Pitching Stat

[theblujay]

ERA almost always regresses back towards xFIP given enough time. This isn't an unfalsifiable abstract idea; it happens. If you want know how well a pitcher will prevent runs going forward, you're almost always better off looking at DIPS than things like WHIP and ERA. This is a fact.

 

ERA is a pretty good measure of effective the pitcher has been at the most important thing: not giving up runs. ERA corrects towards xFIP, but only over a long period of time. That's probably because FIP includes many of the things that control runs allowed. But it is also missing many of the natural aspects of the game of baseball which is why we still play the games.

[z3r0s]

Where else in life can your performance be "normalized" to weed out the effects of "things you don't control"? Let's take my above example and apply it to the real world. My work laptop is low on RAM, but I'm going to download a bunch of moves, games and other s*** on it anyways. Now it's really slow and effects my ability to get my work done on time. Who is going to get the blame? Me of course, even if I yap that I need a better laptop.

 

How the hell does this strawman relate to the argument?

[z3r0s]

ERA is a pretty good measure of effective the pitcher has been at the most important thing: not giving up runs. ERA corrects towards xFIP, but only over a long period of time. That's probably because FIP includes many of the things that control runs allowed. But it is also missing many of the natural aspects of the game of baseball which is why we still play the games.

 

I think you are fundamentally misunderstanding how we use DIPS. Nobody here is saying that FIP is the best representation of how a pitcher HAS performed. ERA is the BEST stat for telling you how many runs a pitcher has let in for each 9 innings he pitched. Its useful enough in evaluating the actual success a pitcher has had over a period of time.

 

DIPS in the other hand is a better predictor of how a pitcher might perform in the future. Its exclusively useful as a predictive number. Nobody here is going to say that a pitcher deserves the Cy Young because he had a crazy good year according to FIP if his results that year were not good. They will however say that that guy will probably be better in the future than his results indicate.

[Nox]

Lol

 

Wow. Epic troll thread. 2 years ago on the old board I probably would have had an aneurysm.

[GD]

Wow. Epic troll thread.

 

It started out with good intentions. Then it all went to hell.

[Nox]

Say the SP knows that his 2B and 1B have poor range.

 

Considering scouts can't evaluate range, I don't think we should be assuming pitchers are any better this. With this line of thinking, I'm sure many would have wanted groundballs hit to Derek Jeter (someone who always looked better at SS than he was) throughout his career. This is a quantifiable losing strategy.

 

I think DIPS were invented by agents to explain why their mediocre SP client isn't all that mediocre.

 

Voros McCracken is not an agent.

 

Where else in life can your performance be "normalized" to weed out the effects of "things you don't control"?

 

Basically everywhere if you hired someone to analyze your performance properly. Teasing out the effects of random variation and understanding the statistical distribution the domain is governed by are about the 2 most fundamental concepts in any sort of analytics (company performance, department performance, employee performance etc), not just sports/baseball. It just so happens that this task is about as easy as it gets in baseball. That's why we use it so confidently.

 

 

The other thing I dislike about DIPS is that they encourage Brandon Morrow circa 2010 type of pitching. I don't want my pitcher going out there and trying to strike everyone out so he can blow up or burn himself out by the 4th, 5th or 6th inning. I have more respect for the SP who can go out there and keep his team in the game with a 5 or 6 K/9IP rate and pitch 230 innings a year by using his defense efficiently and conserving his pitches.

 

This is another common fallacy you see and hear mentioned all the time. High strikeout totals do not drive up pitch counts. Strikeouts and pitch counts have negligible correlation. I know that's counter intuitive, it was for me the first time I saw the study(ies) but it's also true. It's one of the biggest reasons why the pitch-to-contact school of thought is so completely flawed.

[TheHurl]

With this line of thinking, I'm sure many would have wanted groundballs hit to Derek Jeter (

 

"To Jeter" isn't the problem...to his left or right is

[z3r0s]

"To Jeter" isn't the problem...to his left or right is

 

 

Your precious 'stats' can't quantify this s***!

[Nox]

DIPS in the other hand is a better predictor of how a pitcher might perform in the future. Its exclusively useful as a predictive number. Nobody here is going to say that a pitcher deserves the Cy Young because he had a crazy good year according to FIP if his results that year were not good. They will however say that that guy will probably be better in the future than his results indicate.

 

With all due respect, I think you're a little off here too (though obviously not anywhere near the level of some of the mouth breathers in this thread). DIPS are not merely predictive of ERA. They are also a measurement of a result set. The elements of that result set are deemed to be a more direct measure of things the pitcher can control.

 

If this weren't the case, there would be no argument to use FIP in WAR. WAR is simply a valuation framework, an economic tool, for translating a result set into value in a more tangible form.

 

If we want prediction, we use a projection system. While they all use many of the same principles that went into the creation of FIP, xFIP, is a different mechanism.

[Nox]

[ATTACH=CONFIG]317[/ATTACH]

 

Your precious 'stats' can't quantify this s***!

