Remember me

Starting Pitcher Rankings, April 19, 2012

April 20, 2012

                Back-to-back brilliant games by Matt Cain of San Francisco, I think his first two starts since he signed his new contract, have moved Cain from 10th in the Starting Pitcher Rankings to 7th.   Cain’s 1-hit shutout of Pittsburgh on April 13, with 11 strikeouts and no walks, was the best game pitched in the five-day starter window April 9 to 13, and the best game pitched in the majors so far this season.    He followed that up with 9 innings of 2-hit shutout ball against Philadelphia on April 18, a start that missed by only one point of being the best game pitched in the majors in the five-day starter window April 14 to 18; that was a Game Score of 86, while Edwin Jackson’s 2-hitter against Cincinnati scores at 87.   These are the updated rankings through games of April 18:

 

Pitchers1 Pitchers2
Pitchers3
Pitchers4 Pitchers5

 

                We’ve got some new information and some color-coding to explain here.   The season so far can be broken down into four five-day "start windows", which are:

                1)  The late March series in Japan,

                2)  April 4 to April 8,

                3)  April 9 to April 13, and

                4)  April 14 to April 18.

                Generally speaking, an active starting pitcher will have one start in each five-day window.  The number in there represents his Game Score in that Start; the "96" for Matt Cain in the 4-9 to 13 start window indicates that his Game Score there was 96, and this is highlighted because that was the best start by any major league starting pitcher in that five-day start window. 

                The top four starters in the majors are as they were at the start of the season—Verlander, Halladay, Cliff Lee and Clayton Kershaw.   The only change this week in the top ten major league pitchers is that Jon Lester, who moved into the top ten last week (replacing Tim Lincecum) dropped out of the top ten this week, replaced by C. J. Wilson.   Congratulations to C. J. on entering the top ten for the first time.

                To the right of the Game Scores are the pitcher’s ranking number and his current rank among the 187 active starting pitchers.   White-on-black indicates those pitchers who have moved up in the rankings the greatest distance since the start of the season (thus are in the black); white-on-red indicates those who have fallen the furthest (thus are in the red.)   Bartolo Colon has moved up the most spots so far this year, in part because he is maybe the only major league pitcher who has four starts already, but mostly because three of the four starts have been extremely good.    Others who are surging in the rankings include Derek Holland, Chad Billingsley, Jason Vargas and Kyle Lohse.  Francisco Liriano has fallen the furthest, dropping from 79th to 105th after three dismal starts.   Most of the pitchers who have fallen the furthest are those who are not in the starting rotation for one reason or another, but a few of them are pitchers who have pitched, but not well.

                In the past week Yovanni Gallardo has passed Zack Greinke to regain his position as Milwaukee’s #1 starting pitcher in our rankings—the only change of a team’s #1 in the past week.  Gallardo started the season as the #17 starter in the majors, dipped after a bad first start, and has regained his position with two good starts.  

                The highest-ranked pitcher who began the season unranked is Lucas Harrell of Houston, currently ranked 145th.   He’s had two good starts in three. 

                Finally. . . ..are you curious what these rankings would have been on July 1, 1965?   1. Koufax, 2. Marichal, 3. Drysdale, 4. Bob Gibson, 5. Jim Maloney. . …124. Tom Parsons, 125. Catfish Hunter (then 19 years old), 126. Frank Kreutzer, 127. Wes Stock, 128. Floyd Weaver.    How about April 19, 1977?    1. Frank Tanana, 2. Tom Seaver, 3. Jim Palmer, 4. Bert Blyleven, 5. Nolan Ryan. . …142. Mike Krukow, 143. Mark Lemongello, 144. Steve Hargan, 145. Wayne Simpson, 146. Mike Flanagan.   Ten years ago today?    1.  Randy Johnson, 2. Pedro Martinez, 3. Curt Schilling, 4. Mike Mussina, 5. Javier Vazquez. . ..189. Tony McKnight, 190. Nate Cornejo, 191. Gil Heredia, 192. Adam Eaton, 193. Todd Stottlemyre.  

                You can find the rankings for any date since about 1958 (until two years ago today) by clicking on this link:

  

http://www.baseballmusings.com/cgi-bin/PitcherRatings.py

 

                And entering the date for which you would like to see rankings.  

                This is the site (Baseball Musings) of well-known baseball researcher (and good friend) Dave Pinto, who has programmed the rankings and combined it with a historical database.    He has blocked out the scores for the last two years out of consideration for me, so that he’s not trampling on my intellectual property, but it’s fun to pick out a pitcher and follow his progress in the rankings, and see who was the #1 at each moment in history.

                Thanks.  Sorry I’m a little late; I’m in Arizona, so I’m two hours behind.  

 

                Bill James

 
 

COMMENTS (9 Comments, most recent shown first)

bjames

My question, I guess, was how well ERA correlates with game scores. If you have two players with identical ERAs, but Pitcher A's average game score is 5 points higher than Pitcher B's, does that mean Pitcher A is more likely to be better the following year?

