Defense Independent Component ERA (DICE)

by Skirra on Apr 26, 2008 1:33 PM EDT Comment 15 comments

This offseason there was some discussion of Defense Independent Component ERA (DICE) as a way to rate a pitchers performance. I apologize to whomever brought it up for not giving proper credit but let me know and I will give proper attribution. For those of you who are not familiar with DICE here is a explanation.

The basic idea is to evaluate a pitchers performance in a readable way independent of defense. I have been thinking of providing the DICE for the Royals on a regular basis but when I ran the numbers earlier in the year they were all over the place (some pitchers had negative DICE numbers) due to small sample size issues. I hope now that there is some more data I will be able to do this at 2-3 week intervals.

As an observation the focus on defense has apparently not helped out the pitchers as almost all of them do better when defense is removed from the equation, although Nomo is still terrible.

Player	ERA	DICE
Z Greinke	1.25	3.50
B Bannister	2.48	2.94
G Meche	8.00	5.48
B Tomko	5.25	3.92
J Bale	7.63	3.40
R Ramirez	0.87	0.92
R Mahay	2.70	2.90
L Nunez	2.70	2.20
Y Yabuta	9.00	6.70
J Soria	0.00	0.56
L Hochevar	11.57	2.76
J Gobble	6.23	1.78
H Nomo	18.69	13.98
J Peralta	4.15	1.29

Poll

Would you like to see this continue as a regularly updated (every 2-3 weeks) feature?

Yes

25 votes

No

1 votes

26 votes | Poll has closed

3 recs | Comment 15 comments

Story-email Email Printer Print

Comments

tremendous

by royalsreview on Apr 26, 2008 1:37 PM EDT reply actions 0 recs

I brought it up

It was me, in the study, with the laptop.

Finally coming back to the 'K'!!!! I'm going to be there 7/11/08, vs the SEA Mariners, of course!

by RoyalsFanInMarinerTown on Apr 26, 2008 1:49 PM EDT reply actions 0 recs

Thanks for the idea.

Proud father of a budding Royals fan.

by Skirra on Apr 26, 2008 3:17 PM EDT up reply actions 0 recs

SORIA IS SOOOOOOOOOO GOOOD!!

The man is off this planet.. 23 IN A ROW he sat down and still counting.. lets see if Soria can break the consecutive men sat down record that JENKS couldnt cause of Mr. Gathright

A lot of you wont like me... by I assure you my opinion will be voiced GO ROYALS!

by focs on Apr 26, 2008 1:54 PM EDT reply actions 0 recs

Ok, now that I have claimed the original post

Thanks for posting this. I did run all of their DICE’s about 10-14 days ago, but didn’t post it because I ran out of time and was headed to my class. What I like about DICE is it seems to jive with me on some level of my instict from watching the game. I know Meche has been better than 8.00 ERA (though not much). I know Yabuta has been better. I know Greinke is getting mildly lucky sometimes, but would be dazzling still, with a bit less luck.

The point is, I really feel this statistic and tRA have some merit, but with school, work, wife, and 3 kids, I do not have time to do the devoted study of it that I am capable of.

Is there anyone on the board who would be willing to break this formula down and register an intelligent position on why or why it does not work?

Finally coming back to the 'K'!!!! I'm going to be there 7/11/08, vs the SEA Mariners, of course!

by RoyalsFanInMarinerTown on Apr 26, 2008 2:00 PM EDT reply actions 0 recs

Why not just use FIP?

FIP is essentially the same formula and is widely available (hardballtimes.com, fangraphs.com, firstinning.com), so no math homework is necessary.

by Gopherballs on Apr 26, 2008 2:34 PM EDT reply actions 0 recs

Interesting...

What does FIP stand for?

Finally coming back to the 'K'!!!! I'm going to be there 7/11/08, vs the SEA Mariners, of course!

by RoyalsFanInMarinerTown on Apr 26, 2008 2:35 PM EDT up reply actions 0 recs

Fielding Independent Pitching

Tom Tango, of The Book fame, developed it. FIP uses a slightly more conservative adjustment than DICE in attempt to more closely reflect the same scale as ERA.

by Gopherballs on Apr 26, 2008 2:48 PM EDT up reply actions 0 recs

They are the essentialy the same and I already have web connected DICE spreadsheet

so I will continue with it, but thanks for the alternative. The main difference is that FIP removes IBB from the total BB. I will let others debate the value of this difference but it probably comes up only a few times a season for a particular pitcher.

As to the suggestion by RoyalFanInMarinerTown that someone break DICE down and analyze its merits I wholeheartedly agree. I would think that the Grand slam given up by Meche hurt his ERA more than it did his DICE so I suspect that there are some weaknesses with this calculation but perhaps I could be wrong. Overall, I also believe that it has merit but am willing to listen to counter arguments.

Proud father of a budding Royals fan.

by Skirra on Apr 26, 2008 3:33 PM EDT up reply actions 0 recs

Either FIP or DICE works in this situation

FIP is better known, easier to find, and probably a little more accurate, but the difference is in the 0.10-0.20 range (depending on which FIP adjustment is used).

FIP and DICE are light years better than ERA, but you still need to look at hit data (Grounball, Line Drive, and Flyball percentages, BABIP, etc.) to get the full picture. 75 percent of line drives go for hits, so if a pitcher has a higher than average line drive percentage (20% LD% is roughly average), the pitcher is not performing as well as his FIP. Similarly, because groundballs rarely go for extra base hits and often can result in double plays, a pitcher with a high groundball rate (43% GB% is roughly average) and a normal LD% is likely performing better than his FIP.

by Gopherballs on Apr 26, 2008 10:08 PM EDT up reply actions 0 recs

Where can I find BABIP

and the other percentages that you suggest influence the accuracy of FIP/DICE? Baseball-Reference.com has BABIP but it includes MiLB numbers for guys like Hocheaver in its calculations and that would throw off the value of showing his BABIP as a predictor of future performance?

If I can get these numbers next time I will do a more detailed listing and perhaps we can eventually extend this to some of the top prospects near the end of the season to see who deserves a September call-up.

Proud father of a budding Royals fan.

by Skirra on Apr 29, 2008 3:16 PM EDT up reply actions 0 recs

Fangraphs.com has pretty much everything

including Line Drive %, Groundball %, and BABIP, plus other good stuff like average pitch speeds and graphs comparing a player’s performance against league averages. Fangraphs also started a blog that does a very good job of breaking down advanced statistical analysis into very accessible, easy-to-read articles.

Hardballtimes.com also has most of the same data, plus very good articles (which sometimes are a little too dense for casual fans). Firstinning.com also provides some minor league data that Fangraphs does not (like minor league GB%).

Thanks for doing this.

by Gopherballs on Apr 29, 2008 11:08 PM EDT up reply actions 0 recs

Very interesting

Thanks for posting this. I’d like to see it continue as a regularly updated feature. It will be interesting to see how the numbers correlate to the actual ERAs over time.

by Chico Lind with pants on on Apr 26, 2008 2:44 PM EDT reply actions 0 recs

I voted NO

just because nobody else had and I didn’t want to see a shutout.

Disclaimer: Comments may not be suitable for young children or women who are pregnant, or women who think they may be pregnant. Side effects could include nausea, dizziness, or yelling at the monitor in disbelief.

by MileHighKCfan on Apr 28, 2008 3:08 PM EDT reply actions 0 recs