Sabermetric Spring Training: FIP

It’s almost Truck day and the Red Sox players are almost ready to get started.  What better time to start a review of the stats we most commonly use at Fire Brand?  We’ll try to cover as many as we can and hope you’ll ask questions if you have any or feel free to comment on potential pitfalls of the different stats.

FIP is right up there with WAR as one of the most common stats people get introduced to statistical analysis with.  An early and neutral attempt to value a pitchers ability.  FIP or Fielding Independent Pitching is a fairly straight forward computation created by Tom Tango:

FIP=((HR*13+(BB+HBP-IBB)*3-K*2)/IP)+3.2

The resulting answer is an estimate of what a pitchers ERA would equal if his defense had been league average and he had given up a league average number of home runs.  The pitcher has no* control over his defense and mostly no control over how many home runs he gives up.  He can limit fly balls, but how many turn into home runs is mostly a factor of league and ball park factors.

*There is some debate about how much control a pitcher might have on his BABIP.  Most studies have found some effect, but the amount is relatively small and FIP is mostly accounting for the defense.

To fix this Dave Studeman from The Hardball Times has removed a bit of the home run effect.  To do this he replaced home runs with a pitchers fly balls against multiplied by the average league home runs given up per fly balls (11%).  The new equation is as follows:

FIP=(((FB*.11)*13+(BB+HBP-IBB)*3-K*2)/IP)+3.2

 This equation can be helpful, but has some flaws as well.  While almost all pitchers average to a HR/FB near 11 percent it’s not a sure thing.  Some home parks and leagues like the NL West have built plenty of pitchers who benefit by averages below the league average.

Using FIP in Context

As I stated before there is no perfect stat and FIP is not an end all pitching stat. In debating a pitcher you must identify if FIP is actually showing the true skill or is there a reason ERA might be a better measure. A good case I recently debated was Joe Saunders who has a career ERA of 4.16, but a FIP of 4.65. A swing of .50 runs is pretty big and says perhaps Saunders has a skill at controlling those factors that FIP assumes are out of his control.

If he is controlling a stat it would be BABIP, but after playing for several good teams defensively and having two extremely lucky seasons it’s tough to say he actually has an added skill.  Another factor that could explain differences in FIP is LOB% (Left on Base percentage).  Again a stat that could be a fault of defense, skill or luck.

FIP or xFIP are not always the answer and are not always predictive of skill, but both are easy to calculate and have a very high level of projection of future results.  No pitching stat can fully test pitching “skill”, but FIP and xFIP are fair judges of talent as long as you can read the context and see potential reasons to doubt it.

Categories: Boston Red Sox Sabermetric Spring Training

After taking an interest in sabermetrics and statistical analysis Troy began trying to use it to an advantage in fantasy baseball. He started the website RotoSavants.com and also spent time at HardballTimes.com and FantasyPros911.com. After a few years the interest in the Red Sox drew him to start a Red Sox-oriented site (Yawkey Way Academy) with fellow writer Lee Perrault. A short time later he joined Fire Brand. Writer from: December 14, 2009 – July 24, 2010, March 3, 2011 – May 10, 2012.

5 Responses to “Sabermetric Spring Training: FIP” Subscribe

  1. topherdd February 9, 2012 at 10:17 AM #

    As a long time baseball fan who has recently tried to understand the sabremetrics of the game ,these articles are a welcome read.Would like to see a page on this site to reference these stats.League leaders,what is average,good bad, Sox leaders.
    Not allowing the batter to make solid contact,broken bats ,dribblers,weak pop ups(Mariano Riviera) ,are those included in xFIP.? or is that just to difficult to quantify?

    • TroyPatterson February 9, 2012 at 11:21 AM #

      Once I get a few more going we will be working on a glossary, which will include the basic definition as well as links to these pages. Also any time we reference one of these stats there will be a link to our description.

      I agree with your idea for leaders, average and Sox leaders. I will use that on future stats, but FIP and xFIP are fairly easy to understand in this regard. It attempts to match ERA value so below 4.00 is better than average and around 3.00 is great. Anything under 3.00 is elite. Of course this shifted a bit in 2011 as offense was down a bit, but we'll see if that repeats in 2012.

      Most of those things you mentioned are out of the pitchers control and so FIP attempts to remove them, but something like broken bats and weak contact are something we need to identify as a skill by looking at numbers like Line Drive rate and BABIP. If a pitcher repeats this skill we can trust that ERA maybe more precise than FIP.

  2. Dave February 9, 2012 at 5:29 PM #

    Troy – nice addition. A glossary would be great. One minor quibble if you are adding this write-up to a glossary. As it relates to FIP are you sure about the comment, "The pitcher has no* control over his defense and mostly no control over how many home runs he gives up." Whereas I understand the fly ball/home run ratio concept (and, therefore agree with the sentence I quote), I thought FIP was designed to take fielding out of a pitchers performance evaluation and was not developed to correct for all things pitchers don't control. Fielding has no impact on home runs (ignoring Baseball Tonight Top 10 play catches over the fence) and, therefore, home runs were not only left in the FIP measure but given an extremely high weighting as well (i.e., the 13 weight). I would add the home run comment to the babip comment for improved accuracy. Yup, procrastinating at work.

    • TroyPatterson February 13, 2012 at 11:44 AM #

      Good point. I think the point of FIP is to remove fielding, but there is also a wish to remove the effects of luck inherent in the stat. The use of xFIP is to continue the effect of removing luck and obviously not fielding.

Trackbacks/Pingbacks

  1. Great Moments in Intellectual Dishonesty | Fire Brand of the American League - February 10, 2012

    [...] that the pitcher has the ability to control what happens to a ball once it’s put into play.  As Troy mentioned yesterday, pitchers don’t have that kind of control.  (Or at the very least, they have minimal [...]