White Sox

[Ender]

That is the first study I've seen where James came out in the top 2 in anything. It is almost always PECOTA and ZiPs with PECOTA in 1st for everything.

The one thing missing from that article btw is which version of stats they used. For example if you use PECOTA's book stats you aren't really getting the finished product. If you are using CHONE's first run you aren't getting as clear a picture. If you are using ZiPs without adjusting for off season moves which they don't update on the main pages you aren't getting a good picture. etc.

http://www.insidethebook.com/ee/index.p ... aluations/

That is a very interesting thread talking about projection systems (by the people who actually made them) if you are interested. For guys with 3+ years of experience nothing is much better than Marcels. It is really just the young guys that systems like PECOTA significantly improve on and on pitchers since ERA is the wrong stat to be judging projections by anyway.

[bigh0rt]

AVG- He was trending up. .277 in '04, .283 in '05 and .313 in '06. I know he isnt a .313 hitter, but I still saw no reason to have him below .300.

The guy has hit over .300 twice in his career. That wasn't reason enough for you to not expect him to hit .300? Hell, he bat .234 more recently than one of his .300 seasons, and .259 a year ago.

Using your method of only using the last three years to indicate expected performance for this season... we should expect to see Paul bat .285/.369/.525 with 35 HR, 101 RBI, and 89 R... correct? Or are we still using '04 - '06 as our magic 3 years?

[05worldserieschamps]

AVG- He was trending up. .277 in '04, .283 in '05 and .313 in '06. I know he isnt a .313 hitter, but I still saw no reason to have him below .300.

The guy has hit over .300 twice in his career. That wasn't reason enough for you to not expect him to hit .300? Hell, he bat .234 more recently than one of his .300 seasons, and .259 a year ago.

Using your method of only using the last three years to indicate expected performance for this season... we should expect to see Paul bat .285/.369/.525 with 35 HR, 101 RBI, and 89 R... correct? Or are we still using '04 - '06 as our magic 3 years?

Yes, Im still only using 04-06 as indicators. He is a different hitter now than he was in 2003. Thats just a fact. And no, I dont think he is going to hit .300 NOW. But at the beginning of last year, I thought he would. Again, reason being, his power numbers were very slightly declining but his average numbers were getting much better. Im not using his numbers pre-2004, so that first .300 season has minimal importance to me. And again, even though his AVG numbers were trending way up and by going by that alone his average should be been around .320. But I didnt expect that, I expected a slight regression, but certainly not .259. Maybe even in the .290+ range. Regardless to what all these projections were, the 04-06 years, even in OPS, he was trending up and quietly becoming a premiere hitter. He did not live up to that, or even close in 2007. He had a terribly off season. Case closed.

[GotowarMissAgnes]

Yes, Im still only using 04-06 as indicators. He is a different hitter now than he was in 2003. Thats just a fact. And no, I dont think he is going to hit .300 NOW. But at the beginning of last year, I thought he would. Again, reason being, his power numbers were very slightly declining but his average numbers were getting much better. Im not using his numbers pre-2004, so that first .300 season has minimal importance to me. And again, even though his AVG numbers were trending way up and by going by that alone his average should be been around .320. But I didnt expect that, I expected a slight regression, but certainly not .259. Maybe even in the .290+ range. Regardless to what all these projections were, the 04-06 years, even in OPS, he was trending up and quietly becoming a premiere hitter. He did not live up to that, or even close in 2007. He had a terribly off season. Case closed.

And that's the entire problem in a nutshell. You consistently ignore information that contradicts your opinion. In statistical terms, this is what is called "sample selection bias", and is a chief cause of the belief among many that "you can use statistics to prove anything you want". If you follow the general rule that EVERY DATA POINT IS IMPORTANT, then you prevent your bias from coloring your analysis AND you can't prove anything you want, because the data reveal the real story.

[bigh0rt]

The guy has hit over .300 twice in his career. That wasn't reason enough for you to not expect him to hit .300? Hell, he bat .234 more recently than one of his .300 seasons, and .259 a year ago.

Using your method of only using the last three years to indicate expected performance for this season... we should expect to see Paul bat .285/.369/.525 with 35 HR, 101 RBI, and 89 R... correct? Or are we still using '04 - '06 as our magic 3 years?

Yes, Im still only using 04-06 as indicators. He is a different hitter now than he was in 2003. Thats just a fact. And no, I dont think he is going to hit .300 NOW. But at the beginning of last year, I thought he would. Again, reason being, his power numbers were very slightly declining but his average numbers were getting much better. Im not using his numbers pre-2004, so that first .300 season has minimal importance to me. And again, even though his AVG numbers were trending way up and by going by that alone his average should be been around .320. But I didnt expect that, I expected a slight regression, but certainly not .259. Maybe even in the .290+ range. Regardless to what all these projections were, the 04-06 years, even in OPS, he was trending up and quietly becoming a premiere hitter. He did not live up to that, or even close in 2007. He had a terribly off season. Case closed.[/quote]
You're hysterical

So according to you, all that matters is the last three years, except for Paul Konerko, where last year doesn't count, so we'll use his best three years instead... yep, makes perfect sense to me. Wish I'd thought of that sooner; could've tried to pitch it to BBPro or Baseball America.

