All Forums > SimLeague Baseball > MLB > Wins and Losses
12/6/2012 12:54 PM
Posted by burnsy483 on 12/6/2012 12:48:00 PM (view original):
You asked me what pitching to the score, managing games, etc has to do with w-l record. I gave you a scenario, which you understood and accepted. It's an example that shows how you can have a better w-l record, getting your team more wins, and having a higher ERA.

Is games started or w-l record a more useful statistic?
No, I asked you how W/L record reflects a pitcher pitching to the score. Your scenario didn't show that. Having more wins alone doesn't mean that a pitcher managed the game better or pitched to the score.

If a pitcher wins 17 games but his team only wins 19 of his starts is that good? Is it better than when a pitcher gets 14 wins but his team wins 24 of his starts?


12/6/2012 1:05 PM
Posted by burnsy483 on 12/6/2012 12:53:00 PM (view original):
Posted by bad_luck on 12/6/2012 12:46:00 PM (view original):
Regarding W/L and GS, neither is useful on its own.

With other info, IP(nobody uses GS for anything other than establishing that a guy was a starter) is more important.
Ok. So you value IP as a useful statistic.

What's more likely - the 220 IP pitcher being better than the 180 IP pitcher, or, the 20-8 pitcher being better than the 16-12 pitcher?
Knowing innings alone is worth less than knowing only that a guy won 20 games. Obviously, great pitchers usually win a lot of games. Nobody is denying that.

My entire argument is that W/L record adds nothing to the stat line. At best it's useless, at worst it's misleading. See: 2012 Barry Zito, 1992 Jack Morris, 1989 Storm Davis.


12/6/2012 1:19 PM
Posted by bad_luck on 12/6/2012 12:54:00 PM (view original):
Posted by burnsy483 on 12/6/2012 12:48:00 PM (view original):
You asked me what pitching to the score, managing games, etc has to do with w-l record. I gave you a scenario, which you understood and accepted. It's an example that shows how you can have a better w-l record, getting your team more wins, and having a higher ERA.

Is games started or w-l record a more useful statistic?
No, I asked you how W/L record reflects a pitcher pitching to the score. Your scenario didn't show that. Having more wins alone doesn't mean that a pitcher managed the game better or pitched to the score.

If a pitcher wins 17 games but his team only wins 19 of his starts is that good? Is it better than when a pitcher gets 14 wins but his team wins 24 of his starts?


What's the context?

Did the 17 win guy, whose team won only 19 of his starts, play for a crappy team?  Did the 14 win guy, whose team won 24 of his starts, play on a really good team that bailed him out after he left the game in a tie or behind?

You keep debating the merit of pitcher's wins and losses in a vacuum and arguing that they are meaningless because you can't deduce anything from one statistic alone.

Virtually everybody is agreeing with that.  You need to look at the entire picture, which includes many statistics as well as context.

Yet you keep insisting that wins/losses HAS ABSOLUTELY NO VALUE, and should be COMPLETELY DISREGARDED.  Which is absolutely not true.

You've been given numerous examples that show that there is a strong and irrefutable correlation between pitcher's wins/losses and a pitcher's success (or lack of).  Yet you continue to flat out deny that such a correlation exists, despite the mountains of evidence presented and available to you.
12/6/2012 1:25 PM
W/L record doesn't add any context.

It doesn't tell you if a guy pitched well. If he played on a good team. If he managed games well or pitched to the score. If he got good or bad run support.

I'm not denying that there is a correlation between good pitchers and lots of wins. I'm saying that the W/L record adds nothing.
12/6/2012 1:29 PM
Posted by bad_luck on 12/6/2012 1:05:00 PM (view original):
Posted by burnsy483 on 12/6/2012 12:53:00 PM (view original):
Posted by bad_luck on 12/6/2012 12:46:00 PM (view original):
Regarding W/L and GS, neither is useful on its own.

With other info, IP(nobody uses GS for anything other than establishing that a guy was a starter) is more important.
Ok. So you value IP as a useful statistic.

