wOBA and GIDP

by nny on Sep 13, 2009 10:06 AM EDT Comment 6 comments

Hi. :D

Quick summary: I'm trying to predict future team performance and I was using wOBA (The formula at http://www.insidethebook.com/woba.shtml ). However, when comparing the OPS/run totals to OPS/run totals that actually have occurred, my run totals were way higher. For instance, the one I have up right now is a .764 OPS. Teams generally scored between 770-800 runs with that OPS. wOBA said they would finish with 827 runs, quite a bit higher than what past results have said.

I have a friend who basically only uses Extrapolated Runs so I decided to try that out. Same results, 827 runs. Except I had excluded GIDP from XR since I didn't project them (wasn't needed for wOBA after all). Well, once I gave the team the average amount of GIDP (~130), that lowered their XR down to 779, a hell of a lot more reasonable number.

So given that XR and wOBA had the same results without GIDP, and that XR then had reasonable results after including GIDP, the easy conclusion is that the issue with wOBA was GIDP.

Is there a wOBA formula that takes GIDP into account? On Fangraphs I believe I've read their wOBA takes base running into account but they've never actually said how. On your power rankings you use BP's base running, but I believe that doesn't take into account GIDP either.

So, yeah, what's up?

<3Danke sehr.<3

0 recs | Comment 6 comments |

Story-email Email Printer Print

Comments

On FanGraphs they just take stolen bases into account

+0.25 SB, -0.5 CS

As for what you said, there is some merit to this, I think. When calculating individual player’s offensive contributions, Rally comes up with a GIDP runs above average as well.

However, I’m not sure what your methodology here is. Are you using straight wRC (wOBA RC?) to do this, or wOBA runs above a league average total. Whenever I do something like this, I take a 4-year average for runs scored and a four year average for wOBA and do it using wRAA. Is that similar to what you’re getting at?

Marlin Maniac, a Florida Marlins blog

by SFiercex4 on Sep 13, 2009 10:28 AM EDT reply actions

wRC

getting pythagorean win percentage so I need total total runs

by nny on Sep 13, 2009 10:36 AM EDT up reply actions

Can you give us an example of how you're doing it?

I has a blog. | Ask a sabermagician!

by cwyers on Sep 13, 2009 11:43 AM EDT reply actions

I am extremely dumb

So I am an extremely unorganized person. I only have one spreedsheet that I use for projection and it has some 20 tabs on it. When I was first creating it I was working on projecting minor leaguers with a set 650 PA. Rather than project each individual’s HBP, I just gave them the league average over 650 PA (6).

So now it’s awhile later and not wanting to re-type out the entire wOBA formula, I simply copy it from a cell there and paste it onto my new tab. Completely forgetting about the set HBP amount.

Fixed it so that now it’s rounding HBP=PA*.0095 and now I’m getting a lot better results lol

So, yeah, you can ignore this, lol. SORRRRRRRRRYYYY

by nny on Sep 13, 2009 12:34 PM EDT reply actions

Shouldn't you be treating HBP like walks? The distribution of them is not random.

Linda's in the cold ground, won't see her anymore
Somewhere out on the highway tonight, the drunken engines roar
It's just one of those things, one of those things
-- Al Stewart, "Accident on 3rd St."
In memory of Nick Adenhart and all victims of drunk driving

by PaulThomas on Sep 14, 2009 2:57 AM EDT up reply actions