks to his incredible 65% GB rate, that still translated to under 1 HR/G. Chad Billingsley added his own unique aspects with the, by far, highest strikeout and walk rates and lowest GB rate of the trio, though still a healthy 41%.
  4. Cleveland Indians 4.55: It certainly helps to have a pair of 200+ IP horses up top that are both awesome. Dare I say a 1-2 punch even? Sabathia logged 241 IP of 8.3/1.5/0.8 pitching with a batted profile to back it up. Carmona went the Derek Lowe route except without the extra homeruns and reaped the benefits. Behind those two, Paul Byrd and Jake Westbrook were mediocre but not crippling.
  5. Boston Red Sox, 4.58: Josh Beckett was legitimately awesome with a 3.27 tRA, but other than that, nobody on the Sox was outstanding, they just avoided having a boat anchor weigh them down with Matsuzaka, Wakefield, Schilling and Tavaras hovering in the range of 4.5-5.0 tRAs.

Lost a one-game playoff for the playoffs; made playoffs; finished fourth in tough division; made playoffs, made playoffs. Combined record: 457-354 (56.4%). Having a good rotation is a huge boon for a team toward playing games in October.

WORST ROTATIONS IN 2007

  1. Washington Nationals, 6.08: Quick! Name the pitchers that started more than 10 games for the Nationals last year! How many did you get? They were Matt Chico, Jay Bergmann, Mike Bacsik, Shawn Hill, Tim Redding, Jason Simontacchi and Joel Hanrahan. Matt Chico is the only one who made more than 30 starts and he posted a 6.26 tRA. Jay Bergmann and Mike Bacsik are the only other two to make more than 20 and while Bergmann was decent (4.89), Bacsik is now back in AAA.
  2. Florida Marlins, 5.94: Dontrelle Willis. Scott Olsen. Sergio Mitre. Byung-Hyun Kim. Rick VandenHurk. That was the Marlins rotation.
  3. Texas Rangers, 5.73: Perhaps the only surprise here is that the Rangers weren't worse. They had a miserable rotation last year full of pitchers who walked a lot and didn't miss many bats. Park factors help scale this back somewhat, but not enough to hide how atrocious they were.
  4. St. Louis Cardinals, 5.51: Adam Wainwright was good and then there was Kip Wells, Anthony Reyes, Braden Looper and Brad Thompson to muck it all up.
  5. Kansas City Royals, 5.37: Gil Meche was solid enough, but while Zack Greinke looked good, he allowed line drives at a high rate, leading to a tRA barely above average which didn't help enough to overcome luminaries like Odalis Perez and Jorge de la Rosa.

Combined record: 366-444 (45.2%). Probably not as bad as you suspected? Interestingly, teams 6-8 on this list are: Detroit (88 wins), Philadelphia (89 wins) and Seattle (88 wins). Just as having a great rotation is a huge help, it's also no guarantee and vice versa.

BEST BULLPENS IN 2007

  1. Cleveland Indians, 3.91: The dueling Rafael's, Betancourt and Perez were just lights out for the Indians; enough even to outweigh the damage done by Joe Borowski.
  2. Atlanta Braves, 3.99: Peter Moylan came out of nowhere to give Atlanta 90 great innings and oh yeah, it helps when you get to fleece teams out of the likes of Rafael Soriano and Mike Gonzalez.
  3. Milwaukee Brewers, 4.10: The Brewers have felt the departure of Francisco Cordero as he helped them to one of the better bullpens in 2007 and without him, they've fallen to one of the worst in 2008.
  4. Los Angeles Dodgers, 4.13: Takashi Saito and Jonathan Broxton dominated with good support from Rudy Seanez and Joe Beimel.
  5. Toronto Blue Jays, 4.17: That white guy was especially good, but the nowhere near as good as the other white guy, followed closely by the third white guy.

WORST BULLPENS IN 2007

  1. Tampa Bay Devil Rays, 5.66: How bad was this bullpen? They have 60+ innings to Brian Stokes and his 7.07 ERA, 40 IP to Shawn Camp and his 7.20 ERA. They had two pitchers, TWO, with an ERA under 4.89 in their pen.
  2. Cincinnati Reds, 5.33: Thank you Mike Stanton, Todd Coffey, Victor Santos and Kirk Saarloos. Combined they threw 200.3 innings in 2007. So far in 2008, the four of them has thrown 23 innings at the big league level.
  3. New York Yankees, 5.03: Joba Chamberlain is the most-remembered part of this unit, but it also featured an off year from Mariano Rivera and significant time from Sean Henn, Brian Bruney, Edwar Ramirez, etc.
  4. Philadelphia Phillies, 5.03: Brett Myers, Antonio Alfonseca, Tom Gordon's expiring body and Geoff Geary combined to do much harm and in case that wasn't enough they gave 39 innings to Jose Mesa.
  5. Baltimore Orioles, 5.00: The unit we all made fun of last year. The Orioles spent a ton to assemble it, relying on "proven veterans" who promptly shit the bed because, as anyone who valued more than years experience would have told Baltimore, they didn't have actual talent.