What's more likely - the 220 IP pitcher being better than the 180 IP pitcher, or, the 20-8 pitcher being better than the 16-12 pitcher?
Knowing innings alone is worth less than knowing only that a guy won 20 games. Obviously, great pitchers usually win a lot of games. Nobody is denying that.

My entire argument is that W/L record adds nothing to the stat line. At best it's useless, at worst it's misleading. See: 2012 Barry Zito, 1992 Jack Morris, 1989 Storm Davis.


"Nobody is denying that."

You've been denying that for 18 pages.  "W-L record is meaningless" has been said by you numerous times.  If "obviously, great pitchers usually win a lot of games" was understood by you, you wouldn't be telling us how meaningless wins are.  You've said "a 15-8 record tells me nothing."  It's you who have been saying that.

Do you want me to show you a list of players with misleading ERAs? There are tons of them.  If I told you that a pitcher in 1996 went 13-9 with a 3.03 ERA, in 200+ innings, you'd tell me that he actually deserved a better W-L record, just knowing this information.  In fact, Steve Trachsel allowed a ton of home runs, a lot of unearned runs, and stuck out next to nobody.  His ERA should have been much higher.

It is not smart to look at any statistic and assume that you can be 100% certain about his performance.  ERA is a better statistic overall than wins are, but wins are part of the equation, it's part of the story.  Steve Carlton won 27 games for a terrible Phillies team.  I've read actual stories about his season, about how much more confident the team was behind him in games he pitched, about how much better they hit because of it, of how much better they played in the field because of it.  
12/6/2012 1:30 PM
Posted by bad_luck on 12/6/2012 12:54:00 PM (view original):
Posted by burnsy483 on 12/6/2012 12:48:00 PM (view original):
You asked me what pitching to the score, managing games, etc has to do with w-l record. I gave you a scenario, which you understood and accepted. It's an example that shows how you can have a better w-l record, getting your team more wins, and having a higher ERA.

Is games started or w-l record a more useful statistic?
No, I asked you how W/L record reflects a pitcher pitching to the score. Your scenario didn't show that. Having more wins alone doesn't mean that a pitcher managed the game better or pitched to the score.

If a pitcher wins 17 games but his team only wins 19 of his starts is that good? Is it better than when a pitcher gets 14 wins but his team wins 24 of his starts?


The scenario gives an example about how it's possible.  A pitcher who pitches to put his team in the best situation to win is more likely to win that the pitcher who just pitches the same way always.
12/6/2012 1:40 PM
"Knowing innings alone is worth less than knowing only that a guy won 20 games."

It's interesting that knowing how many games someone won is "COMPLETELY WORTHLESS" when there are stats about a pitcher that are "worth less" than wins.  What does that make IP as a statistic?
12/6/2012 1:46 PM
Posted by burnsy483 on 12/6/2012 1:29:00 PM (view original):
Posted by bad_luck on 12/6/2012 1:05:00 PM (view original):
Posted by burnsy483 on 12/6/2012 12:53:00 PM (view original):
Posted by bad_luck on 12/6/2012 12:46:00 PM (view original):
Regarding W/L and GS, neither is useful on its own.

With other info, IP(nobody uses GS for anything other than establishing that a guy was a starter) is more important.
Ok. So you value IP as a useful statistic.

What's more likely - the 220 IP pitcher being better than the 180 IP pitcher, or, the 20-8 pitcher being better than the 16-12 pitcher?
Knowing innings alone is worth less than knowing only that a guy won 20 games. Obviously, great pitchers usually win a lot of games. Nobody is denying that.

My entire argument is that W/L record adds nothing to the stat line. At best it's useless, at worst it's misleading. See: 2012 Barry Zito, 1992 Jack Morris, 1989 Storm Davis.


"Nobody is denying that."

You've been denying that for 18 pages.  "W-L record is meaningless" has been said by you numerous times.  If "obviously, great pitchers usually win a lot of games" was understood by you, you wouldn't be telling us how meaningless wins are.  You've said "a 15-8 record tells me nothing."  It's you who have been saying that.

