Sabermetrics are a bunch of malarkey . . . - Page 11

Russ · 02-04-2008, 01:39 PM

Quote:

Originally Posted by Anisoptera

Maybe I'll look at it again later and try to explain the difference (it's been awhile and my memory is a bit fuzzy so maybe someone fresher on it can step in), but for now I'll just say I'm pretty sure it is different. The basic idea may be the same, but it's different in an important way.

The only real difference is putting it into a team concept like you stated earlier. I wasn't sure what exzactly you were talking about. But the basic RC formula has always been OBP*SLG*AB. I don't personally have a problem with it and think that the MLV is pretty accurate.

satchel · 02-04-2008, 01:40 PM

Quote:

Originally Posted by lynchjm24

And logging out of this message board once and for all.....

This community has a good mix of traditionalists and SABR guys, I'd say. Did Lynchjm leave because he thought the "idiots" outweighed the "eggheads?" Or was it too much for him to have to deal with even one person whose viewpoint was different than his own?

The potential for debate is one thing that makes this whole scene fun.

Who Killed Kenny? · 02-04-2008, 01:48 PM

I read the whole thread! Only took 2 hours at work (dont tell my boss). It grew from 6 pages to 10 during the time and I wasn't sure I was gonna finish.

I agree VORP and Sabermetrics are tough to understand. Even as a math major in college I had to read about VORP a couple times in order to understand it. It is very valuable in studying individual players. You can reliably find out who the best player is every time. For choosing the Most Outstanding Batter (the default name of the OOTP award) it is terrific. I love stats.

HOWEVER, Most Valuable Player (IMHO) can be something completely different. Yes, it can be the best player on the team, but sometimes it goes to the player that makes his team the best. Who would you rather have Derek Jeter (a very good player, but hardly the best on the team) or Matsui (a better hitter than Jeter)? I think Jeter is his team's MVP many years, and as his team is the normally the amongst the best, he should be MVP more often. Is he the Most Outstanding Batter? No.

In essence the word Valuable causes the most problems. There should really be two separate awards. One based on performance, and one based on performance and human intangiables.

sfeldkamp · 02-04-2008, 01:55 PM

Of course all stat's are descriptive, they are a record of what happened, no matter how you slice and dice them.

Instead of saying that VORP was predictive I should have said that it was a better predictor than RBI, and that's because it isolates a players performance as mentioned.

andymac · 02-04-2008, 01:58 PM

Quote:

Originally Posted by sfeldkamp

Of course all stat's are descriptive, they are a record of what happened, no matter how you slice and dice them.

Instead of saying that VORP was predictive I should have said that it was a better predictor than RBI, and that's because it isolates a players performance as mentioned.

Well, yea, it is a better predictor because it is a better descriptor.

The Wolf · 02-04-2008, 02:18 PM

Quote:

Originally Posted by satchel

Did Lynchjm leave because he thought the "idiots" outweighed the "eggheads?" Or was it too much for him to have to deal with even one person whose viewpoint was different than his own?

ROFL

Anisoptera · 02-04-2008, 02:19 PM

One argument the traditionalists could make, I think, is that oftentimes a player comes to the plate with an idea of a certain thing he wants to do be it hit a home run, sac fly, bunt, or whatever based on what someone else tells him or what he tells himself instead of just going out there and doing the best with what he's presented with. And with those raw stats that are influenced by these mindets, can you really use stats like VORP and what not to determine who the best player is? I guess over so many plate appearances the major saber guys feel it's acceptable noise, but I wonder whether it really is. They say things like RBI and Runs are team dependent and so are not very useful, well, many plate appearances are situation dependent so are they still so much more useful? I wonder.

The Wolf · 02-04-2008, 02:22 PM

Quote:

Originally Posted by andymac

Well, yea, it is a better predictor because it is a better descriptor.

Whichever available stat yields the best approximation of a player's overall performance will, of course, be both, when adjusted for aging and injury. This year it's VORP. Next year it may be something else. The times change, and analytical technology changes along with them. The only constant is change.

Jestre · 02-04-2008, 02:25 PM

I find it amusing that those that believe VORP is the be all end all immediately assume those of us that don't must not understand it. I completely understand VORP and what it is supposed to represent. I merely don't believe it is any thing more than a predictor of future tendencies and a measure of team independant effectiveness and that is it. I can look at counting stats and come up with basically the same conclusions on players that VORP goes through computations to determine...it aint rocket science.

Anisoptera · 02-04-2008, 02:31 PM

Quote:

Originally Posted by Jestre

I find it amusing that those that believe VORP is the be all end all immediately assume those of us that don't must not understand it. I completely understand VORP and what it is supposed to represent. I merely don't believe it is any thing more than a predictor of future tendencies and a measure of team independant effectiveness and that is it. I can look at counting stats and come up with basically the same conclusions on players that VORP goes through computations to determine...it aint rocket science.

