Posted by oriolemagic on 2/20/2012 5:02:00 PM (view original):
If you insist on using earned runs as a measure, at least adjust for innings.  Use Pitching Runs stats you come up with..  

Sheldon - 45.72
Ducey - 45.31

Fraction of points off.  For those fractions I will take the 81 extra innings.

Posted by JFerg on 2/20/2012 2:48:00 PM (view original):
The formula I want to use is WAR, and you are right it is a big time formula.  Which is a good thing cause it's trying to account for everything it possibly can to evaluate a person on one number, we can argue all day.  You have to read some to understand where I'm coming from, I read what you read.  I use to love RBI's and Runs, and Batting Average and counting stats, but in the end it's not the best way to evaluate a player. 

Posted by MikeT23 on 2/20/2012 5:12:00 PM (view original):
You'll have to explain that one to me.

FWIW, I've already said I'm not impressed by 81 more innings.   I think I can find enough pitchers to cover that.

How much are you going to pay for 81 innings of that quality?  $6 million? $7 million?  $8 million?  I'd rather have that in one player and invest that money elsewhere in my team.

Posted by gjello10 on 2/20/2012 7:24:00 PM (view original):

Posted by JFerg on 2/20/2012 2:48:00 PM (view original):
The formula I want to use is WAR, and you are right it is a big time formula.  Which is a good thing cause it's trying to account for everything it possibly can to evaluate a person on one number, we can argue all day.  You have to read some to understand where I'm coming from, I read what you read.  I use to love RBI's and Runs, and Batting Average and counting stats, but in the end it's not the best way to evaluate a player. 

JFerg, you and I will agree all day long about the value of advanced metrics in evaluating real life players. 

But you're applying it to HBD in a way they shouldn't be applied.  I don't think this point has been driven home fully.  But we all know that splits are very important.  And, having not done a detailed enoungh survey of stats to say that I'm 100% sure, I am nonetheless pretty certain that in HBD pitchers and hitters have far more control over BABIP than they do in real life, and that splits are the mechanism.  For this reason, I think the real-life logic of FIP/xFIP fails to qualify here. 

Posted by oriolemagic on 2/20/2012 7:48:00 PM (view original):
Pitcher runs is simply a matter of determining the a pitchers ability above a replacement pitcher.  It takes the league average era and divides it by 9 to get a average runs allowed per inning.  Then subtract out the actual runs the pitcher gave up.  That is your base runs better then the league average, and multiply it by the number of innings a pitcher has thrown.  That gives you a correlation between the pitchers runs they gave up vs. the league average spread out over the innings they pitched.  It does count innings so you could say your counting the innings twice, but the point in showing this is that on runs and innings alone they are fractions apart.  Ducey wins in FIP by a long margin.  Start looking at all the stats and Ducey is the overall winner.  And for the person who wants W-L record, although I dont believe it means anything in this argument..

Ducey 25-8
Sheldon 20-3

Posted by oriolemagic on 2/20/2012 7:48:00 PM (view original):
Pitcher runs is simply a matter of determining the a pitchers ability above a replacement pitcher.  It takes the league average era and divides it by 9 to get a average runs allowed per inning.  Then subtract out the actual runs the pitcher gave up.  That is your base runs better then the league average, and multiply it by the number of innings a pitcher has thrown.  That gives you a correlation between the pitchers runs they gave up vs. the league average spread out over the innings they pitched.  It does count innings so you could say your counting the innings twice, but the point in showing this is that on runs and innings alone they are fractions apart.  Ducey wins in FIP by a long margin.  Start looking at all the stats and Ducey is the overall winner.  And for the person who wants W-L record, although I dont believe it means anything in this argument..

Ducey 25-8
Sheldon 20-3

Posted by jvford on 2/20/2012 9:25:00 PM (view original):

Posted by oriolemagic on 2/20/2012 7:48:00 PM (view original):
Pitcher runs is simply a matter of determining the a pitchers ability above a replacement pitcher.  It takes the league average era and divides it by 9 to get a average runs allowed per inning.  Then subtract out the actual runs the pitcher gave up.  That is your base runs better then the league average, and multiply it by the number of innings a pitcher has thrown.  That gives you a correlation between the pitchers runs they gave up vs. the league average spread out over the innings they pitched.  It does count innings so you could say your counting the innings twice, but the point in showing this is that on runs and innings alone they are fractions apart.  Ducey wins in FIP by a long margin.  Start looking at all the stats and Ducey is the overall winner.  And for the person who wants W-L record, although I dont believe it means anything in this argument..

Ducey 25-8
Sheldon 20-3

You don't seem to understand pitching runs. What your calculations show is that Sheldon plus 81 innings of replacement value is the same as Ducey. So, as long as you can find a SuB type to give you 81 innings at a little better than replacement value, Sheldon is clearly better.

And as others have said, forget FIP unless you have developed a modification for the fact that hitters and pitchers have more control over BABIP than in real life.

Posted by dedelman on 2/20/2012 9:37:00 PM (view original):

Posted by jvford on 2/20/2012 9:25:00 PM (view original):

Posted by oriolemagic on 2/20/2012 7:48:00 PM (view original):
Pitcher runs is simply a matter of determining the a pitchers ability above a replacement pitcher.  It takes the league average era and divides it by 9 to get a average runs allowed per inning.  Then subtract out the actual runs the pitcher gave up.  That is your base runs better then the league average, and multiply it by the number of innings a pitcher has thrown.  That gives you a correlation between the pitchers runs they gave up vs. the league average spread out over the innings they pitched.  It does count innings so you could say your counting the innings twice, but the point in showing this is that on runs and innings alone they are fractions apart.  Ducey wins in FIP by a long margin.  Start looking at all the stats and Ducey is the overall winner.  And for the person who wants W-L record, although I dont believe it means anything in this argument..

Ducey 25-8
Sheldon 20-3

You don't seem to understand pitching runs. What your calculations show is that Sheldon plus 81 innings of replacement value is the same as Ducey. So, as long as you can find a SuB type to give you 81 innings at a little better than replacement value, Sheldon is clearly better.

And as others have said, forget FIP unless you have developed a modification for the fact that hitters and pitchers have more control over BABIP than in real life.

jvford-- your point about the 81 innings of 4.5 ERA being the key difference is the most important thing about this discussion, but I disagree about the value of those innings.  Those are 81 innings at about half a run better than replacement value (in most worlds); that costs a million or two in free agency, and a roster spot. 

The precise reason that Ducey had the better season than Sheldon is that he enables you to use that roster spot, and that $1M, to buy a win or so somewhere else, a defensive C or 4th OF or defensive SS or lefty-mashing 1B or something.  Those are valuable innings, even at an ERA of 4.5.