Paul Konerko for MVP. .300 AVG, 40 HR, 100 RBI -- move him up your rankings, folks. Albert who?

He is a different hitter now than he was in 2003.

Why is 2003 singled out? Hell, in 2002 he bat .300, a staple bench mark in your line of reasoning. He had hit .290 twice previous to 2003 also. You have to love the quoted sentence -- it's you in a nutshell. His RBI were 'trending upwards' up until 2003, as you'd put it -- and hell, he was in his 20's then. But I guess now that he's in his 30's, a "different hitter" and in a lineup that was the best offense in 2006, a career year only makes sense.

[Mayor Daley]

Unless the Sox get some decent pitching it will be the same thing as past year.
I see Crede and Uribe traded (before the season starts) for some proven starting pitchers.
(I have no idea who is desperate enough to want Uribe)
Replacing them - Fields at third, Alexei Ramirez at second.
Add two more GOOD starting pitchers to the current mix and you have a team that will contend.
As things stand right now..... Well, it's going to be a looong season.

[Old_Style]

Unless the Sox get some decent pitching it will be the same thing as past year.
I see Crede and Uribe traded (before the season starts) for some proven starting pitchers.
(I have no idea who is desperate enough to want Uribe)
Replacing them - Fields at third, Alexei Ramirez at second.
Add two more GOOD starting pitchers to the current mix and you have a team that will contend.
As things stand right now..... Well, it's going to be a looong season.

I don't think the Sox are going to get a quality, proven pitcher by trading Crede before the season starts.
I think they should start him at third until May and than ship him off. As of now, I think he has little trade value because of his back. Maybe spring training will be enough to up his value some.
As for Uribe, they'd be lucky to get a bag of balls for him.

[05worldserieschamps]

The guy has hit over .300 twice in his career. That wasn't reason enough for you to not expect him to hit .300? Hell, he bat .234 more recently than one of his .300 seasons, and .259 a year ago.

Using your method of only using the last three years to indicate expected performance for this season... we should expect to see Paul bat .285/.369/.525 with 35 HR, 101 RBI, and 89 R... correct? Or are we still using '04 - '06 as our magic 3 years?

Yes, Im still only using 04-06 as indicators. He is a different hitter now than he was in 2003. Thats just a fact. And no, I dont think he is going to hit .300 NOW. But at the beginning of last year, I thought he would. Again, reason being, his power numbers were very slightly declining but his average numbers were getting much better. Im not using his numbers pre-2004, so that first .300 season has minimal importance to me. And again, even though his AVG numbers were trending way up and by going by that alone his average should be been around .320. But I didnt expect that, I expected a slight regression, but certainly not .259. Maybe even in the .290+ range. Regardless to what all these projections were, the 04-06 years, even in OPS, he was trending up and quietly becoming a premiere hitter. He did not live up to that, or even close in 2007. He had a terribly off season. Case closed.

You're hysterical

So according to you, all that matters is the last three years, except for Paul Konerko, where last year doesn't count, so we'll use his best three years instead... yep, makes perfect sense to me. Wish I'd thought of that sooner; could've tried to pitch it to BBPro or Baseball America.

Paul Konerko for MVP. .300 AVG, 40 HR, 100 RBI -- move him up your rankings, folks. Albert who?

He is a different hitter now than he was in 2003.

Why is 2003 singled out? Hell, in 2002 he bat .300, a staple bench mark in your line of reasoning. He had hit .290 twice previous to 2003 also. You have to love the quoted sentence -- it's you in a nutshell. His RBI were 'trending upwards' up until 2003, as you'd put it -- and hell, he was in his 20's then. But I guess now that he's in his 30's, a "different hitter" and in a lineup that was the best offense in 2006, a career year only makes sense.[/quote]

1. Acording to me, the last 3 years are the only real years of importance when evaluating a player for the upcoming season, yes. Which makes perfect sense. 4 Years ago, a player is, well, 4 years younger(obviously). A 35 year olds stats at the end of his prime shouldnt be considered. They just arent a good indicator. Likewise, a player near his prime's stats shouldnt matter when he was 25-26. And once again, when a player is 27, his stats from his rookie season,when he was barely in the Majors shouldnt be held against him. In some cases, a player's stats are better 4 years ago in some cases, visa versa. Im not saying a player cant have an out of the blue year and relive(whether be good or bad) what happened 4 years ago. But a player shouldnt have his projections based on that.

