[quote]therajraj wrote:
meant Verlander’s 2011 Cy Young/MVP year. Not 2012.
Verlander had a 6.8 WAR in 2011 and a 7 WAR in 2012
[/quote]
Why does WAR even matter in this discussion? You’re the one who admitted last year that it wasn’t intended to be used regarding pitchers after I pointed out that it was a flawed statistic, given that the team with the 5th highest WAR in the NL amongst their pitchers was Colorado’s staff, which I believe ended up with one of the highest team ERAs in the last 30 years.
[quote]therajraj wrote:
meant Verlander’s 2011 Cy Young/MVP year. Not 2012.
Verlander had a 6.8 WAR in 2011 and a 7 WAR in 2012
[/quote]
Why does WAR even matter in this discussion? You’re the one who admitted last year that it wasn’t intended to be used regarding pitchers after I pointed out that it was a flawed statistic, given that the team with the 5th highest WAR in the NL amongst their pitchers was Colorado’s staff, which I believe ended up with one of the highest team ERAs in the last 30 years.[/quote]
I dont’ remember saying that. I remember discussing how the cumulative WARs of a pitching staff didn’t necessarily reflect who had the best staff. But on the individual level it’s perfectly valid.
[quote]therajraj wrote:
2012 cumulative WAR of starting pitchers only (no bullpen)
[/quote]
I have to agree with DB about WAR.
Have you read how WAR is calculated? Good lord is it a bunch of bullshit. I mean, I love the statistics side of baseball, especially as an avid fantasy fan, BUT, WAR is an overly-complicated computation created to keep sabermetrics companies in business. Nothing like basing a statistic off of a hypothetically playing backup.
I also think there are a bunch of other useless and trivial stats like isolated power. There’s a stat to measure power already…it’s called slugging.
[quote]therajraj wrote:
meant Verlander’s 2011 Cy Young/MVP year. Not 2012.
Verlander had a 6.8 WAR in 2011 and a 7 WAR in 2012
[/quote]
Why does WAR even matter in this discussion? You’re the one who admitted last year that it wasn’t intended to be used regarding pitchers after I pointed out that it was a flawed statistic, given that the team with the 5th highest WAR in the NL amongst their pitchers was Colorado’s staff, which I believe ended up with one of the highest team ERAs in the last 30 years.[/quote]
I dont’ remember saying that. I remember discussing how the cumulative WARs of a pitching staff didn’t necessarily reflect who had the best staff. But on the individual level it’s perfectly valid.
But yes, every statistic is flawed.
[/quote]
You don’t remember that either? You said that WAR was not intended to be used in reference to pitching. And now you turn around and try to supplement your argument with WAR? Do you even believe anything you say, or do you just try to manipulate what little you do know about the game in order to suit whatever argument you’re trying to make at the time? It seems like a big-time sabermetrics guy like yourself would remember how you felt about the application of one of the most (over) utilized advanced metrics in the game.
And just for the sake of argument, could you please explain to me why individual WARs of pitchers is perfectly valid but adding up their individual WARs to achieve a staff-wide WAR can be so out of whack that Colorado ended up with the 5th-best in the NL last year?
How can that be a valid statistic, Raj? Think about that for a minute. Doesn’t it seem like something is completely amiss with your statement?
[quote]therajraj wrote:
2012 cumulative WAR of starting pitchers only (no bullpen)
[/quote]
I have to agree with DB about WAR.
Have you read how WAR is calculated? Good lord is it a bunch of bullshit. I mean, I love the statistics side of baseball, especially as an avid fantasy fan, BUT, WAR is an overly-complicated computation created to keep sabermetrics companies in business. Nothing like basing a statistic off of a hypothetically playing backup.
I also think there are a bunch of other useless and trivial stats like isolated power. There’s a stat to measure power already…it’s called slugging.
WAR is based on a completely arbitrary assumption of what a “replacement” player is. It’s mind-boggling.
Some teams don’t even have a “replacement” player available in their minor-league system. Look at the Giants starting rotation right now. If they suffer another injury they’re fucked because all of their top pitching prospects are younger guys down in Single A right now. I doubt they have a pitcher at Triple A who could come up and even approach “replacement player” quality.
So right there the theoretical replacement player who the current starters are being compared against, the wins above and beyond what that guy could contribute to, is wildly off-base and not rooted in reality at all. The Giants’ starting staff should have much higher WARs than they actually do because they are theoretically contributing to way more wins than even the best “replacement player” available to them could contribute to.
You don’t remember that either? You said that WAR was not intended to be used in reference to pitching. And now you turn around and try to supplement your argument with WAR? Do you even believe anything you say, or do you just try to manipulate what little you do know about the game in order to suit whatever argument you’re trying to make at the time? It seems like a big-time sabermetrics guy like yourself would remember how you felt about the application of one of the most (over) utilized advanced metrics in the game. [/quote]
I don’t believe what I say, you got me!
