xFIP
Description:
Expected Fielding Independent Pitching (xFIP) is a regressed version of FIP, developed by Dave Studeman from The Hardball Times. It’s calculated exactly the same as FIP, except it replaces a pitcher’s homerun rate with the league-average rate (10.6% HR/FB) since pitcher homerun rates have been shown to be very unstable over time. A pitcher may allow a homeruns on 12% of their flyballs one year, yet then turn around and only allow 7% the next year. Homerun rates can be very difficult to predict, so xFIP attempts to correct for that. Along with FIP, xFIP is one of the best metrics at predicting a pitcher’s future performance.
Context:
2010 xFIP Values
Things to Remember:
- While homerun rates are generally unstable over time, some pitchers are still more prone to allowing homeruns than others. If a pitcher has a long history of out- or under-performing the league average with their HR/FB rate, then you can reasonably expect them to perform closer to their career average than the league-average. For example, C.C. Sabathia has a career 8.5% homerun rate, and has never posted a homerun rate above league average. In cases like this, xFIP may not be the best judge of the player’s true talent level.
- Similarly, groundball pitchers typically have a higher HR/FB rate than flyball pitchers.
- xFIP values are scaled higher than FIP and ERA by a slight amount, meaning a 4.09 xFIP is better than a 4.09 FIP.
- xFIP has the highest correlation with future ERA of all the pitching metrics.
Links for Further Reading:
Batty for Baseball Stats – Hardball Times
1
Quick note on this:
“- Similarly, groundball pitchers typically have a higher homerun rate than flyball pitchers.”
I understand what you’re saying there — that groundball pitchers have higher HR/FB rates than flyball pitchers (due I think to the fact that their flyballs are more often the result of bad pitches). But the term “homerun rate” is potentially ambiguous if read out of context, as someone might think you were saying that groundball pitchers give up more home runs (per nine innings) than flyball pitchers do.
P.S. This dictionary idea is awesome! I especially like the percentile tables.
Great call, I’ll clarify. I feel like “homerun rate” looks less weird to people than HR/FB rate, but it can be confusing especially in this instance. Many thanks.
Wait, doesn’t xFIP taken into account flyball percentage? Thus a pitcher with a 5 K/9, 3 BB/9, and a 48 FB% will have a higher xFIP than a pitcher with a 2 K/9, 3 BB/9, and a 40 FB%
Nope, it only considers walks, strikeouts, and HR/FB rate….and the HR/FB rate is regressed to league-average for all players.
tERA and SIERA (BPro) are the two pitching stats that include batted ball data.
According to the calculator link, what Jon says is true. FB%*.106 gets you the expected HR portion.
Could you add the formula to this post (I see it on FIP, but not xFIP)?
Ahh bugger, yeah, right. I saw Dave Cameron address it in the chat this morning, and you’re right…FB% is included to help you get the expected HR portion. So Jon, you were basically right…FB% is included and helps get the expected HR portion of xFIP.
I would hesitate to say that Jon’s above example is correct without running it through the calculator, though…it all depends on the relative weights of each section. But yeah, sorry Jon…good catch all around.
Yeah I asked Dave about it in the chat and got clarification that my original thought of it was correct. And no problem, I’m just happy I have a clearer definition of xFIP now.
I messed this example up by the way, I meant to give the pitchers identical K and BB rates, ha.
Ah ok, thanks for the clarification! Yikes, I can’t believe I had it wrong the whole time.
Why can’t I type anything in the cells in the calculator?
I have been reviewing pitchers statistics for 2010 – 2011. I notice that their xFIP for that period is different than just for 2011. For instance, Jered Weaver has an xFIP of 3.42 for 2011 and 3.35 for the period 2010-11. Which is the one I should use when comparing him to other pitchers ? Is the 2011 a more current expectation and thus the better indicator ?