2. No, Big H0rt. Im not using Paul Konerko's 3 best years. GotoWarMrsAgnes argued that Paul Konerko was achieving about what he should have in 2007, as it was right in line with career averages. Then I pointed out than from 2004-2006, his OPS was .894, .909 and .932. I simply pointed out that anyone that actually thought Konerko was going to go back to his career average of .848 was an idiot.(pre-2007) But of course, he had a terrible year and went .841. That happens. My entire point though, was that Konerko(along with others) had freak down years/injuries that caused the White Sox to lose as many games as they did. Im not saying they were good or would have won 90+ games, just that they should not have won 72 games.

3. Yes, in 2002 his average was good. But his OPS was one of his lowest. And, againm you're right that his RBI were trending up until 2003, but again, like 2007, HE HAD A DOWN YEAR. No one projected him to have that kind of year in 2003 just like no one projected him to have that kind of year in 2007. I'd love to see any projections that did. Thats my whole point! They were down years, and guess what? In 2004, he bounced back! Which is exactly what I expect him to do this time. Thats enough about Konerko. I think anyone in their right mind sees that Konerko just had an awful year and not to expect it again.

[05worldserieschamps]

Unless the Sox get some decent pitching it will be the same thing as past year.
I see Crede and Uribe traded (before the season starts) for some proven starting pitchers.
(I have no idea who is desperate enough to want Uribe)
Replacing them - Fields at third, Alexei Ramirez at second.
Add two more GOOD starting pitchers to the current mix and you have a team that will contend.
As things stand right now..... Well, it's going to be a looong season.

I think thats right for the most part and what Ive said all along. We need another pitcher. However, I dont think regardless of what else we add or dont add we will be as bad as last year. Just wont. I think right now they are a 78-80 win team at least. But like you said, they need more SP to get over that next huge step. And I love your idea about Crede. That makes perfect sense. (unfortunatly, you are correct about Uribe as well)

[GotowarMissAgnes]

1. Acording to me, the last 3 years are the only real years of importance when evaluating a player for the upcoming season, yes. Which makes perfect sense. 4 Years ago, a player is, well, 4 years younger(obviously). A 35 year olds stats at the end of his prime shouldnt be considered. They just arent a good indicator. Likewise, a player near his prime's stats shouldnt matter when he was 25-26. And once again, when a player is 27, his stats from his rookie season,when he was barely in the Majors shouldnt be held against him. In some cases, a player's stats are better 4 years ago in some cases, visa versa. Im not saying a player cant have an out of the blue year and relive(whether be good or bad) what happened 4 years ago. But a player shouldnt have his projections based on that.

So, tell us. Why should we buy your view, when it is directly contradictory to every one of the good projection systems. ZiPS uses a minimum of 4 years. PECOTA uses not only an entire player's career, but the entire career information on closely comparable players, and it's the best around. Every good projection system uses at least 4 years, and often more than that.

But, you want we should use your method.

2. No, Big H0rt. Im not using Paul Konerko's 3 best years. GotoWarMrsAgnes argued that Paul Konerko was achieving about what he should have in 2007, as it was right in line with career averages. Then I pointed out than from 2004-2006, his OPS was .894, .909 and .932. I simply pointed out that anyone that actually thought Konerko was going to go back to his career average of .848 was an idiot.(pre-2007) But of course, he had a terrible year and went .841. That happens. My entire point though, was that Konerko(along with others) had freak down years/injuries that caused the White Sox to lose as many games as they did. Im not saying they were good or would have won 90+ games, just that they should not have won 72 games.

While no one predicted Konerko to fall that far, many predicted a fall to about .875 or so, including James and PECOTA. And, the good projection system do not say silly things like "oh, that was a freak down year." and do silly things like throw that important piece of evidence out the window. Instead, they update their model to reflect the fact that Paul Konerko has given evidence that he can have that bad of a year. And that's why pretty much no one is predicting Konerko to rebound the way you are. All of the systems have adjusted their expectations downward, and have Konerko in the .860-.870 range for OPS--a rebound, but a much milder one than you seem to think will happen.

3. Yes, in 2002 his average was good. But his OPS was one of his lowest. And, againm you're right that his RBI were trending up until 2003, but again, like 2007, HE HAD A DOWN YEAR. No one projected him to have that kind of year in 2003 just like no one projected him to have that kind of year in 2007. I'd love to see any projections that did. Thats my whole point! They were down years, and guess what? In 2004, he bounced back! Which is exactly what I expect him to do this time. Thats enough about Konerko. I think anyone in their right mind sees that Konerko just had an awful year and not to expect it again.

No one projected him to fall as far as he did in 2007, but almost every system projected a decline. ZiPS projected a 30 point decline in OPS for Konerko in 2007, and James and PECOTA projected a 50-60 point decline in OPS for Konerko in 2007.

So, once again, you're wrong. Despite the fact Konerko's OPS had risen 3 years in a row. EVERY good projection system projected a fairly large decline in his OPS in 2007. And every one of them projects only a small rebound this year (ZiPS, in fact, predicts another decline). Maybe, just maybe, these systems and the guys that made them know a little bit more about this than you do.