Zambrano 0-4 since signing

[Soul]

Yup lets lay all the blame on Z while nobody is talking about this pathetic excuse for an offense that our GM has assembled. Guess what guys, if Z goes 9 innings and gives up 2 runs, we still lose this damn game.

 

Runs scored/Game

 

Cubs 4.60

NL Ave 4.64

 

average? yes

sub-par? maybe

pathetic? unjustifiable.

 

Wrong.

 

Keener's not referring to season totals. It's recently that matters, because that's when we stopped winning. In August the Cubs have scored 112 runs, that's 15th in the N.L.

 

The offense is pathetic, period.

[davearm]

Yup lets lay all the blame on Z while nobody is talking about this pathetic excuse for an offense that our GM has assembled. Guess what guys, if Z goes 9 innings and gives up 2 runs, we still lose this damn game.

 

Runs scored/Game

 

Cubs 4.60

NL Ave 4.64

 

average? yes

sub-par? maybe

pathetic? unjustifiable.

 

Wrong.

 

Keener's not referring to season totals. It's recently that matters, because that's when we stopped winning. In August the Cubs have scored 112 runs, that's 15th in the N.L.

 

The offense is pathetic, period.

Keener was griping about the offense "that our GM has assembled."

 

Hendry assembled this offense in late July?

[Mephistopheles]

So Zambrano has been less effective, yet better?

 

I think we're going to need another longwinded rambling paragraph out of you to explain this one.

 

people got it.

 

Zambrano is not that good of a hitter.

 

you are incorrect. he certainly is. run through the numbers if you'd like.

 

and let me provide this link viewtopic.php?t=43717

So your argument is that Zambrano's hitting makes up for the pitching deficit, huh?

 

Well I took you up on your invitation to run through the numbers.

 

Zambrano has produced 11 runs (8 runs + 5 RBI - 2 HR = 11 RP).

Lilly has produced 9 runs (5 runs + 4 RBI - 0 HR = 9 RP).

Hill has produced 5 runs (0 runs + 5 RBI - 0 HR = 5 RP).

 

So let's just net those runs produced with the bat off of each guy's earned runs allowed on the mound and see what we get.

 

Lopping off 11 earned runs drops Zambrano's ERA from 4.06 to 3.51.

With 9 fewer ER, Lilly's ERA falls from 3.85 to 3.16.

5 fewer ER nudges Hill's ERA from 3.68 to 3.06.

 

Some of us here are smart and don't use RBI and R to gauge offensive production.

[Flames24Rulz]

Yup lets lay all the blame on Z while nobody is talking about this pathetic excuse for an offense that our GM has assembled. Guess what guys, if Z goes 9 innings and gives up 2 runs, we still lose this damn game.

 

Runs scored/Game

 

Cubs 4.60

NL Ave 4.64

 

average? yes

sub-par? maybe

pathetic? unjustifiable.

 

Wrong.

 

Keener's not referring to season totals. It's recently that matters, because that's when we stopped winning. In August the Cubs have scored 112 runs, that's 15th in the N.L.

 

The offense is pathetic, period.

Keener was griping about the offense "that our GM has assembled."

 

Hendry assembled this offense in late July?

 

More or less, the offense has been pathetic the whole season. When the offense was good in April, the bullpen was horrid. The offense has been about as pathetic as it can possibly be, ever since the All-Star Break.

[Bull]

So Zambrano has been less effective, yet better?

 

I think we're going to need another longwinded rambling paragraph out of you to explain this one.

 

people got it.

 

Zambrano is not that good of a hitter.

 

you are incorrect. he certainly is. run through the numbers if you'd like.

 

and let me provide this link viewtopic.php?t=43717

So your argument is that Zambrano's hitting makes up for the pitching deficit, huh?

 

Well I took you up on your invitation to run through the numbers.

 

Zambrano has produced 11 runs (8 runs + 5 RBI - 2 HR = 11 RP).

Lilly has produced 9 runs (5 runs + 4 RBI - 0 HR = 9 RP).

Hill has produced 5 runs (0 runs + 5 RBI - 0 HR = 5 RP).

 

So let's just net those runs produced with the bat off of each guy's earned runs allowed on the mound and see what we get.

 

Lopping off 11 earned runs drops Zambrano's ERA from 4.06 to 3.51.

With 9 fewer ER, Lilly's ERA falls from 3.85 to 3.16.

5 fewer ER nudges Hill's ERA from 3.68 to 3.06.

 

Some of us here are smart and don't use RBI and R to gauge offensive production.

 

we have rules of debate here. Please review the guidelines.

