Jonah Keri Trade Value Column

[witesoxfan]

My point is that the stat is flawed and it, along with most sabermetric stats, should be used as part of the picture, not the entire thing. Such as Jose Quintana was not better than Cole Hamels last year just because his FIP was better.

 

No one has ever said it's the entire thing. No one has ever said WAR, FIP, xFIP, SIERA, GB/FB, any stat ever has ever been the entire thing, the end all, be all of all statistics. Never has anyone said that. We are saying from a pure talent perspective and how well the pitcher actually pitched, FIP is a better indicator than ERA. ERA is a much, much better indicator of how many runs a pitcher has given up on average but that it is also far less predictive both in the short and long terms.

 

Quintana and Hamels were both incredibly good last year. I don't think it'd be a stretch to say that you can choose either one. The bottom line is that Hamels and Quintana are not as far apart talent wise as you or anyone may believe, Jose Quintana is absolutely more valuable than Cole Hamels is for a number of reasons, and Jose Quintana would get a $20+ mill AAV contract in a rational free market.

[Quin]

My point is that the stat is flawed and it, along with most sabermetric stats, should be used as part of the picture, not the entire thing. Such as Jose Quintana was not better than Cole Hamels last year just because his FIP was better.

 

Jose Quintana is probably the unluckiest pitcher in the history of the universe when it comes to the rest of the team holding a lead for him, getting him offensive support, etc.

 

He is also possibly the most underrated player in baseball. He'd get $20M per year over 7-8 years in this current market easily.

[Eminor3rd]

I don't think any of it is dumb luck, that was sarcasm. Jamie Moyer was a magician who got guys to swing at pitches they couldn't hit hard. That's what made him good, same with Buehrle. There is an art to throwing pitches in spots guys can't square up and even more so to getting guys to swing out in front of the ball by changing speeds. You can't just dismiss these things as luck when there is an actual skill to it. Roy Halladay is another pitcher who has some crazy ERA/FIP years because he's a guy that often pitched to contact.

 

My overall point is that the stat is flawed and it, along with most sabermetric stats, should be used as part of the picture, not the entire thing. Such as Jose Quintana was not better than Cole Hamels last year just because his FIP was better.

 

Then make a case that Hamels was better. You say it should only be "part of the picture," yet you're making the exact same type of absolute claim the opposite direction, with less evidence to back it up.

[Eminor3rd]

Jamie Moyer

1996- 3.98 ERA and a 4.95 FIP

2001- 3.43 ERA and a 4.17 FIP

2002- 3.32 ERA and a 4.03 FIP

2003- 3.27 ERA and a 4.01 FIP

2008- 3.71 ERA and a 4.32 FIP

 

Basically a huge part of Jamie Moyer's success is considered "lucky" according to FIP was my point.

 

Look at Garland's '05 season for a point of reference...4.24 FIP and a 3.5 ERA. Now was he lucky that year or maybe was he perhaps spotting his sinker well and inducing weak ground ball outs? Either way I can assure you that it wasn't Juan Uribe's fantastic defense at SS that made him good that year.

 

So you picked 5 non-consecutive years out of a 25+ year career where those numbers didn't align, and you call that evidence that the system is wrong? Because that seems like evidence that the system is right.

 

What I was trying to explain was that we currently do not know how much of any given pitcher's success or failure can be attributed to luck/good defense specifically, but we CAN tell you how much of it came from the pitcher -- with one notable exception, which is the effect of limiting hard contact. But that factor appears to vary from year to year across every pitcher's career (which you illustrated nicely with the Moyer example), which makes it not predictable, which implies either that it isn't a skill that pitchers can control or that it doesn't make enough difference to ensure sustained success or failure.

 

As wite said, NO one thinks that limiting hard contact isn't a factor, but none of the numbers in existence currently capture it, and FIP is closer to the truth than ERA. Time will give us an even better model if we're patient.

[TheFutureIsNear]

So you picked 5 non-consecutive years out of a 25+ year career where those numbers didn't align, and you call that evidence that the system is wrong? Because that seems like evidence that the system is right.

