MLBN Flexin' their "Sabr" chops: Base Rate
The Lodo Magic Man over on Purple Row pointed out MLBN introducing a new metric they called "Base Rate" or something similar.
Their formula reads as follows:
OPS-AVG.
That's it.
Now, there are two ways to look at this. OPS-AVG could be looked at as SLG+walks (or just the IsoD), or it could be looked at as OBP + IsoP.
My immediate reaction was to eschew it, thinking "well, we have wOBA, why do we really need something else? Linear Weights own." Some very brief discussion followed, and Rox Girl brought up the point that OPS double-counts singles, as they appear in both OBP and SLG. So at least it seems there's some merit.
What I've noticed is that it over-rewards guys with low batting averages. Pujols and Fielder appear near the top, Mauer and Votto are definitely on there, but BaseRate provides some surprises.
Here's the top-10 in the majors by metric:
|
|
OPS |
|
wOBA |
|
BaseRate |
|
|
1 |
1.101 |
Albert Pujols |
.449 |
Albert Pujols |
.774 |
|
|
2 |
1.031 |
Joe Mauer |
.438 |
.715 |
||
|
3 |
Prince Fielder |
1.014 |
Prince Fielder |
.420 |
.681 |
|
|
4 |
.981 |
Joey Votto |
.418 |
.666 |
||
|
5 |
.972 |
.413 |
Joe Mauer |
.666 |
||
|
6 |
Kevin Youkilis |
.961 |
Derrek Lee |
.412 |
Derrek Lee |
.666 |
|
7 |
Adrian Gonzalez |
.958 |
.410 |
.661 |
||
|
8 |
Hanley Ramirez |
.954 |
.408 |
Joey Votto |
.659 |
|
|
9 |
Ben Zobrist |
.948 |
.405 |
Kevin Youkilis |
.656 |
|
|
10 |
.948 |
Ryan Braun |
.405 |
Mark Teixeira |
.656 |
I'm not much of a deep thinker, or someone who formulates metrics, so attempting to figure out what exactly is wrong with this ( or why I should be accepting it) isn't really my strong suit, so I'm turning to the minds over here at BtB.
Just for the record, I'm not suggesting this should overthrow wOBA or anything, I'm pretty on board with linear weights, etc, but I'm wondering if BaseRate can actually provide any value that isn't completely redundant or overly faulty.
(If this has been discussed already, I'll delete this. I ran a quick search to see if there was already a discussion, and I haven't seen anything.)
12 comments
|
0 recs |
Do you like this story?
Comments
I will be closer to the total bases for each player where one triple and 2 outs is the same as 3 single.
- .-. ..- … – / – …. . / .—. .-. - .. . … …
by Jeff Zimmerman on Feb 25, 2010 12:00 PM EST reply actions
I'm sure Tango will dissect it and show what (incorrect) linear weights it implies...
OBP and ISO are good stats with important uses. But once you start wanting an overall metric to answer different questions, you might as well get it right and not keep creating B- stats.
Beyond the Boxscore Not a member? Sign up.
this is roughly what I thought as well
I like the idea of OBP + ISO, but it seems as if by removing one of the doubled set of singles, it’s undervaluing the runner-advancing potential or the on-base potential
Purple Row: Take this personally
http://www.youtube.com/user/rockiesmagicnumber
Learn about Batting Metrics
Learn about Pitching Metrics
by Andrew Martin on Feb 25, 2010 12:49 PM EST up reply actions
That sounds about right.
Double counting singles is fine, because singles are more valuable than walks. By making it OBP+ISO, you’re discounting the movement value of a single.
I’m not sure if it’s necessary, but it is simple. I don’t think it adds anything to OPS though.
Marlin Maniac, a Florida Marlins blog
Come attend Intro to Sabermetrics 101!
Check me out at Beyond the Box Score as well.
that last sentence is kind of the conclusion I drew.
Purple Row: Take this personally
http://www.youtube.com/user/rockiesmagicnumber
Learn about Batting Metrics
Learn about Pitching Metrics
by Andrew Martin on Feb 25, 2010 1:36 PM EST up reply actions
Yeap.
Marlin Maniac, a Florida Marlins blog
Come attend Intro to Sabermetrics 101!
Check me out at Beyond the Box Score as well.
I liked the part where Joe Magrane freaked out about baseball becoming too complicated.
And the part where Harold Reynolds indicated he’s been getting too much heat as an anti-stat guy.
Magrane's a great play-by-play guy
not so much a great analyst
http://citrusjuicing.com/ An SRQ focused-Tampa Bay area sports blog
by CubFanRaysaddict on Feb 25, 2010 11:25 PM EST up reply actions
Gobbelty Goop
Base rate. It’s gobbelty goop. It doesn’t give us any useful information we can’t get with a) stats that more accurately measure actual run values or b) make mathematical sense or c) are already commonly used.
I think what they were attempting to do is correct for the fact that OPS double-counts hits. If they want to do this, the correct stat to use, in my opinion, is APS (Average plus Secondary Average). Give it’s easy to compute and makes some mathematical sense. The formula is simply (TB+BB+SB-CS)/AB.
OPS is generally a bad starting point, given it’s well documented flaws and the mathematical gibberish involved. OPS derivatives are even worse.
http://www.capitolavenueclub.com/
It's not easy to get accurate statistics
derived from already flawed statistics. Alas, MLBN has yet to recognize this.
I like baseball.
I write for Beyond the Box Score and The Hardball Times Fantasy
by Satchel Price on Feb 26, 2010 1:52 PM EST up reply actions
True, absolutely.
But the video of the analysis and stuff is pretty hard to watch for anyone with decent exposure to sabermetrics.
I like baseball.
I write for Beyond the Box Score and The Hardball Times Fantasy
by Satchel Price on Feb 26, 2010 11:02 PM EST via mobile up reply actions

by 


























