Well, i wasn't planning on writing anything today, but it appears we have some time in between football games and I decided on dropping some content on you guys.
Today I want to talk about and demonstrate why I decided to use Baseball-Reference's pitching WAR instead of the FanGraphs version, just to clarify, I'm a big fan of their writers, Cistulli always cracks me up in his podcast and overall really enjoy their work, but to me this is very debatable at the very least.
While both sites use very similar formulas there's one big diference, FG uses FIP, BR uses ERA. This alone changes everything, before i make my point, a quick rundown of ehat is WAR.
WAR stands for Wins Above Replacement, which basically means, that number is the amount of wins that player provides you in relation to the number of wins you'd get from a replacement level AAAA player, when considering that you cannot tell me it's plausible to use a stat (FIP) that isolates runs from the long ball and completely disregards every other run scored, when in the game a wild pitch with a runner on 3rd base scoring and a solo shot means the same, any average Joe knows the responsibility of a pitcher on the long ball its far bigger than on a wild pitch, sac fly, single that could be caught, but that doesn't matter when counting for a stat that measures added value.
Let me give you an example on my NL Cy Young post I started to really notice this, more people every year are turning towards WAR when talking about awards, and even some voters also, which is fine but not fine when looking at a FIP based version, you just cannot do it. I'll give you an example, I recently compared Elsbury and Cain and their stats when entering Free Agency, now I am going to compare two pitchers and their numbers this year.
First One
K,% 24.2. BB% 3.8. K-BB% 20.4
AVG .252. BABIP .303. LOB% 67.5
ERA 4.42 FIP 3.61 HR/FB 13.8%
WAR 3.8
Second One
K% 22.7. BB% 9.6. K/BB% 13.2
AVG .213. BABIP .258. LOB% 81.6
ERA 2.96 FIP 3.93. HR/FB 11.1 %
WAR 3.3
The takeaway from this is easy, for two talented pitchers, one got really unlucky, the second one got very lucky, and if you had to take the over under on either one of those you'd have to take the first pitcher to bounce back and the second one to decline some, it doesn't take a genius to know who added more value/wins to his team, not even close, despite that the first pitcher had a higher WAR, with only 6.2 more innings, 207.2 to 201, both started 32 games.
By the way the first one is Jeff Samardzija and the second one Gio Gonzalez, Gio finished sixth on the Cy Young vote, Jeff was a huge disappointment and looking at the BR's WAR version you see Gonzalez at 6.6, and Samardzija at 2.4, I could give you other examples but it would be redundant.
While I know FG's WAR has some qualities it goes against what WAR was designed to represent.
Bottom line is: Luck affects results, results determine added value, WAR is added value, therefore Luck must affect WAR, and FIP eliminates luck
Sorry for the overuse of I, Big Narcissist.
Today I want to talk about and demonstrate why I decided to use Baseball-Reference's pitching WAR instead of the FanGraphs version, just to clarify, I'm a big fan of their writers, Cistulli always cracks me up in his podcast and overall really enjoy their work, but to me this is very debatable at the very least.
While both sites use very similar formulas there's one big diference, FG uses FIP, BR uses ERA. This alone changes everything, before i make my point, a quick rundown of ehat is WAR.
WAR stands for Wins Above Replacement, which basically means, that number is the amount of wins that player provides you in relation to the number of wins you'd get from a replacement level AAAA player, when considering that you cannot tell me it's plausible to use a stat (FIP) that isolates runs from the long ball and completely disregards every other run scored, when in the game a wild pitch with a runner on 3rd base scoring and a solo shot means the same, any average Joe knows the responsibility of a pitcher on the long ball its far bigger than on a wild pitch, sac fly, single that could be caught, but that doesn't matter when counting for a stat that measures added value.
Let me give you an example on my NL Cy Young post I started to really notice this, more people every year are turning towards WAR when talking about awards, and even some voters also, which is fine but not fine when looking at a FIP based version, you just cannot do it. I'll give you an example, I recently compared Elsbury and Cain and their stats when entering Free Agency, now I am going to compare two pitchers and their numbers this year.
First One
K,% 24.2. BB% 3.8. K-BB% 20.4
AVG .252. BABIP .303. LOB% 67.5
ERA 4.42 FIP 3.61 HR/FB 13.8%
WAR 3.8
Second One
K% 22.7. BB% 9.6. K/BB% 13.2
AVG .213. BABIP .258. LOB% 81.6
ERA 2.96 FIP 3.93. HR/FB 11.1 %
WAR 3.3
The takeaway from this is easy, for two talented pitchers, one got really unlucky, the second one got very lucky, and if you had to take the over under on either one of those you'd have to take the first pitcher to bounce back and the second one to decline some, it doesn't take a genius to know who added more value/wins to his team, not even close, despite that the first pitcher had a higher WAR, with only 6.2 more innings, 207.2 to 201, both started 32 games.
By the way the first one is Jeff Samardzija and the second one Gio Gonzalez, Gio finished sixth on the Cy Young vote, Jeff was a huge disappointment and looking at the BR's WAR version you see Gonzalez at 6.6, and Samardzija at 2.4, I could give you other examples but it would be redundant.
While I know FG's WAR has some qualities it goes against what WAR was designed to represent.
Bottom line is: Luck affects results, results determine added value, WAR is added value, therefore Luck must affect WAR, and FIP eliminates luck
Sorry for the overuse of I, Big Narcissist.
No comments:
Post a Comment