 

What I was trying to explain was that we currently do not know how much of any given pitcher's success or failure can be attributed to luck/good defense specifically, but we CAN tell you how much of it came from the pitcher -- with one notable exception, which is the effect of limiting hard contact. But that factor appears to vary from year to year across every pitcher's career (which you illustrated nicely with the Moyer example), which makes it not predictable, which implies either that it isn't a skill that pitchers can control or that it doesn't make enough difference to ensure sustained success or failure.

 

As wite said, NO one thinks that limiting hard contact isn't a factor, but none of the numbers in existence currently capture it, and FIP is closer to the truth than ERA. Time will give us an even better model if we're patient.

 

Jamie Moyer had 10 seasons with an ERA under 4, every single one of them his FIP was over his ERA. The 5 I chose to highlight were the most egregious examples. My point being that pitchers who pitch to contact and don't strike a lot of guys out are always going to have a high FIP and fWAR regardless

 

And to the bolded....Or maybe it just points out that having pinpoint control as a pitcher and avoiding mistakes is a very difficult skill to master and maintain? There's a reason only a small percentage of guys can do it year in and year out.

 

FIP suggests that any batter can hit any pitch hard at any time. Just not true. You're telling me there's no skill in getting a guy to roll over a 2-2 change-up for an easy out to the SS? Or to get someone to beat a low and away 2 seamer to the second baseman? If I'm getting this straight you do indeed acknowledge that this is a factor in a pitcher's success, but yet choose to use a stat that says it doesn't exist? That may work for you my friend, but not me. If it doesn't pass my common sense test I'm not going to use the stat, and common sense tells me that a pitcher can make certain pitches in certain spots to keep a hitter from hitting the ball hard. It's just that some are better at it than others and some can do it more consistently

 

 

 

[witesoxfan]

Jamie Moyer had 10 seasons with an ERA under 4, every single one of them his FIP was over his ERA. The 5 I chose to highlight were the most egregious examples. My point being that pitchers who pitch to contact and don't strike a lot of guys out are always going to have a high FIP and fWAR regardless

 

And to the bolded....Or maybe it just points out that having pinpoint control as a pitcher and avoiding mistakes is a very difficult skill to master and maintain? There's a reason only a small percentage of guys can do it year in and year out.

 

FIP suggests that any batter can hit any pitch hard at any time. Just not true. You're telling me there's no skill in getting a guy to roll over a 2-2 change-up for an easy out to the SS? Or to get someone to beat a low and away 2 seamer to the second baseman? If I'm getting this straight you do indeed acknowledge that this is a factor in a pitcher's success, but yet choose to use a stat that says it doesn't exist? That may work for you my friend, but not me. If it doesn't pass my common sense test I'm not going to use the stat, and common sense tells me that a pitcher can make certain pitches in certain spots to keep a hitter from hitting the ball hard. It's just that some are better at it than others and some can do it more consistently

 

No it's not.

 

Here

[TheFutureIsNear]

Then make a case that Hamels was better. You say it should only be "part of the picture," yet you're making the exact same type of absolute claim the opposite direction, with less evidence to back it up.

 

Jose Quintana

200.1 IP

8.00 K/9

2.34 BB/9

3.42 K/BB

3.32 ERA

.254 BA Against

1.24 WHIP

6.3 IP/Start

1.35 Ground ball/Fly ball

 

Cole Hamels

204.2 IP

8.7 K/9

3.36 K/BB

2.46 ERA

.232 BA against

1.15 WHIP

6.8 IP/Start

.93 Ground ball/Fly ball

 

Paints a much broader picture in my eyes and Hamels was better. Not by a lot, but he was better.

[TheFutureIsNear]

No it's not.

 

Here

 

I was using some hyperbole but in so many words that's what this means "the amount of balls that fall in for hits against pitchers do not correlate well across seasons". I call BS on that. A pitcher who can consistently hit his spots and keep hitters off balance with varying speeds will have significantly more success on balls in play being outs. Will there be exceptions every once in a while where a guy inside outs a blooper or something? Of course, but there have been plenty of guys who made successful careers out of pitching to contact and not striking a lot of guys out.