If you honestly think RBIs are a better indicator of who had a better season, I find it impossible to believe you really understand VORP. I'll admit, VORP was not the easiest thing in the world for me to understand (I had to write out the formulas, try to understand each part individually and that took a good long while), but once you do I think you have to admit it is far better of an indicator.

The Wolf · 02-04-2008, 02:34 PM

Quote:

Originally Posted by Jestre

I find it amusing that those that believe VORP is the be all end all

It's not. It's just the best tool we have today. Soon there will no doubt be better ones.

1998 Yankees · 02-04-2008, 02:38 PM

Quote:

Originally Posted by The Wolf

It's not. It's just the best tool we have today. Soon there will no doubt be better ones.

Jeez, I hope not.

The Wolf · 02-04-2008, 02:39 PM

Reactionary!

Jestre · 02-04-2008, 02:47 PM

Quote:

Originally Posted by Anisoptera

If you honestly think RBIs are a better indicator of who had a better season, I find it impossible to believe you really understand VORP. I'll admit, VORP was not the easiest thing in the world for me to understand (I had to write out the formulas, try to understand each part individually and that took a good long while), but once you do I think you have to admit it is far better of an indicator.

I never said RBI's are a better indicator but they are a component. Those of us that do not worship at the altar of VORP consider many factors in determining a players season. To ignore some stats because they are team oriented or situational in grading a players performance is simply not realistic in that that decision can penalize or benefit some players artificially, which puts VORP firmly in the realm of theoretical. I would bet that the majority of non VORP'ers would rate players extremely similar to VORP'ers on a value basis but perhaps significantly different on a seasonal performance basis.

Russ · 02-04-2008, 02:47 PM

Jestre, I see where you're coming from.

The basis of RC is that OBP and SLG have the greatest effect on runs created. Hypothetically, the player with the highest OPS will have the highest RC if playing time is equal. IRL, that's not always the case, though.

CBL-Commish · 02-04-2008, 04:08 PM

Quote:

Originally Posted by Jestre

I can look at counting stats and come up with basically the same conclusions on players that VORP goes through computations to determine...it aint rocket science.

Not if you're looking at a list of 100s of players playing different positions and in wildly different run contexts. Maybe you can swag it close 66% of the time or something, but actually running the numbers gets you orders of magnitude more accuracy.

Jestre · 02-04-2008, 04:25 PM

Quote:

Originally Posted by CBL-Commish

Not if you're looking at a list of 100s of players playing different positions and in wildly different run contexts. Maybe you can swag it close 66% of the time or something, but actually running the numbers gets you orders of magnitude more accuracy.

I've never worked with a group of 100's of players with similar numbers, nor do I see any situation where that would be the case. It is quite easy to take a field of players from a season and break them down into manageable similar groupings from there it is just as easy to rate them in order of quality. We are talking about a typical season wherein normal bell curve rules apply....once again it aint rocket science.

CBL-Commish · 02-04-2008, 04:35 PM

Quote:

Originally Posted by Jestre

I've never worked with a group of 100's of players with similar numbers, nor do I see any situation where that would be the case. It is quite easy to take a field of players from a season and break them down into manageable similar groupings from there it is just as easy to rate them in order of quality. We are talking about a typical season wherein normal bell curve rules apply....once again it aint rocket science.

I suppose that'll work if you're not that concerned with accuracy, or if you'd rather spend the time to break things up into manageable groupings and rate them yourself rather than just clicking on a list and sorting.

Neags23 · 02-04-2008, 04:41 PM

Wouldn't some kind of RBI% formula like,

RBI-HR/(RBI-HR)+LOB

(where LOB is runners left on base for the entire season)

at least give you a better idea of production than RBIs?

I don't know if that's really a formula since I just made it up in my head, but I'm basically trying to get an RBI efficiency stat. So the number of RBIs you get minus the number of times you drove in yourself, divided by the total # of runners on base when you came to the plate.

Shouldn't that give you how efficient you were at driving in runs? That wouldn't penalize leadoff hitters who have no one on base in front of them, etc.

So if a hitter has 120 RBIs, 35 homers and left 130 runners on base would have an RBI% of .395.

Or for a leadoff guy who has 40 RBIs, 5 homers, and left 40 runners on base would have an RBI% of .467.

Of course, I don't know if they keep up with the total # of runners left on base for the entire season for each player, and I don't know if that formula actually does what I'm trying to get it to do.

I got a C in STAT405...

BoofBonser26 · 02-04-2008, 04:59 PM

Quote:

Originally Posted by Neags23

Wouldn't some kind of RBI% formula like,

RBI-HR/(RBI-HR)+LOB

(where LOB is runners left on base for the entire season)

at least give you a better idea of production than RBIs?

