AVG v OBP - Page 3

Luis_Rivera · 06-05-2006, 08:51 AM

Quote:

Originally Posted by sixto

This may not be the right thread to bring this up in, but...I was just editing some players, and I noticed that editing their strikeout abilities had no effect on their predicted stats. That is interesting because aren't we all in general agreement that players with higher strikeout totals have a statistically significant depression in BA? It makes me nervous that we could be generating players who can hit .380 but strike out 199 times in 550 PA. And I thought we dealt with this problem like years ago.

But maybe this is the exact right thread for this question, because if I'm way off base, no one will really care that much.

Ugh... this is the one thing I wanted fixed above all else from the last version. For a game engine that's supposed to be based off the principles of DIPS, this is very, very bad.

Luis_Rivera · 06-05-2006, 08:59 AM

...and here's a link to the thread I started back in December that sums up what I wanted done basically:
Revisiting Avoid K's and Contact

Dagrims · 06-05-2006, 01:18 PM

Quote:

Originally Posted by sixto

First column is AB, second is hits, third is SO, fourth is BA, fifth is AB-SO, and the last column is BA on non-SO ABs (including HR).

Sixto 06 is awesome.

Hammer755 · 06-05-2006, 01:43 PM

Quote:

Originally Posted by Dagrims

Sixto:BM is awesome.

Fixed!

Luis_Rivera · 06-05-2006, 02:31 PM

Oh jesus... I just noticed that in not reading very thoroughly the first bunch of posts before I saw the chart and freaked out, that I missed how Sixto included HR in his calculation of BABIP. So, is everything actually fixed with regards to this??

Dagrims · 06-05-2006, 02:48 PM

Quote:

Originally Posted by Hammer755

Fixed!

I was actually looking at the last line of his attached chart.

I like your edit, though.

mlyons · 06-05-2006, 03:25 PM

Quote:

Originally Posted by Luis_Rivera

Oh jesus... I just noticed that in not reading very thoroughly the first bunch of posts before I saw the chart and freaked out, that I missed how Sixto included HR in his calculation of BABIP. So, is everything actually fixed with regards to this??

I just finished running a twenty-five year test league and am now slowly going through retired players to check variabilities and distributions and other things of that nature. Hopefully, I'll have some sort of tentative data at some point later this week on that and other things.

Long answer: I don't know.

Luis_Rivera · 06-05-2006, 03:32 PM

Quote:

Originally Posted by mlyons

I just finished running a twenty-five year test league and am now slowly going through retired players to check variabilities and distributions and other things of that nature. Hopefully, I'll have some sort of tentative data at some point later this week on that and other things.

Long answer: I don't know.

Good man... keep us posted on what you find!!

BoSox Rule · 06-05-2006, 03:54 PM

Quote:

Originally Posted by sixto

This is far from an exhaustive list, but whether you attack the problem from high average or high strikeouts, there are obviously limits to how many hits a man can get in ABs where he does not strike out. Andres Galarraga probably is the ultimate testament to what we're talking about here. He could not have won his batting title striking out like he did early in his career.

First column is AB, second is hits, third is SO, fourth is BA, fifth is AB-SO, and the last column is BA on non-SO ABs (including HR).

Jim Thome's 2001 BABIP was .356. I'd imagine all those numbers aren't right.

The formula for BABIP is (H-HR)/(AB-HR-SO).

sixto · 06-05-2006, 05:33 PM

It was my mistake to call the column BABIP. It should be more like BANSO (batting average no strikeouts).

I apologize though for those who were confused by the fact that those were real-life stats, not simmed stats, in the attached Excel image.

The idea behind BABIP was to determine fielding-independent statistics. HR were included with BB and SO was because HR do not depend on the fielders. That is the only reason they were included - to determine the effect fielders have on statistics.

If you want to determine how a batter was at getting hits when he was at the mercy of fielders, then BABIP will obviously help you get there. However, I don't believe that a player's propensity to strike out has any bearing on that. It was my mistake using the term BABIP. That is not what that column signifies.