 

you can attack arguments, but don't question another poster's intelligence.

 

You have some insight and understanding of statistical data (though not as much as you think), but your approach makes your posts unreadable. If you would exercise some sense of human decency, I think you could be a valuable member of this community.

[davearm]

Yup lets lay all the blame on Z while nobody is talking about this pathetic excuse for an offense that our GM has assembled. Guess what guys, if Z goes 9 innings and gives up 2 runs, we still lose this damn game.

 

Runs scored/Game

 

Cubs 4.60

NL Ave 4.64

 

average? yes

sub-par? maybe

pathetic? unjustifiable.

 

Wrong.

 

Keener's not referring to season totals. It's recently that matters, because that's when we stopped winning. In August the Cubs have scored 112 runs, that's 15th in the N.L.

 

The offense is pathetic, period.

Keener was griping about the offense "that our GM has assembled."

 

Hendry assembled this offense in late July?

 

More or less, the offense has been pathetic the whole season. When the offense was good in April, the bullpen was horrid. The offense has been about as pathetic as it can possibly be, ever since the All-Star Break.

The statistics shown above disprove this conclusion: 4.60 RPG against an NL average of 4.64 is not pathetic, it's average.

[Mephistopheles]

Actually they don't. If you want to disprove your statement you also have to show that the offense is not prone to more low scoring games than the average offense (in other words that they're not scoring a lot on bad pitchers late. running up the score so to speak in blowouts).

 

Also you didn't consider park effects. You didn't consider strength of schedule (or lack thereof).

 

In fact, something that does include most of these factors (EqA) argues that the Cubs offense has been pathetic most of the season, posting an alarming .252 EqA which is good for 13th in the league. It, however, doesn't prove this either. It's worth pointing out. Now respond to my other points.

[Bull]

Now respond to my other points.

 

I'm here to teach.

 

Most people don't respond well to orders. At least not people with dignity.

[Ender]

The statistics shown above disprove this conclusion: 4.60 RPG against an NL average of 4.64 is not pathetic, it's average.

 

They also play in a hitters park in a division with horrible pitching. Not to mention actual runs scored has a ton of noise involved, you'd be better off looking at something like RC or even OPS. The Cubs have scored more runs than they probably should have on the season.

 

Same thing with the pitching, ERA is a horrible stat to use to judge pitchers by because of how much statistical noise and variance is involved in the stat.

[jaydee]

Yup lets lay all the blame on Z while nobody is talking about this pathetic excuse for an offense that our GM has assembled. Guess what guys, if Z goes 9 innings and gives up 2 runs, we still lose this damn game.

 

Runs scored/Game

 

Cubs 4.60

NL Ave 4.64

 

average? yes

sub-par? maybe

pathetic? unjustifiable.

 

Wrong.

 

Keener's not referring to season totals. It's recently that matters, because that's when we stopped winning. In August the Cubs have scored 112 runs, that's 15th in the N.L.

 

The offense is pathetic, period.

 

And Zambrano has been just as pathetic.

[Bull]

The statistics shown above disprove this conclusion: 4.60 RPG against an NL average of 4.64 is not pathetic, it's average.

 

They also play in a hitters park in a division with horrible pitching. Not to mention actual runs scored has a ton of noise involved, you'd be better off looking at something like RC or even OPS. The Cubs have scored more runs than they probably should have on the season.

 

Same thing with the pitching, ERA is a horrible stat to use to judge pitchers by because of how much statistical noise and variance is involved in the stat.

 

Every stat has value (well maybe except W-L record for a pitcher).

 

Despite the fact they "should have scored less run", it doesn't change the fact that they did score those runs. Somehow you have to account for that.

 

If runs scored is worthless then you're saying that pythagorean record as a predictor is worthless. Do you really believe this?

[Bull]

Actually they don't. If you want to disprove your statement you also have to show that the offense is not prone to more low scoring games than the average offense (in other words that they're not scoring a lot on bad pitchers late. running up the score so to speak in blowouts).

 

Also you didn't consider park effects. You didn't consider strength of schedule (or lack thereof).

 

In fact, something that does include most of these factors (EqA) argues that the Cubs offense has been pathetic most of the season, posting an alarming .252 EqA which is good for 13th in the league. It, however, doesn't prove this either. It's worth pointing out. Now respond to my other points.

 

So runs scored against bad pitchers are less valuable than runs scored against good ones? Runs scored against bad teams (strength of schedule) don't count for as much as those against elite teams?

 

If you use this as a statement to that effect, you'd have to break down what runs were scored in what kids of games. Your sample size gets smaller and smaller. Now it sounds like you're valuing "close and late stats" Does this mean you believe in "clutch"?