Careful!

http://www.firejoemorgan.com/2008/01...ssed-this.html

02-04-2008, 02:25 PM	#209
Jestre All Star Reserve Join Date: Dec 2001 Location: North Smithfield,Ri,USA Posts: 612	I find it amusing that those that believe VORP is the be all end all immediately assume those of us that don't must not understand it. I completely understand VORP and what it is supposed to represent. I merely don't believe it is any thing more than a predictor of future tendencies and a measure of team independant effectiveness and that is it. I can look at counting stats and come up with basically the same conclusions on players that VORP goes through computations to determine...it aint rocket science. __________________ My eyes perceive the present, but my roots are imbedded deeply in the grandeur of the past. "Chief Meyers"

02-04-2008, 02:47 PM	#215
Russ All Star Starter Join Date: Oct 2005 Location: Essex HON! Posts: 1,923 Infractions: 0/1 (1)	Jestre, I see where you're coming from. The basis of RC is that OBP and SLG have the greatest effect on runs created. Hypothetically, the player with the highest OPS will have the highest RC if playing time is equal. IRL, that's not always the case, though. __________________ If you don't love Russ, you don't love America. This post brought to you by Carl's Jr.

02-04-2008, 04:41 PM	#219
Neags23 All Star Starter Join Date: Apr 2006 Posts: 1,577	Wouldn't some kind of RBI% formula like, RBI-HR/(RBI-HR)+LOB (where LOB is runners left on base for the entire season) at least give you a better idea of production than RBIs? I don't know if that's really a formula since I just made it up in my head, but I'm basically trying to get an RBI efficiency stat. So the number of RBIs you get minus the number of times you drove in yourself, divided by the total # of runners on base when you came to the plate. Shouldn't that give you how efficient you were at driving in runs? That wouldn't penalize leadoff hitters who have no one on base in front of them, etc. So if a hitter has 120 RBIs, 35 homers and left 130 runners on base would have an RBI% of .395. Or for a leadoff guy who has 40 RBIs, 5 homers, and left 40 runners on base would have an RBI% of .467. Of course, I don't know if they keep up with the total # of runners left on base for the entire season for each player, and I don't know if that formula actually does what I'm trying to get it to do. I got a C in STAT405... __________________ GM Havana Sugar Kings, World Baseball League - 2000, 2003, 2005 WBL Champions Former GM Washburn Sea Wolves Dog Days Baseball - 1981 & 1986 Kennel Cup Champions

02-04-2008, 01:48 PM	#203
Who Killed Kenny? Major Leagues Join Date: Mar 2007 Posts: 410	I read the whole thread! Only took 2 hours at work (dont tell my boss). It grew from 6 pages to 10 during the time and I wasn't sure I was gonna finish. I agree VORP and Sabermetrics are tough to understand. Even as a math major in college I had to read about VORP a couple times in order to understand it. It is very valuable in studying individual players. You can reliably find out who the best player is every time. For choosing the Most Outstanding Batter (the default name of the OOTP award) it is terrific. I love stats. HOWEVER, Most Valuable Player (IMHO) can be something completely different. Yes, it can be the best player on the team, but sometimes it goes to the player that makes his team the best. Who would you rather have Derek Jeter (a very good player, but hardly the best on the team) or Matsui (a better hitter than Jeter)? I think Jeter is his team's MVP many years, and as his team is the normally the amongst the best, he should be MVP more often. Is he the Most Outstanding Batter? No. In essence the word Valuable causes the most problems. There should really be two separate awards. One based on performance, and one based on performance and human intangiables.

02-04-2008, 01:55 PM	#204
sfeldkamp Minors (Double A) Join Date: Mar 2006 Posts: 184	Of course all stat's are descriptive, they are a record of what happened, no matter how you slice and dice them. Instead of saying that VORP was predictive I should have said that it was a better predictor than RBI, and that's because it isolates a players performance as mentioned.

02-04-2008, 02:19 PM	#207
Anisoptera Minors (Double A) Join Date: Aug 2007 Posts: 169	One argument the traditionalists could make, I think, is that oftentimes a player comes to the plate with an idea of a certain thing he wants to do be it hit a home run, sac fly, bunt, or whatever based on what someone else tells him or what he tells himself instead of just going out there and doing the best with what he's presented with. And with those raw stats that are influenced by these mindets, can you really use stats like VORP and what not to determine who the best player is? I guess over so many plate appearances the major saber guys feel it's acceptable noise, but I wonder whether it really is. They say things like RBI and Runs are team dependent and so are not very useful, well, many plate appearances are situation dependent so are they still so much more useful? I wonder.