[Balta1701]

Paints a much broader picture in my eyes and Hamels was better. Not by a lot, but he was better.

When a guy in the NL is "better but not by a lot" and gets to face pitchers...the guy in the AL wins.

[LDF]

When a guy in the NL is "better but not by a lot" and gets to face pitchers...the guy in the AL wins.

 

a very good point. remember the old saying of the pitchers face tougher

comp b/c they don't face the pitchers batting.

[Y2Jimmy0]

Brief snippet from Jayson Stark 10 trade candidates column sums up nicely from an executive exactly why Quintana has more value than Hamels: "Here's my take," said an official of one club that has traded an ace in the not-so-distant past. "I think the Phillies can make a good trade for Cole Hamels. I don't think it will be a grand slam. If he was making $5 million a year, then they might get a grand slam. But he's not. I think they can make a good trade, one that makes sense for both teams. But I think there are too many different starting-pitcher options available for a team to give up the kind of prospects they're asking and pay that money. Teams don't have to do that right now. And they're not going to do that."

[shysocks]

I was using some hyperbole but in so many words that's what this means "the amount of balls that fall in for hits against pitchers do not correlate well across seasons". I call BS on that. A pitcher who can consistently hit his spots and keep hitters off balance with varying speeds will have significantly more success on balls in play being outs. Will there be exceptions every once in a while where a guy inside outs a blooper or something? Of course, but there have been plenty of guys who made successful careers out of pitching to contact and not striking a lot of guys out.

There is actual data that says it's not BS. It's not like people just make this stuff up out of thin air.

[TheFutureIsNear]

There is actual data that says it's not BS. It's not like people just make this stuff up out of thin air.

 

No there's not. It's an interpretation, there's no tangible evidence either way. I've already posted 2 articles in this thread with supporting evidence that contradicts what FIP says about the pitchers influence on contact. We'll agree to disagree on this because I'm honestly tired of arguing about it at this point.

 

[Eminor3rd]

No there's not. It's an interpretation, there's no tangible evidence either way. I've already posted 2 articles in this thread with supporting evidence that contradicts what FIP says about the pitchers influence on contact. We'll agree to disagree on this because I'm honestly tired of arguing about it at this point.

 

I don't think you're reading what people are writing. Are my posts too long? Serious question

 

No one thinks that FIP is complete. Everyone knows there's an element of soft/hard contact that no one has been able to isolate from any model. Those articles you're linking are attempts to try to find what's missing and add it to the model. But FIP is mathematically a better estimator of future performance than is ERA. The truth is in the middle of the two, but it's closer to FIP, and so FIP is better.

 

There's more research to be done, but until the next breakthrough, fWAR is a better indicator of true talent pitcher performance than bWAR.

 

To put it another way: You're bringing up gaps in the model that everyone agrees are there, but that doesn't change the fact that it's still the best model we have right now.

Edited December 7, 2014 by Eminor3rd

[TheFutureIsNear]

I don't think you're reading what people are writing. Are my posts too long? Serious question

 

No one thinks that FIP is complete. Everyone knows there's an element of soft/hard contact that no one has been able to isolate from any model. Those articles you're linking are attempts to try to find what's missing and add it to the model. But FIP is mathematically a better estimator of future performance than is ERA. The truth is in the middle of the two, but it's closer to FIP, and so FIP is better.

 

There's more research to be done, but until the next breakthrough, fWAR is a better indicator of true talent pitcher performance than bWAR.

 

To put it another way: You're bringing up gaps in the model that everyone agrees are there, but that doesn't change the fact that it's still the best model we have right now.

 

Posts too long?This post has been edited by the Soxtalk staff to remove objectionable material. Soxtalk encourages a free discussion between its members, but does not allow personal attacks, threats, graphic sexual material, nudity, or any other materials judged offensive by the Administrators and Moderators. Thank you.

Did you even bother to read the post I responded to?

[Eminor3rd]

Did you even bother to read the post I responded to?