Do you want me to show you a list of players with misleading ERAs? There are tons of them.  If I told you that a pitcher in 1996 went 13-9 with a 3.03 ERA, in 200+ innings, you'd tell me that he actually deserved a better W-L record, just knowing this information.  In fact, Steve Trachsel allowed a ton of home runs, a lot of unearned runs, and stuck out next to nobody.  His ERA should have been much higher.

It is not smart to look at any statistic and assume that you can be 100% certain about his performance.  ERA is a better statistic overall than wins are, but wins are part of the equation, it's part of the story.  Steve Carlton won 27 games for a terrible Phillies team.  I've read actual stories about his season, about how much more confident the team was behind him in games he pitched, about how much better they hit because of it, of how much better they played in the field because of it.  
I've never denied that good pitchers get wins. I'm saying the W/L record adds nothing.

Knowing that Trachsel went 13-9 doesn't tell me that he should have had a higher ERA. All those other things you mentioned (K rate, HR rate, etc) do.




12/6/2012 1:53 PM
You're saying that you know good pitchers get wins, but knowing a win total tells you nothing.

...

OK.


Re: Trachsel: You seem to like ERA as a stat, it "tells you something." In this example, it tells you nothing.  There are other examples of this, just like there are examples of a pitcher's w-l record being misleading.  And I'm not asking you to try to determine if Trachsel should have had a higher ERA based on his W-L record.
12/6/2012 1:55 PM
I'd also love if you could answer this question:

It's interesting that knowing how many games someone won is "COMPLETELY WORTHLESS" when there are stats about a pitcher that are "worth less" than wins.  What does that make IP as a statistic?
12/6/2012 1:57 PM
Good pitchers get wins. But so do bad pitchers. And sometimes pitchers pitch great and don't get the win.

In your Trachsel example, what does his 13-9 record tell you? I wouldn't say he deserved a better W/L record, I would ignore his W/L record because it's useless. 
12/6/2012 1:57 PM
Any statistic in a vacuum tells you nothing.

Example 1: a guy has a WHIP of 1.00.  Is he a good pitcher?

Example 2: a guy has an ERA of 2.50.  Is he a good pitcher?
12/6/2012 1:57 PM
Posted by burnsy483 on 12/6/2012 1:55:00 PM (view original):
I'd also love if you could answer this question:

It's interesting that knowing how many games someone won is "COMPLETELY WORTHLESS" when there are stats about a pitcher that are "worth less" than wins.  What does that make IP as a statistic?
If you want me to answer question, making 6 posts in a row after asking it isn't the best way to do that.
12/6/2012 2:00 PM
Posted by bad_luck on 12/6/2012 1:57:00 PM (view original):
Good pitchers get wins. But so do bad pitchers. And sometimes pitchers pitch great and don't get the win.

In your Trachsel example, what does his 13-9 record tell you? I wouldn't say he deserved a better W/L record, I would ignore his W/L record because it's useless. 
You forgot to also say this:

"Good pitchers get good ERAs.  So do bad pitchers.  And sometimes pitchers pitch great and have bad ERAs.

In your Trachsel example, what does his 3.03 ERA tell me?  I wouldn't say he deserved a worse ERA, I would ignore his ERA because it's useless."
12/6/2012 2:02 PM
Posted by tecwrg on 12/6/2012 1:57:00 PM (view original):
Any statistic in a vacuum tells you nothing.

Example 1: a guy has a WHIP of 1.00.  Is he a good pitcher?

Example 2: a guy has an ERA of 2.50.  Is he a good pitcher?
I agree with that.

But I stand by my point that W/L record adds nothing. Pedro Martinez was 219-100 over his career. He was a great pitcher. Is he less great if he went 185-105, while all his other stats stay exactly* the same? Is he suddenly better if his record shifts to 235-93 and all his other stats stay exactly the same?

*by exactly I don't just mean cumulative ERA, Whip, etc, but literally everything. Pitch sequences, innings...nothing else changes.
of 28
All Forums > SimLeague Baseball > MLB > Wins and Losses

Terms of Use Customer Support Privacy Statement

Popular on WhatIfSports site: Baseball Simulation | College Basketball Game | College Football Game | Online Baseball Game | Hockey Simulation | NFL Picks | College Football Picks | Sports Games

© 1999-2014 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.