 

Career leader in wJT+

[Nox]

"To Jeter" isn't the problem...to his left or right is

 

Touche.

[z3r0s]

That's a fair comment. Went too far with the 'exclusively' bit.

[Angrioter]

Stupid people always will do and will say stupid-s***.

FIP = ((13*HR)+(3*(BB+HBP))-(2*K))/IP + constant (3.20)

Real variables: HR, BB, HBP and IP

 

Every damn offensive event has a different value; While more bases you get, more value has the offensive event in question...

ERA doesn't tell much of a pitcher's true talent. Depend on where you play, the defense and luck.

[Nox]

I knew Nox wouldn't be able to keep silent on this thread.

 

Obvious Darrell bait was obvious.

[theblujay]

Obvious Darrell bait was obvious.

 

FIP doesn't even take hits or runs allowed into account. How can you take that seriously?

[Angrioter]

Greinke

Pitching 1.6 WAR

Hitting 1.0 WAR (.358 wOBA, 132 wRC+)

[Nox]

FIP doesn't even take hits or runs allowed into account. How can you take that seriously?

 

Because there's alot more than goes into run and hit prevention than what the pitcher brings to the table.

[theblujay]

Because there's alot more than goes into run and hit prevention than what the pitcher brings to the table.

 

But that's always there. So why wouldn't you take that into account. There's much more that goes into giving up a hit that's under the pitchers control that isn't in FIP and it's almost impossible to put into stats. Why should we ignore that.

[theblujay]

I feel like you just don't undestand things like DIPS on a fundamental level, and it makes discourse impossible.

 

I'm not talking about DIPS.

[z3r0s]

I feel like you just don't undestand things like DIPS on a fundamental level, and it makes discourse impossible.

I'm not talking about DIPS.

 

Classic!

[Angrioter]

I'm not talking about DIPS.

 

what? You say this after 13 pages.

 

DIPS = FIP, xFIP, tERA

[Nox]

But that's always there. So why wouldn't you take that into account. There's much more that goes into giving up a hit that's under the pitchers control that isn't in FIP and it's almost impossible to put into stats. Why should we ignore that.

 

"So why wouldn't you take that into account"

 

Because they're hugely variable.

 

Nobody argues that DIPS theory is effective because pitchers actually have 0 control over batted balls, itt's just that teasing out that effect is exceptionally difficult. It's been shown that ignoring the effect completely gets us far closer to the truth (underlying pitcher talent) than run and ball in play metrics. It's not perfect, it's better.

 

There's much more that goes into giving up a hit that's under the pitchers control that isn't in FIP and it's almost impossible to put into stats.

 

There's also way less than what you're implying. Tell me this: why do position players who come into pitch have a cumulative BABIP against that is essentially in line with the league average produced by full time pitchers?

[theblujay]

what? You say this after 13 pages.

 

DIPS = FIP, xFIP, tERA

I'm talking about people who think they can use stats that do not include crucial information to show how effective a pitcher is.

[Nox]

I'm talking about people who think they can use stats that do not include crucial information to show how effective a pitcher is.

 

But they do use the information. They use it in the smartest way possible, which was determined empirically to ignore it.

[theblujay]

"So why wouldn't you take that into account"

 

Because they're hugely variable.

 

Nobody argues that DIPS theory is effective because pitchers actually have 0 control over batted balls, itt's just that teasing out that effect is exceptionally difficult. It's been shown that ignoring the effect completely gets us far closer to the truth (underlying pitcher talent) than run and ball in play metrics. It's not perfect, it's better.

 

 

 

There's also way less than what you're implying. Tell me this: why do position players who come into pitch have a cumulative BABIP against that is essentially in line with the league average produced by full time pitchers?

 

I guess ignoring the effects works over a very large sample, but that's only useful in some circumstances.

 

 

2nd point: Again that's over a huge sample and the larger the sample size the less the luck or "real" part of baseball effects the data.

[Angrioter]

I'm talking about people who think they can use stats that do not include crucial information to show how effective a pitcher is.

 

What statistics?

[theblujay]

But they do use the information. They use it in the smartest way possible, which was determined empirically to ignore it.

 

This whole debate was about people who only use SABR stats to evaluate players and ignore the human parts of the game. It might be best to ignore it all when you are looking are stats from years of data and looking for average production in the years to come, but there are other important things to look at from game to game.

[theblujay]

What statistics?

 

Lots of SABR stats

[Angrioter]

Lots of SABR stats

 

All of these stats are based on offensive and defensive events.

Score runs and prevent runs; Hit and catch the ball.

[Angrioter]

[ATTACH=CONFIG]317[/ATTACH]

 

Your precious 'stats' can't quantify this s***!

 

http://www2.baseballinfosolutions.com/assets/1/7/OF-out-ratio.jpg

http://www2.baseballinfosolutions.com/assets/1/7/IF-out-ratio.jpg