We’ve had several page discussions on WAR. How the fuck am I supposed to remember every single thing I’ve said on the matter? If you really care to continue this discussion, link me to what comment you’re referring to.
[quote]DBCooper wrote:
And just for the sake of argument, could you please explain to me why individual WARs of pitchers is perfectly valid but adding up their individual WARs to achieve a staff-wide WAR can be so out of whack that Colorado ended up with the 5th-best in the NL last year?
How can that be a valid statistic, Raj? Think about that for a minute. Doesn’t it seem like something is completely amiss with your statement?[/quote]
Every statistic has its limitations. Just because you have identified one in-congruence in the stat, doesn’t mean it invalidates the stat.
Tell me, how far off from your perception is individual WAR for pitchers in 2012?
WAR is based on a completely arbitrary assumption of what a “replacement” player is. It’s mind-boggling.
Some teams don’t even have a “replacement” player available in their minor-league system. Look at the Giants starting rotation right now. If they suffer another injury they’re fucked because all of their top pitching prospects are younger guys down in Single A right now. I doubt they have a pitcher at Triple A who could come up and even approach “replacement player” quality.
So right there the theoretical replacement player who the current starters are being compared against, the wins above and beyond what that guy could contribute to, is wildly off-base and not rooted in reality at all. The Giants’ starting staff should have much higher WARs than they actually do because they are theoretically contributing to way more wins than even the best “replacement player” available to them could contribute to.[/quote]
You’re still not getting it, none of what you wrote here matters. You are again, hung up on the name of the stat.
It’s the number of times a Major leaguer is better than the average minor league/replacement type player. Not in any particular team’s system.
Replacement player is just a measuring stick to gauge a players worth, it’s not meant as an illustration of a major league-minor league dynamic that some seem to think it is.
Have you read how WAR is calculated? Good lord is it a bunch of bullshit. I mean, I love the statistics side of baseball, especially as an avid fantasy fan, BUT, WAR is an overly-complicated computation created to keep sabermetrics companies in business. Nothing like basing a statistic off of a hypothetically playing backup.
Have you read how WAR is calculated? Good lord is it a bunch of bullshit. I mean, I love the statistics side of baseball, especially as an avid fantasy fan, BUT, WAR is an overly-complicated computation created to keep sabermetrics companies in business. Nothing like basing a statistic off of a hypothetically playing backup.
What exactly is your criticism of WAR? That it’s a complicated statistic to calculate?
[/quote]
Moreso that you can’t really define what a replacement player is.
Also that it’s a totally hypothetical statistic…why would you base a statistic on someone who is NOT playing? That makes absolutely no sense. It’s like saying “Well, in an alternate universe, where your team isn’t made up of the same players…you have x less/more wins.” It tells you nothing.
But yes, also that it is cleary overly complicated - “If you can’t explain it simply, you don’t understand it well enough.”
WAR is based on a completely arbitrary assumption of what a “replacement” player is. It’s mind-boggling.
Some teams don’t even have a “replacement” player available in their minor-league system. Look at the Giants starting rotation right now. If they suffer another injury they’re fucked because all of their top pitching prospects are younger guys down in Single A right now. I doubt they have a pitcher at Triple A who could come up and even approach “replacement player” quality.
So right there the theoretical replacement player who the current starters are being compared against, the wins above and beyond what that guy could contribute to, is wildly off-base and not rooted in reality at all. The Giants’ starting staff should have much higher WARs than they actually do because they are theoretically contributing to way more wins than even the best “replacement player” available to them could contribute to.[/quote]
You’re still not getting it, none of what you wrote here matters. You are again, hung up on the name of the stat.
It’s the number of times a Major leaguer is better than the average minor league/replacement type player. Not in any particular team’s system.
Replacement player is just a measuring stick to gauge a players worth, it’s not meant as an illustration of a major league-minor league dynamic that some seem to think it is.
[/quote]
No, you don’t get it. A player’s worth to his team increases in a way that WAR does not account for when the average replacement player is not available. Two guys with the same WAR do not have the same value in reality if one guy has no viable alternative and the other has a competent replacement player down in the minors or sitting on the bench. So, if WAR cannot gauge that, what good is the statistic? What good is a stat that doesn’t even approach the reality of the situation? Isn’t that part of what stats do, quantify reality?
Moreso that you can’t really define what a replacement player is.
Also that it’s a totally hypothetical statistic…why would you base a statistic on someone who is NOT playing? That makes absolutely no sense. It’s like saying “Well, in an alternate universe, where your team isn’t made up of the same players…you have x less/more wins.” It tells you nothing. [/quote]
It’s basically a player who would cost no extra resources to acquire and would take the league minimum salary. Someone who has no competition to be signed. They look at roughly what value this type of player provides and determine what a replacement player is.