[haltz]

If runs scored is worthless then you're saying that pythagorean record as a predictor is worthless. Do you really believe this?

I don't know about statistical noise. ERA gives you exactly the target value that it purports to. That happens to be a poor estimation of true talent and how a pitcher has pitched. It gets better the larger the sample.

[Mephistopheles]

Actually they don't. If you want to disprove your statement you also have to show that the offense is not prone to more low scoring games than the average offense (in other words that they're not scoring a lot on bad pitchers late. running up the score so to speak in blowouts).

 

Also you didn't consider park effects. You didn't consider strength of schedule (or lack thereof).

 

In fact, something that does include most of these factors (EqA) argues that the Cubs offense has been pathetic most of the season, posting an alarming .252 EqA which is good for 13th in the league. It, however, doesn't prove this either. It's worth pointing out. Now respond to my other points.

 

So runs scored against bad pitchers are less valuable than runs scored against good ones? Runs scored against bad teams (strength of schedule) don't count for as much as those against elite teams?

 

If you use this as a statement to that effect, you'd have to break down what runs were scored in what kids of games. Your sample size gets smaller and smaller. Now it sounds like you're valuing "close and late stats" Does this mean you believe in "clutch"?

 

No you're missing that. If the Cubs schedule was very easy and faced a lot of bad pitching staffs, wouldn't we expect the Cubs offensive output to increase because of this, in runs? Thats all I was saying. A run scored against a bad pitcher is easier than a run scored off of a good pitcher. It's relative. And no, it's not clutch, it's called being able to hit good pitching. A good offense is one that can hit all types of pitching at a better than average clip (not better than league average clip, better than the leagues average off that pitching).

Edited August 30, 2007 by Mephistopheles

[CubinNY]

If runs scored is worthless then you're saying that pythagorean record as a predictor is worthless. Do you really believe this?

I don't know about statistical noise. ERA gives you exactly the target value that it purports to. That happens to be a poor estimation of true talent and how a pitcher has pitched. It gets better the larger the sample.

define "true talent"

[Mephistopheles]

If runs scored is worthless then you're saying that pythagorean record as a predictor is worthless. Do you really believe this?

I don't know about statistical noise. ERA gives you exactly the target value that it purports to. That happens to be a poor estimation of true talent and how a pitcher has pitched. It gets better the larger the sample.

define "true talent"

 

He shouldn't have to define "true talent." You know damn well what he means.

[CubinNY]

If runs scored is worthless then you're saying that pythagorean record as a predictor is worthless. Do you really believe this?

I don't know about statistical noise. ERA gives you exactly the target value that it purports to. That happens to be a poor estimation of true talent and how a pitcher has pitched. It gets better the larger the sample.

define "true talent"

 

He shouldn't have to define "true talent." You know damn well what he means.

No I don't.

[Fred Hornkohl]

If runs scored is worthless then you're saying that pythagorean record as a predictor is worthless. Do you really believe this?

I don't know about statistical noise. ERA gives you exactly the target value that it purports to. That happens to be a poor estimation of true talent and how a pitcher has pitched. It gets better the larger the sample.

define "true talent"

 

He shouldn't have to define "true talent." You know damn well what he means.

 

It depends on what the meaning of is is. :wink:

[haltz]

define "true talent"

A crystal ball projection. Or I should say an "accurate" projection.

[Mephistopheles]

If runs scored is worthless then you're saying that pythagorean record as a predictor is worthless. Do you really believe this?

I don't know about statistical noise. ERA gives you exactly the target value that it purports to. That happens to be a poor estimation of true talent and how a pitcher has pitched. It gets better the larger the sample.

define "true talent"

 

He shouldn't have to define "true talent." You know damn well what he means.

No I don't.

 

how good they really are.

 

 

see darrell may 2003. his true talent was a 5.00 pitcher but his era was below 4.00

[CubinNY]

If runs scored is worthless then you're saying that pythagorean record as a predictor is worthless. Do you really believe this?

I don't know about statistical noise. ERA gives you exactly the target value that it purports to. That happens to be a poor estimation of true talent and how a pitcher has pitched. It gets better the larger the sample.

define "true talent"

 

He shouldn't have to define "true talent." You know damn well what he means.

No I don't.

 

how good they really are.

 

 

see darrell may 2003. his true talent was a 5.00 pitcher but his era was below 4.00

How do you define how good they really are? There has to be some objective anchor. Words like "ture" are supposed to reflect reality not a statistical model of the way the world is supposed to work. In other words, it is not reality that is wrong. Reality is what really happend. The trick is getting a model that is relatively accurate in relation to the "true value" of something you are observing.

