Wrestling With FIP

Mandatory Credit: Daniel Shirey-USA TODAY Sports

I don’t really like FIP as a stat.

FIP is another tool for measuring how well a pitcher is performing, but I have had concerns that it is getting too much credit.  For example, among though who speak ‘Advanced Metrics,’ it is a fashionable thing to say something like:  “Well, he’s clearly outpitching his peripherals – he’s due for a regression.”  When you hear that, the speaker is most often referring to a pitcher’s FIP being notably higher than his ERA.  Even the professional writers say things like that:

"[Luis] Avilan has proven himself as a valuable setup man.  But the 24-year-old southpaw has already made a professional-high 43 appearances and his 3.38 FIP (Fielding Independent Pitching) is not nearly as comforting as his 1.40 ERA.” (Mark Bowman, mlb.com)"

But what does this really mean, anyway?

FIP stands for ‘Fielding Independent Pitching.’  It is supposed to be a measurement about how well a pitcher controls the things he can control by himself… taking fielding considerations out of the equation.  What would that be exactly?  Outcomes that involve no bat-on-ball contact:  walks, strikeouts, and hit batsmen; plus homers.  On this point, I have some concerns… but no real solution to suggest:  but realize that there is a great deal of space between ‘no contact’ and ‘full contact’ (the homer).  And it’s that gap that I think leads to some issues with the use of FIP.

The formula that fangraphs uses for FIP is the following:

• ((13*HR)+(3*(BB+HBP))-(2*K))/IP + fudge factor

Don’t worry about the fudge factor – it’s their attempt to make FIP balance with ERA, and the point of this is not to argue with those details.   But the formula breaks down like this:

• Homers effectively count 13 points against a pitcher.
• Walks and hit batters count 3 points against a pitcher each.
• Strikeouts count 2 points in your favor.
• The more innings you throw, the better – which works against relievers and pretty much magnifies all of their mistakes.

Overall, if you give up a lot of homers or you’re not a ‘strikeout pitcher’, then FIP is not your friend.

Here are my quarrels about FIP’s shortcomings:

• The best thing you can do as a pitcher is get a strikeout.  The worst is to yield a homer.  But there are a lot of good things that you can do in between there – and to be effective, the ‘finesse pitcher’ must do those things, which includes:  changing speeds, locating pitches effectively, strategic use of breaking balls mixed with fastballs.  When done well, these things make for less effective contact when hitters swing.  But FIP doesn’t care what happens when hitters swing… unless they homer or miss… and that leaves out a bunch of possible outcomes.
• The bad pitcher will not do those things, and everything hit is a line drive.  But if those line drives stay in the park, then a bad pitcher can end up with a similar FIP rating as a great pitcher (Edinson Volquez has a 3.86 FIP; Julio Teheran 3.97.  But their ERAs are 5.74 and 3.35 respectively).
• Pitchers throwing ‘to contact’ will often pitch around hitters that they have trouble with.  This results in more walks and a worse FIP score.  It doesn’t mean that they did something wrong; it does mean they employed a strategy that is scored against them… though it also annoys their fantasy owners.
• FIP takes a lot of run prevention activities completely out of the equation.  This, of course, is the primary mission of the pitcher.

The Braves have a bunch of starters who pitch ‘to contact.’  In terms of ERA, they sit 5th in baseball (3.59), with a higher FIP of 3.78 (7th).  This year, both numbers are actually pretty good because their walks are down (5th-lowest walk rate) even as the staff strikeout rate is 16th overall.

Is Tim Hudson‘s pitching “unlucky” this year since his FIP is close to half a run lower than his ERA?  Note there were days in which Ramiro Pena played instead of a lesser fielder simply to help Hudson out.  But Huddy generally pitches in front of the same defense that Julio Teheran does (who has the reverse situation).   Hmmmm….

The team with the friendliest FIP is Detroit.  This is the number one staff in strikeout rate, and it’s not even close.  For them, FIP is instructive because that team does have a relatively porous infield defense, having limited range:  their starters’ ERA is half a point higher than FIP (3.73 vs. 3.13).

Okay, so what’s the point here?

When you’re evaluating pitchers – or evaluating competitors… possibly ones to acquire via trade – don’t use FIP exclusively to bash somebody without taking the full context into mind:  his pitching ‘style’, the defense behind him, the tendency of hitters to make solid contact (line drives) vs. grounders or pop-ups.  Is he tougher with runners on base?  Finally, is he good at getting through trouble successfully (limiting the damage)?

Also:  know that there are still things that elude the numbers:

• Remember the incredible first half turned in by Jair Jurrjens in 2011?  His ERA was 1.87… with a FIP of 3.92.  The only other number that really stands out is that he stranded 84% of runners allowed (which weren’t many – his WHIP was 1.07).
• How about Kris Medlen‘s second half of 2012:  0.97 ERA… and a 2.22 FIP.  He was stranding an incredible 93.3% of runners allowed.  He was also getting a bunch of Infield Fly Balls (10%).  This year, though, his rate is down to 3.3%… big difference.
• Mike Minor had a great second half of 2012 (overshadowed by Medlen).  He, too, had 10% infield flys… but that has continued this year (Teheran leads the starters this year with 12.4% IFFBs).

In short:  FIP is a tool… but it’s not the only one available.  So please be mindful of how you wield it.