 

shysocks's? Yes. He's saying that studies have shown FIP to be more accurate, despite its flaws. That's really the crux of the argument. I think that if you were to argue against it, you'd need data showing that bWAR/ERA is a better true talent estimator, not simply data showing FIP as imperfect.

Edited December 7, 2014 by Eminor3rd

[TheFutureIsNear]

shysocks's? Yes. He's saying that studies have shown FIP to be more accurate, despite its flaws. That's really the crux of the argument. I think that if you were to argue against it, you'd need data showing that bWAR/ERA is a better true talent estimator, not simply data showing FIP as imperfect.

 

Try reading the exchange again....he bolded a specific part of my post and then said that there is "actual data" saying that a pitcher doesn't have any affect on the type of contact a hitter makes, and I informed him that there isn't data that confirms anything. Something you yourself have stated. So maybe you should brush up on your own reading skills before you question mine.

 

I've read every word you've said, and I respect your opinion. Doesn't mean I have to agree with it. You keep stating your opinion as if its fact. FIP isn't nearly as widely accepted as you're making it out to be. There are tons of people and tons of articles that question FIP. If it works for you, and you want to you use it that's fine, but don't act like I'm wrong and you're right just because I don't agree with your view.

[Eminor3rd]

Try reading the exchange again....he bolded a specific part of my post and then said that there is "actual data" saying that a pitcher doesn't have any affect on the type of contact a hitter makes, and I informed him that there isn't data that confirms anything. Something you yourself have stated. So maybe you should brush up on your own reading skills before you question mine.

 

I've read every word you've said, and I respect your opinion. Doesn't mean I have to agree with it. You keep stating your opinion as if its fact. FIP isn't nearly as widely accepted as you're making it out to be. There are tons of people and tons of articles that question FIP. If it works for you, and you want to you use it that's fine, but don't act like I'm wrong and you're right just because I don't agree with your view.

 

The fact that people question FIP and that there are imperfections in FIP does NOT mean it still isn't the most accurate model we have to isolate pitcher performance from the "noise" of defense and randomness. That is a fact, not my opinion. It's not really something that can be "disagreed with" unless you want to argue that all of the research is fake. And that's exactly why FIP IS as widely accepted as I'm making it out to be -- because it works better than anything else, even if we know there's a better mousetrap that will hopefully eventually be built.

[TheFutureIsNear]

The fact that people question FIP and that there are imperfections in FIP does NOT mean it still isn't the most accurate model we have to isolate pitcher performance from the "noise" of defense and randomness. That is a fact, not my opinion. It's not really something that can be "disagreed with" unless you want to argue that all of the research is fake. And that's exactly why FIP IS as widely accepted as I'm making it out to be -- because it works better than anything else, even if we know there's a better mousetrap that will hopefully eventually be built.

 

You saying that things are fact doesn't make it so....hate to break this news to you.

[Eminor3rd]

You saying that things are fact doesn't make it so....hate to break this news to you.

 

Of course not. The decade and a half of research that is freely available for you and anyone else to read, though, DOES make it fact. I'm just telling you what I've read. My opinion has nothing to do with it.

 

I mean, if you're interested in this stuff, I can point out some stuff for you to read so you can see it for yourself. I know that some random dude just SAYING something is true isn't convincing, but you can look and scrutinize these studies yourself and see how it works and that it's real. I don't have any reason to lie to you, I have no vested interest in making sure Voros McCracken's legacy lives on. I'd change my tune TOMORROW if some new study came out that proved all of this wrong.

Edited December 8, 2014 by Eminor3rd

[TheFutureIsNear]

Of course not. The decade and a half of research that is freely available for you and anyone else to read, though, DOES make it fact. I'm just telling you what I've read. My opinion has nothing to do with it.

 

I mean, if you're interested in this stuff, I can point out some stuff for you to read so you can see it for yourself. I know that some random dude just SAYING something is true isn't convincing, but you can look and scrutinize these studies yourself and see how it works and that it's real. I don't have any reason to lie to you, I have no vested interest in making sure Voros McCracken's legacy lives on. I'd change my tune TOMORROW if some new study came out that proved all of this wrong.