Taking HR out of the stats may 'narrow the band' but if anything I wanted to show that no matter how good a player could possibly be at getting hits on balls in play, he could not possibly hit .833 or whatever you would have to do to hit .380 in 550 PA with 199 SO. Whether you take the HR out or not, there is still a virtual cap on how well a player can perform on balls in play.

jgross68.5 · 06-05-2006, 06:22 PM

Quote:

Originally Posted by Gastric ReFlux

jgross68PA = GPA * player's age

edit cuz my typing skillz totally suk today

Interesting.

jgross68.5 · 06-05-2006, 06:23 PM

Quote:

Originally Posted by sixto

It was my mistake to call the column BABIP. It should be more like BANSO (batting average no strikeouts).

No, no, NO! It's BACON!!!! (Batting Average on CONtact)

Hammer755 · 06-05-2006, 06:53 PM

Thsi Thred Iz Usless!!! 6tos Tabul Sux!!!

Luis_Rivera · 06-05-2006, 06:59 PM

Quote:

Originally Posted by sixto

Whether you take the HR out or not, there is still a virtual cap on how well a player can perform on balls in play.

Yep... well, taking homers out of the equation results in less noise for your purposes, as HR correlate positively with strikeouts, and thus causes more extreme BACON numbers like Thome's '01.

BoSox Rule · 06-05-2006, 07:10 PM

By the way, I think the best stat you could use in OOTP2006 is VORP/PA.

kroney37 · 06-06-2006, 08:08 PM

If anyone's interested, this is a link to some interesting research regarding players who reach base on errors much more than average.

http://www.retrosheet.org/Research/RuaneT/error_art.htm

06-05-2006, 02:31 PM	#45 (permalink)
Luis_Rivera Hall Of Famer Join Date: Apr 2004 Location: Massachusetts Posts: 4,177 Thanks: 0 Thanked 1x in 1 post	Oh jesus... I just noticed that in not reading very thoroughly the first bunch of posts before I saw the chart and freaked out, that I missed how Sixto included HR in his calculation of BABIP. So, is everything actually fixed with regards to this?? Last edited by Luis_Rivera; 06-05-2006 at 02:35 PM.

06-05-2006, 05:33 PM	#50 (permalink)
sixto Hall Of Famer Join Date: Jul 2002 Posts: 2,251 Thanks: 1 Thanked 1x in 1 post	It was my mistake to call the column BABIP. It should be more like BANSO (batting average no strikeouts). I apologize though for those who were confused by the fact that those were real-life stats, not simmed stats, in the attached Excel image. The idea behind BABIP was to determine fielding-independent statistics. HR were included with BB and SO was because HR do not depend on the fielders. That is the only reason they were included - to determine the effect fielders have on statistics. If you want to determine how a batter was at getting hits when he was at the mercy of fielders, then BABIP will obviously help you get there. However, I don't believe that a player's propensity to strike out has any bearing on that. It was my mistake using the term BABIP. That is not what that column signifies. Taking HR out of the stats may 'narrow the band' but if anything I wanted to show that no matter how good a player could possibly be at getting hits on balls in play, he could not possibly hit .833 or whatever you would have to do to hit .380 in 550 PA with 199 SO. Whether you take the HR out or not, there is still a virtual cap on how well a player can perform on balls in play. Last edited by sixto; 06-05-2006 at 05:34 PM.

06-05-2006, 06:53 PM	#53 (permalink)
Hammer755 Hall Of Famer Join Date: Dec 2001 Location: Houston, TX Warnings: 11 Posts: 2,345 Thanks: 0 Thanked 0x in 0 posts	Thsi Thred Iz Usless!!! 6tos Tabul Sux!!! __________________

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

06-05-2006, 08:59 AM	#42 (permalink)
Luis_Rivera Hall Of Famer Join Date: Apr 2004 Location: Massachusetts Posts: 4,177 Thanks: 0 Thanked 1x in 1 post	...and here's a link to the thread I started back in December that sums up what I wanted done basically: Revisiting Avoid K's and Contact

06-05-2006, 07:10 PM	#55 (permalink)
BoSox Rule Bat Boy Join Date: Sep 2003 Posts: 16 Thanks: 0 Thanked 0x in 0 posts	By the way, I think the best stat you could use in OOTP2006 is VORP/PA.

06-06-2006, 08:08 PM	#56 (permalink)
kroney37 Minors (Double A) Join Date: Mar 2006 Posts: 132 Thanks: 0 Thanked 0x in 0 posts	If anyone's interested, this is a link to some interesting research regarding players who reach base on errors much more than average. http://www.retrosheet.org/Research/RuaneT/error_art.htm