A Graphic Look at Redefining Cy Young

by David Fung on Nov 14, 2011 10:00 AM EST in Visuals Above Replacement

To go along with a piece I did two weeks ago about taking a different look at MVPs, lets take another look at past Cy Young winners and see if there's some better metrics we can use to determine quality or worth. Unlike the previous study, it's been harder to predict as winners have one anywhere from one to all of the six categories. To go with basic units of measure, ERA, SO, Wins, the three advanced metrics we'll be looking at is:

-FIP (Fielding Independent Pitching)
-K/9 (Strikeouts per 9 innings)
-(K-BB)/PA (Strikeout/BB differential per plate appearance)

Were there generalizations to be made? Read more after the jump.

All I could take away from looking at this from a high level is that most winners or candidates that should have won, had really good (K-BB)/PA numbers, and generally were also leaders in FIP (in most years, under 3.00 FIP, with the exception of 2006-2007).

So who were good candidates according to have won? The most recent was Josh Johnson last year and CC Sabathia in 2008. I know, it's highly debatable right?

For full-size graphic, click the image or here. See more at http://FUNGraphs.tumblr.com or follow me on Twitter @cobradave

Poll

Which one of these candidates should have won the Cy Young?

Josh Johnson over Roy Halliday in 2010

5 votes

CC Sabathia over Cliff Lee in 2008

3 votes

Eric Bedard over CC Sabathia in 2007

6 votes

Johan Santana over Bartolo Colon in 2005

25 votes

Randy Johnson over Roger Clemens in 2004

19 votes

Pedro Martinez over Roy Halliday in 2003

2 votes

Pedro Martinez over Barry Zito in 2002

29 votes

Mike Mussina over Roger Clemens in 2001

10 votes

99 votes | Poll has closed

Tweet Comment 8 comments | 1 recs |

Do you like this story?

Comments

It’s Roy Halladay

"We are all agreed that your theory is crazy. The question that divides us is whether it is crazy enough to have a chance of being correct."
- Niels Bohr

by Frag on Nov 14, 2011 10:16 AM EST reply actions

Good catch.

by David Fung on Nov 14, 2011 11:29 AM EST up reply actions

Nice idea, but why FIP? FIP has its own flaws, particularly with respect to HR/FB ratio. That’s why we have xFIP. Halladay was better according to both xFIP and SIERA.

by Rujasu on Nov 14, 2011 11:56 AM EST reply actions

xFIP's tendency to regress to HR/FB isn't necessarily an improvement

And many would argue it’s a flaw. Just like not all pitchers regress to a .300 BABIP, not all pitchers regress to 10.6 HR/FB, either.

Among the class of pitchers who actually do possess BABIP and HR/FB skill, Mariano Rivera for instance, FIP underestimates them and xFIP underestimates them further.

Jesse-Douglas Mathewson, Ph.D. Candidate in Government and Politics at UMD-College Park.
Columnist for Beyond the Box Score specializing in projections, PitchF/X and infographics.
Blogger and Editor, Rational Pastime Blog. Twitter: @RationalPastime.

by J-Doug on Nov 14, 2011 3:19 PM EST up reply actions

Neither is perfect, hence why it’s good to look at the whole crop of them. I would say that xFIP is more indicative in smaller samples (one season), whereas over a full career it may be better just to look at FIP or ERA+, in case there’s a long term trend like you see with some guys (Rivera, Matt Cain). Those guys are definite outliers, though. Since you’re looking at just one season with the Cy, I think it makes more sense to look at xFIP rather than FIP, since you don’t have enough information to know if there’s a long term trend of BABIP or HR/FB prevention. Taking the example of Halladay and Johnson, Halladay’s better xFIP and SIERA for the one season seem to be more in line with his better career numbers vs. Johnson’s, and it’s the FIP that looks like the big outlier. Again, I’d say it’s useful to look at both. I’d give xFIP a little more weight, though, over a single season.

by Rujasu on Nov 14, 2011 3:57 PM EST up reply actions

Also...

Many would argue it’s flawed because of its dependence on batted ball data, which has its obvious issues.

I write a Giants blog. I also write for MLB Daily Dish and Beyond the Box Score

by Julian Levine on Nov 17, 2011 1:05 AM EST up reply actions

I simply picked FIP as it was easier for most people to understand, including myself. I didn’t take a look at xFIP, but SIERA got a little too complex. When I update this, I’ll take a stab at xFIP though, I’ve already started updating my previous Most Valuable graphic with new values.

Thanks for the feedback.

by David Fung on Nov 14, 2011 1:32 PM EST reply actions

Innings pitched

All three of the stats you have chosen for the graph are rate stats. I would imagine that every single year you could find a reliever who outperformed the starters in these categories. It simply doesn’t say a whole lot about what is more valuable when you only look at the quality of pitching but not the quantity of innings pitched. This is the reason why you could argue for Halladay over Johnson last year (250.2 vs. 183.2 IP).

by Indian Bob on Nov 15, 2011 8:49 AM EST reply actions