Yes, it does, because Pitcher A would have better Strikeouts and Walks, and the Strikeouts and Walks are independantly predictive; thus, Pitcher A would be more likely to succeed in the next season. Pitchers give up home runs on 11% of fly balls. If a pitcher gives up more home runs than that one year, it doesn't tend to predict that he will do so again the next year.
7:13 AM Apr 21st
 
enamee
Well, my comment wasn't very well written, and the Gallardo comparison just confused matters. Basically, I found it odd that Lilly's park-adjusted ERA was worse than average, but his average game score was 53. Going into this season, I casually assumed that Lilly had slipped and was no longer a top starter, but that was mainly based on his mediocre ERA. His game scores (and thus his ranking in your system) suggest that he's just fine.

My question, I guess, was how well ERA correlates with game scores. If you have two players with identical ERAs, but Pitcher A's average game score is 5 points higher than Pitcher B's, does that mean Pitcher A is more likely to be better the following year?

That's only tangentially relevant to Ted Lilly, whom I have come to appreciate more in light of this ranking system.
11:29 PM Apr 20th
 
bjames
Sorry, Matthew. I can't figure our what your question was. Ted Lilly's ERAs the last three seasons have been under 4.00 all three years--3.10, 3.62, 3.97. His ERA for the three seasons is better than Gallardo's. His ERA for 2009 and 2010 is better than Gallardo; for 2011 not much worse. Down the stretch in 2012, the MOST relevant rating data, Lilly pitched brilliantly.

In spite of these things, Gallardo still rates ahead of him. What's your point? You're trying to break off some sub-set of games in which his ERA was over 5.00, but his ERA was under 4.00 You're park-adjusting his ERA when he is a pitcher's park but letting it sit when it is in a hitter's park. I can't relate to the fractions. What is the whole of what you're trying to get to?
10:45 PM Apr 20th
 
enamee
Oh, my comment below wasn't meant to suggest that the system overrates Lilly. I just found it interesting that Lilly's 2011 game scores and ERA were so out of sync.

Apologies if I seemed to be beating a dead horse. That wasn't my intention.

Matthew
9:50 PM Apr 20th
 
bjames
Ted Lilly is not overrated in the system. This is my last comment on the issue.
3:44 PM Apr 20th
 
enamee
In the comments to last week's starting pitchers article, I expressed surprise that Ted Lilly was ranked so highly, after a mediocre 2011 and a merely solid 2010. I dug a bit into his gamelogs, and I noticed something interesting (to me, at least): While Lilly had a high ERA and a terrible home runs allowed rate from April-July 2011, his game scores weren't THAT bad. From April-July, his average game score was 48, but his ERA was 5.02.

I've never studied this, but I would guess that someone with an average game score of 48 usually has an ERA that is slightly (but not dramatically) worse than average. But Lilly's 48 produced an ERA over five, which, in today's game, is unplayable.

Over the balance of 2011, Lilly's average game score was 53, but according to Baseball-Reference.com, his park-adjusted ERA was 6 percent worse than average. The previous year, Lilly's average game score was 56 -- but his ERA was 13 percent BETTER than average. If you go by game scores -- which the pitcher ranking system does -- Lilly was a little worse in 2011 than 2010, but he was still an above-average pitcher who rarely missed a start.

For comparison, Yovani Gallardo's past two years (by game scores) are basically like Ted Lilly's in reverse. In 2010, Gallardo had an average game score of 54 and an ERA 5 percent better than average. In 2011, he had an average game score of 56 and an ERA 11 percent better than average.

Because Gallardo was a bit better than Lilly in the more recent season, it's no surprise that he entered this season ranked a hair better than Lilly, 488 points to 478.

I guess my only remaining question is, why would Lilly's 2011 game scores be so out of line with his ERA? Is this something that happens pretty often, or is it unusual? Put another way, how well do game scores correlate with ERA (or component ERA, or Fielding Independent Pitching)?

-Matthew Namee
1:23 PM Apr 20th
 
StatsGuru
Thanks for the kind words, Bill.
12:07 PM Apr 20th
 
wovenstrap
I just wanted to say, an obvious use for this amazing tool for me is fantasy baseball. I don't know about other people, but I think my team it typical in that it has 2 top pitchers (Wilson/Greinke) and then several guys who are all solid B-type choices (Latos, Marcum, Gonzalez (increasingly elite), Hudson, Sanchez, Worley). That's 7 pitchers, and our league forces a 5-man rotation over the course of a season, so you constantly have to assess if Worley is really better than Hudson or whatever. This tool prevents me from getting emotional about a pitcher's last start or two, and also quantifies an eye-popping start and places it in its proper context. Fantasy baseball is partly all about the hot hand, but -- is there really such a thing? do pitchers just actually get substantially better for a season or two? how do you spot that? -- this tool does a great job of letting me capture a true increase/decline in quality in recent starts. For all I know, it'll help me detect the next Cliff Lee-type vault into the elite class of pitchers too. Amazing tool.
11:35 AM Apr 20th
 
tangotiger
Funny... I figured that no matter the time zone, you'd always be ahead.
8:22 AM Apr 20th
 
 
©2024 Be Jolly, Inc. All Rights Reserved.|Powered by Sports Info Solutions|Terms & Conditions|Privacy Policy