These guys are a dime a dozen.
[quote]Anonymity wrote:
But yes, also that it is cleary overly complicated - “If you can’t explain it simply, you don’t understand it well enough.”
[/quote]
I understand it. SOmething being complicated isn’t a reason to dismiss it.
No, you don’t get it. A player’s worth to his team increases in a way that WAR does not account for when the average replacement player is not available. Two guys with the same WAR do not have the same value in reality if one guy has no viable alternative and the other has a competent replacement player down in the minors or sitting on the bench. [/quote]
Yes, this is true.
[quote]DBCooper wrote:
So, if WAR cannot gauge that, what good is the statistic?[/quote]
To compare the value of different players across the league, determine how they’re performing.
[quote]Anonymity wrote:
Also Raj, I think that if WAR is just a measuring stick, then a more appropriate stat would be something like FIP-(league avg FIP).[/quote]
Not a whole lot would change except all the numbers would be lower
Moreso that you can’t really define what a replacement player is.
Also that it’s a totally hypothetical statistic…why would you base a statistic on someone who is NOT playing? That makes absolutely no sense. It’s like saying “Well, in an alternate universe, where your team isn’t made up of the same players…you have x less/more wins.” It tells you nothing. [/quote]
It’s basically a player who would cost no extra resources to acquire and would take the league minimum salary. Someone who has no competition to be signed. They look at roughly what value this type of player provides and determine what a replacement player is.
These guys are a dime a dozen.
[quote]Anonymity wrote:
But yes, also that it is cleary overly complicated - “If you can’t explain it simply, you don’t understand it well enough.”
[/quote]
I understand it. SOmething being complicated isn’t a reason to dismiss it.[/quote]
It’s a game played with a bat and a ball. It’s not complicated.
Moreso that you can’t really define what a replacement player is.
Also that it’s a totally hypothetical statistic…why would you base a statistic on someone who is NOT playing? That makes absolutely no sense. It’s like saying “Well, in an alternate universe, where your team isn’t made up of the same players…you have x less/more wins.” It tells you nothing. [/quote]
It’s basically a player who would cost no extra resources to acquire and would take the league minimum salary. Someone who has no competition to be signed. They look at roughly what value this type of player provides and determine what a replacement player is.
These guys are a dime a dozen.
[quote]Anonymity wrote:
But yes, also that it is cleary overly complicated - “If you can’t explain it simply, you don’t understand it well enough.”
[/quote]
I understand it. SOmething being complicated isn’t a reason to dismiss it.[/quote]
I wasn’t saying YOU personally don’t understand it. More that the people who created the stat don’t really understand baseball that well. I find it a looping, roundabout, overly-complex way to say “This guy is X units better than average”.
I think there are much easier ways to calculate such a statistic that don’t involve placing ficticious numbers on hypothetical players, and then basing the calculation off of that.
Moreso that you can’t really define what a replacement player is.
Also that it’s a totally hypothetical statistic…why would you base a statistic on someone who is NOT playing? That makes absolutely no sense. It’s like saying “Well, in an alternate universe, where your team isn’t made up of the same players…you have x less/more wins.” It tells you nothing. [/quote]
It’s basically a player who would cost no extra resources to acquire and would take the league minimum salary. Someone who has no competition to be signed. They look at roughly what value this type of player provides and determine what a replacement player is.
These guys are a dime a dozen.
[quote]Anonymity wrote:
But yes, also that it is cleary overly complicated - “If you can’t explain it simply, you don’t understand it well enough.”
[/quote]
I understand it. SOmething being complicated isn’t a reason to dismiss it.[/quote]
I wasn’t saying YOU personally don’t understand it. More that the people who created the stat don’t really understand baseball that well. I find it a looping, roundabout, overly-complex way to say “This guy is X units better than average”.
I think there are much easier ways to calculate such a statistic that don’t involve placing ficticious numbers on hypothetical players, and then basing the calculation off of that.
[/quote]
Right.
If you want to do the nitty gritty, compare a whole host of stats to build your impression of players x vs player y (2 major leaguers) then yeah WAR is not needed.
But WAR is essentially a bunch of different stats, weighted in their perceived value to performance and then presented in a nice pretty single number.
It’s a game played with a bat and a ball. It’s not complicated.[/quote]
Open a physics textbook some time.[/quote]
What does physics have to do with the game of baseball. The PHYSICS of baseball are certainly complicated. The GAME of baseball, which is what these convoluted sabermetrics attempt to quantify, is not as complicated as you’re making it out to be.
The guy wasn’t that good but we didn’t have anything anywhere close to being capable of replacing him, so he had some value to us.
The guy wasn’t that good and we had a similar caliber player in the minors at a third of the cost, so we went with the cheaper player. The other guy didn’t have much value to us.