 

I fully understand McCracken's interpretation and have read plenty on it. And despite that I still choose not to put heavy stock into FIP. What I think you're not understanding is that isn't NOT fact, it's merely 1 man's interpretation that gained traction. As you said, there is no way to factor in pitcher's influence...yet. Will you honestly be surprised if FIP is irrelevant in 5 years? Look, if the stat works for you then go ahead and use, I'm not trying to tell you you're wrong. All I'm saying is that a stat that interprets the skill of keeping hitters off balance and missing the barrel of the bat as "luck" or "random" isn't for me. Not sure why that's so hard to understand.

[Eminor3rd]

I fully understand McCracken's interpretation and have read plenty on it. And despite that I still choose not to put heavy stock into FIP. What I think you're not understanding is that isn't NOT fact, it's merely 1 man's interpretation that gained traction. As you said, there is no way to factor in pitcher's influence...yet. Will you honestly be surprised if FIP is irrelevant in 5 years? Look, if the stat works for you then go ahead and use, I'm not trying to tell you you're wrong. All I'm saying is that a stat that interprets the skill of keeping hitters off balance and missing the barrel of the bat as "luck" or "random" isn't for me. Not sure why that's so hard to understand.

 

The last line just isn't correct, and I think that's what I'm having trouble with. It doesn't claim or "interpret" that as luck -- it just can't isolate the effect so it isn't included. But nothing else can isolate the effect either. The "fact" part of the situation is that, even without being able to factor the "off-balance-ness" in, it's still mathematically more stable and a better predictor of future performance, so it's an objectively better tool for player evaluation than anything else out there right now. And so that's why most people put more stock in FIP-based WAR than they do in ERA-based WAR. And that's the argument that Quintana was better than Hamels last year.

 

When I'm saying it is "factually" the best, I'm not making any sort of claim based on philosophy or comfort or intuition, and I'm not saying that its assumptions are infallible, I'm just saying it provides the most accurate results right now. That's what the studies have shown.

 

Basically, FIP based WAR is more accurate because it strips away factors that don't belong to the pitcher, but is an incomplete model because it also strips away the ability to limit or allow hard contact. bWAR is a complete model which catches everything, but it is an inaccurate one because it gives the pitcher credit for things he cannot control. Mentally, we have to make concessions when using both, such as "yeah well he played in front of a crappy defense" or "yeah, but he's shown a consistent ability to repress homeruns throughout his career" and make the appropriate subjective adjustment in our valuations. But at the end of the day, the one that does a better job predicting future outcomes is the more useful one, and that is currently FIP-based WAR.

 

You're right though, that it almost certainly won't be in five years.

 

 

[witesoxfan]

This is also why you use a combination of statistics and you do not rely on one single statistic. If you see a guy that has a WAR of 5, all you know is that he's a good player but you have no idea what else he's done. The same thing can be said for FIP - if you see a guy with a FIP of 3.75, you only know that he's a guy who appears to be a good pitcher, but that may be under or overrating his ability to induce weak contact.

 

It wasn't that many years ago that FGs started keeping track of IFFB%. I don't believe they are capable of including this in any sort of formula for FIP or xFIP or SIERA or whatever other ones you'd like to use, but they've seen over the years that some guys, especially relievers, are capable of inducing infield flyballs at a higher rate than your normal pitcher. This matters because it's an inherently easy ball to catch and, while you typically want your GB/FB as high as possible, it's OK to allow IFFB's because 1) the BABIP of an IFFB is going to be very close to 0 and 2) the HR/FB of an IFFB WILL be 0%. We still have no way of incorporating that, so we look at that in conjunction with a stat like FIP, not as a mutually exclusive option.

 

We may see a guy put up a FIP of 3.50 but an ERA of 4.50 and wonder if it was defense or if it was allowing hard contact. We see that it was Javy Vazquez, who has a tendency to leave balls up and out over the plate, and can see that it was probably his ERA that was a bit closer to the truth.