 

ERA is a poor predictor and is loaded with "noise" but it can also be an objective anchor for how "good" a pitcher has been.

 

Every statistic becomes a better predictor with more data.

[soccer10k]

so zambrano's hitting coupled with hill's inability to bunt makes up the difference between their pitching numbers? i have a hard time buying that.

 

VORP

 

38.1 - Carlos Zambrano (31.4 + 6.7)

34.6 - Rich Hill (35.3 - 0.7)

33.9 - Ted Lilly (35.6 - 1.7)

 

Interesting. Thanks for those numbers. I didn't realize that a good hitting pitcher could make up so much production.

[Soul]

 

The offense is pathetic, period.

 

And Zambrano has been just as pathetic.

 

I wish that wasn't true. He did have a couple months where he found it. Now he's back to giving up the big inning again. It reminds me of April.

[haltz]

How do you define how good they really are? There has to be some objective anchor. Words like "ture" are supposed to reflect reality not a statistical model of the way the world is supposed to work. In other words, it is not reality that is wrong. Reality is what really happend. The trick is getting a model that is relatively accurate in relation to the "true value" of something you are observing.

The statistical model is trying to reflect reality, not the other way around. But what reality are we talking about? Players have different talent levels, which is obviously not a hard concept.

 

If Matt Murton's true OBA talent is .360, there are going to be years that he under or overperforms, or that is changed due to variance outside his control, or both. This is what projection systems are getting at when they use 50th lines. It's a theoretical construct, but, a perfect 50th line would represent exactly how good that player is at every point in his career.

 

The more data we have, the closer we can get to pinpointing how good that player is. There's no perfect projection, of course. It's just about taking everything that player cannot control out of the equation. We can put the context back in and talk about that "reality" but that's not relevant to the discussion here. E.g., there is a lot of context and things outside a pitcher's control involved in Earned Runs.

 

ERA is a poor predictor and is loaded with "noise" but it can also be an objective anchor for how "good" a pitcher has been.

 

Every statistic becomes a better predictor with more data.

Career ERA+ works well for judging Roger Clemens' career, but if it represents Ian Snell's true talent at this point then it's a coincidence to a certain extent. Wainwright and Looper have identical FIPs to their ERAs this year. Sometimes these things happen, but it doesn't make ERA a good way to judge a pitcher.

[CubinNY]

How do you define how good they really are? There has to be some objective anchor. Words like "ture" are supposed to reflect reality not a statistical model of the way the world is supposed to work. In other words, it is not reality that is wrong. Reality is what really happend. The trick is getting a model that is relatively accurate in relation to the "true value" of something you are observing.

The statistical model is trying to reflect reality, not the other way around. But what reality are we talking about? Players have different talent levels, which is obviously not a hard concept.

 

If Matt Murton's true OBA talent is .360, there are going to be years that he under or overperforms, or that is changed due to variance outside his control, or both. This is what projection systems are getting at when they use 50th lines. It's theoretical, because we'll never know (there is even variance over an entire career) and by the time we have a really good amount of data a player is, at the very least, in a decline phase. But, a perfect 50th line would represent exactly how good that player is at every point in his career.

 

The more data we have, the closer we can get to pinpointing how good that player is. There's no perfect projection, of course. It's just about taking everything that player cannot control out of the equation. We can put the context back in and talk about that "reality" but that's not relevant to the discussion here. E.g., there is a lot of context and things outside a pitcher's control involved in Earned Runs.

 

ERA is a poor predictor and is loaded with "noise" but it can also be an objective anchor for how "good" a pitcher has been.

 

Every statistic becomes a better predictor with more data.

Career ERA+ works well for judging Roger Clemens' career, but if it represents Ian Snell's true talent at this point then it's a coincidence to a certain extent. Wainwright and Looper have identical FIPs to their ERAs this year. Sometimes these things happen, but it doesn't make ERA a good way to judge a pitcher.

 

I am not following you.

 

I don't get a lot of what you have written but I don't get this the most, "If Matt Murton's true OBA talent is .360". How do you know what his true level of anything is?

 

Truth lies is how he has preformed in the past. Eveything is else is a projection and every projection gets better with more data, that obvious.

 

ERA, ERA+, and other descriptive statistics are rooted in reality. VORP, Pyth. record, etc. are inferential models whose main value lies in their predicitve utility. And that predictive utility lies in how much they actually reflect reality.

 

This has gotten tangental. Anyway, I'm slightly worried about Big Z.