1. FIP vs. xFIP

For the statistically-inclined and curious, which do you prefer and why? I know it boils down to "HR are a pitcher's fault" vs. "HR are a percentage of FB rates and park effects" but I'm not exactly sure which one "feels" more accurate.

What say you?

3. Re: FIP vs. xFIP

Originally Posted by nate
For the statistically-inclined and curious, which do you prefer and why? I know it boils down to "HR are a pitcher's fault" vs. "HR are a percentage of FB rates and park effects" but I'm not exactly sure which one "feels" more accurate.

What say you?
I kind of prefer FIP. While I do think extreme HR/FB rates would artificially inflate or deflate FIP numbers of a pitcher, since most guys are going to be within standard deviation, I don't think those rates are skewed badly enough to concern myself with the adjustment.

I think, in a correlation test, xFIP would (and does) perform better, the difference is negligible when looking for a quick and dirty assessment of a guy.

4. Re: FIP vs. xFIP

Originally Posted by nate
For the statistically-inclined and curious, which do you prefer and why? I know it boils down to "HR are a pitcher's fault" vs. "HR are a percentage of FB rates and park effects" but I'm not exactly sure which one "feels" more accurate.

What say you?
Firstly, here's a good summary of xFIP. The bolded part is just "HR" in the standard FIP formula.

Expanding on FIP, xFIP realizes that some pitchers may be incredibly (un)lucky on their Home Run per Fly ball (HR/FB) rates in a given year, while starting pitchers should expect that rate to be around 10.6%. Theoretically, this should be a better predictor of a pitcher’s future ERA. xFIP is calculated exactly the same as FIP, with the exception of the Home Run component, which makes the formula look like this:

xFIP = ((13*(.106*# of fly balls))+(3*BB+HBP-IBB)-(2*K))/IP+constant

This newly introduced term helps normalize some of the variation that is inherent in FIP due to the aforementioned luck factor on fly balls leaving the park...

Things to Remember:

• Pitchers that don’t throw a lot of fly balls are going to look better.
• xFIP has the highest correlation with future ERA of all the pitching metrics.
• xFIP is just as vulnerable to small sample sizes as the other metrics, though still an upgrade over ERA.
It depends on what I'm doing. If I'm trying to assess a guy's likely ERA moving forward (for his current team), I'll usually stick with FIP and use xFIP only if his HR rate is out of line.

If I'm simply trying to compare pitcher A to pitcher B from a "which guy is better" or "how well has that guy pitched" perspective, I'll use xFIP.

My problem with xFIP is that it ignores park effects. While 10.6% HR/FB may be league average for starters, we would expect that figure to vary significantly by park. HR factors for GABP have averaged in the 125 range (25% more HR than expected) over the past few years. Petco, by comparison, is around 75%.

So in estimating ERA, xFIP is generally going to be overly optimistic about Reds pitchers, assuming a lower HR rate baseline than they can expect in reality.

I'd be much happier if instead of .103 it used a league and home park adjusted figure, using a ratio of where the guy has actually pitched (home or away). Of course, this takes you away from the simplicity and that's really the core attraction of FIP to begin with.

5. Re: FIP vs. xFIP

Originally Posted by RedsManRick
Firstly, here's a good summary of xFIP. The bolded part is just "HR" in the standard FIP formula.

It depends on what I'd doing. If I'm trying to assess a guy's likely ERA moving forward (for his current team), I'll usually stick with FIP and use xFIP only if his HR rate is out of line.

If I'm simply trying to compare pitcher A to pitcher B from a "which guy is better" or "how well has that guy pitched" perspective, I'll use xFIP.

My problem with xFIP is that it ignores park effects. While 10.6% HR/FB may be league average for starters, we would expect that figure to vary significantly by park. HR factors for GABP have averaged in the 125 range (25% more HR than expected) over the past few years. Petco, by comparison, is around 75%.

So in estimating ERA, xFIP is generally going to be overly optimistic about Reds pitchers, assuming a lower HR rate baseline than they can expect in reality.

I'd be much happier if instead of .103 it used a league and home park adjusted figure, using a ratio of where the guy has actually pitched (home or away). Of course, this takes you away from the simplicity and that's really the core attraction of FIP to begin with.
Right on, Rick. I guess the cool thing with xFIP is if you _wanted_ to, you could calculate based on park effect. But yeah, as you mention, the "ease" of calculation goes out the window.

Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•

 Board Moderators may, at their discretion and judgment, delete and/or edit any messages that violate any of the following guidelines: 1. Explicit references to alleged illegal or unlawful acts. 2. Graphic sexual descriptions. 3. Racial or ethnic slurs. 4. Use of edgy language (including masked profanity). 5. Direct personal attacks, flames, fights, trolling, baiting, name-calling, general nuisance, excessive player criticism or anything along those lines. 6. Posting spam. 7. Each person may have only one user account. It is fine to be critical here - that's what this board is for. But let's not beat a subject or a player to death, please. Thank you, and most importantly, enjoy yourselves!

RedsZone.com is a privately owned website and is not affiliated with the Cincinnati Reds or Major League Baseball

Contact us: Boss | GIK | cumberlandreds | dabvu2498 | Gallen5862 | Plus Plus | RedlegJake | RedsfaninMT | redsfan1995 | The Operator | Tommyjohn25