From colintj at South Side Sox. Find a player's position on the left side and find that player's...

From colintj at South Side Sox. Find a player's position on the left side and find that player's fielding ability relative to his position along the top. The corresponding cell tells you what a player has to hit in terms of wOBA in order to be a league-average player, overall. Remember, wOBA is on the OBP scale, where average is about .335. Good stuff.

about 3 years ago Sky Kalkman 23 comments 0 recs |

Story-email Email Printer Print

Comments

Comparing extremes...

A +1 win catcher can have a .286 wOBA (that’s really really crappy) and be as valuable as a -1 win first baseman with a .384 wOBA (that’s really really good).

Beyond the Boxscore // Calling BJ Upton lazy is lazy.

by Sky Kalkman on Dec 3, 2008 10:58 AM EST reply actions

Hmmm.... I can't find the post there

So is this using positional offensive averages rather than a post-offense-and-defense positional adjustment?

OMG Banny. FWIW I am only crdtng u w/3 runs allwd bc of DDJ OMFG

by Matt Klaassen on Dec 3, 2008 11:57 AM EST reply actions

here's the link

http://www.southsidesox.com/2008/12/1/676779/link-dump-12-1

And it uses Tango’s position adjustments, I’m pretty sure.

Beyond the Boxscore // Calling BJ Upton lazy is lazy.

by Sky Kalkman on Dec 3, 2008 12:23 PM EST up reply actions

if so, shouldn't 3B be the same as 2B and CF?

OMG Banny. FWIW I am only crdtng u w/3 runs allwd bc of DDJ OMFG

by Matt Klaassen on Dec 3, 2008 12:30 PM EST up reply actions

Yes.

Given that 10 points of wOBA is a five-run gap, things look good, though. And everything else lines up. I’m sure colin will see this eventually and give us a heads up.

Beyond the Boxscore // Calling BJ Upton lazy is lazy.

by Sky Kalkman on Dec 3, 2008 12:38 PM EST up reply actions

I used -.25 for 3B

i kind of like it better intuitively, but i intended to use Tango’s numbers.

nothing gets ‘em wetter than infrequent postings on the city’s second favorite team

by colintj on Dec 3, 2008 5:51 PM EST up reply actions

nice job anyway

OMG Banny. FWIW I am only crdtng u w/3 runs allwd bc of DDJ OMFG

by Matt Klaassen on Dec 3, 2008 7:28 PM EST up reply actions

ty

nothing gets ‘em wetter than infrequent postings on the city’s second favorite team

by colintj on Dec 3, 2008 7:48 PM EST up reply actions

good... I hope that didn't sound like a backhanded compliment

someone at Royals review was asking if these were empirical averages, and I expained about Tango’s adjustments, but just for the heck of it, I’m pluggin in the positional splits from b-r — its pretty surprising how very close they are.

OMG Banny. FWIW I am only crdtng u w/3 runs allwd bc of DDJ OMFG

by Matt Klaassen on Dec 3, 2008 10:13 PM EST up reply actions

yeah, they shouldn't be crazy different.

DH is by far the largest difference.

Beyond the Boxscore // Calling BJ Upton lazy is lazy.

by Sky Kalkman on Dec 4, 2008 9:50 AM EST up reply actions

they should be pretty close

speaks to how good the model is, really.

nothing gets ‘em wetter than infrequent postings on the city’s second favorite team

by colintj on Dec 4, 2008 2:13 PM EST up reply actions

Link for defensive stats?

Does anybody have a link to defensive wins stats?

by KMils on Dec 3, 2008 1:53 PM EST reply actions

So much out there

Of the free stuff that is in plays/runs format already (so you don’t have to convert them yourself)

Justin has total value stats at the link… go to position players and look at defensive runs… to convert to wins, divide the number of runs by 10.5 (or 10 for short.

If you don’t like that, check out Pintos Probabilistic Model of Range archives. Take the players out-expected outs, then multiply by .8 to get runs, then divide by 10.5 to get wins (might want to prorate that number over a full season if the player didn’t play close to that in the position in question, something like def. wins times 162/games played)

Those are two quick sources that are relatively complete. There are others.

OMG Banny. FWIW I am only crdtng u w/3 runs allwd bc of DDJ OMFG

by Matt Klaassen on Dec 3, 2008 3:04 PM EST up reply actions