0 recs  |  Comment 13 comments

Story-email Email Printer Print

Comments

Display:

Rick VandenHurk

God I love that name.

by JI on Jul 2, 2008 11:53 AM PDT reply actions   0 recs

His first name is actually 'Henricus' so it's even better than you think

I’ve had a huge mancrush on the Dutchman since his Hawaiian winter league season.

by marc w on Jul 2, 2008 3:07 PM PDT up reply actions   0 recs

So, is uderstanding the specifics of tRA absolutely necessary?

I understand that its goal is to take the result of a play out of the equation as much as is possible, and I certainly understand that evaluating processes is highly preferable to evaluating results. I understand that it’s scaled to resemble more traditional pitching metrics, which means I have a decent idea of what’s good and what’s bad, and as time goes on and tRA becomes more familiar, that understanding of what’s good and what’s bad will certainly improve. I also know that people who are much more well informed and gifted than I in these areas, such as Jeff, Matthew, David Cameron and R.J. Anderson, think very highly of tRA and its utility. Not to mention the fact that Graham is some sort of organically created supercomputer.

That being said, I don’t fully grasp how tRA is calculated. I’ve read Graham’s posts several times, and I will continue to try and wrap my head around it, but as of now, it’s outside of my comfort zone. We all have our strengths, and suffice to say, this sort of thing isn’t one of mine. So, while I can read posts such as these and understand what they mean in a larger sense, am I missing out on something major by not really getting it 100%?

As a young boy, I dreamed of being a baseball.

by acblue on Jul 2, 2008 11:56 AM PDT reply actions   0 recs

No, not really

and FWIW, it’s not really “scaled to resemble more traditional pitching metrics” as it is just a fixed up version of the same inputs that go into ERA.

ERA = earned runs allowed / innings pitched * 9
tRA = expected runs allowed / (expected outs generated / 3) * 9

by Matthew on Jul 2, 2008 12:33 PM PDT up reply actions   0 recs

Yeah, I regretted that as soon as I hit post.

I meant that the scale is somewhat similar.

As a young boy, I dreamed of being a baseball.

by acblue on Jul 2, 2008 1:04 PM PDT up reply actions   0 recs

re: "am I missing out on something major by not really getting it 100%?"

What exactly do you want to know?
I’m sure Graham can explain it (or Matthew, Jeff et al) if he has a sense of what it is that’s troubling you.

at its heart, tRA appears to use the run value of each plate appearance – walk, strike-out or batted ball – to create an ‘expected’ runs allowed. So it’s basically using the run values of each type of batted ball (and K’s, strike-outs, etc.) instead of, y’know, what actually happened.

The tougher part to follow (esp. from a math point of view – I’ve stopped trying) is calculating the run values themselves. That’s not something that tRA itself is doing – Tango and a bunch of other really smart people have done a ton of work on this, and those values are being used in lots of metrics now (prOPS, for example). As mathematically tricky as the run values can be, that concept too is really useful and intuitive in a way.

If you’ve got the run values, and you’ve got enough play-by-play data to find the batted ball types for each batter a pitcher faced, calculating it should be fairly straightforward (Graham is probably laughing maniacally at this and inscribing ‘marc w’ on a bullet).

by marc w on Jul 2, 2008 3:18 PM PDT up reply actions   0 recs

Well fine then

that’s nice and all, but…why?
Just trying to make sure the data’s up to date (Tango’s values are based on several seasons in the 80s/90s)?

by marc w on Jul 2, 2008 4:25 PM PDT up reply actions   0 recs

Partly that but mostly because we had to.

Batted balls are relative ergo we had to have run and out values that are based on our interpretation of batted ball types.

by Matthew on Jul 2, 2008 4:53 PM PDT up reply actions   0 recs

Yeah, that makes total sense

Given the league, year, sp-rp context, new parks, using off-the-shelf run values could certainly skew the results.

OK, so yeah, the work required is a lot more, but I still think the concept is ‘elegant’ as you sciencey types say.

by marc w on Jul 2, 2008 5:08 PM PDT up reply actions   0 recs

It is indeed pretty straightforward to calculate tRA itself, so no bullet.

The hard part for the initial run was data input and regression, but as Matthew said below, we’re doing some pretty heavy lifting with the run and out values these days

by Graham on Jul 2, 2008 5:21 PM PDT up reply actions   0 recs

Comments For This Post Are Closed


User Tools

By reading a game thread of your own volition you agree to accept all liability for any and all damage done to your delicate sensibilities.
Start posting about the Mariners »

Join SB Nation and dive into communities focused on all your favorite teams.

FanPosts

Community blog posts and discussion.

Recommended FanPosts

Max_small
FootbaLL IV (Nov 21st)
Small
The Sabermetrics of Cats

Recent FanPosts

Small
WAR gains major exposure
Laughing_man_elmex_small
AFL Championship game thread
Smell-the-glove_small
OFFTOP 11/19 - The Hand of Fraud
Small
Tony Blengino interview on 710 AM ESPN Seattle's Hot Stove League
Ichirocameron_small
OTFPOTD 11/16: Recipes Edition
Small
Lookout Landing as a communtiy
Small
Royals determined to unload Callaspo...
Ichiro_small
Ackley and Triunfel on MLB TV right now

+ New FanPost All FanPosts >


Sexy People

1_small Graham

Small Matthew

Small Jeff