In an effort to make "all that is old, new again", I have resurrected an old series of articles that I put together in a previous "Mack's Mets" lifetime that focused on the new wave of statistical analysis that has shaped baseball scouting and player rankings. Some would refer to them as "Sabermetrics" and others would argue that it isn't necessarily new anymore.
However, to keep things somewhat fresh, I will go over a new statistic each week and then I will attempt to relate that measure to our favorite team and one or more of our current players to see how we rate, so to speak. In the third installment of this series, we took a look at BABIP and how it relates to hitters AND pitchers. For this article, we will return to the pitching side of the ledger and take a look at a statistic known as FIP and how it attempts to look deeper into a pitcher’s performance.
So, what the heck does the acronym FIP stand for? FIP stands for “fielding independent pitching” and others generally refer to the statistic as “pitching independent from team defense”. The measurement was developed by Voros McCracken (cool name) way back in the early 2000’s as a way to level the playing field when comparing pitchers from different teams. In short, different teams have different players manning the variety of positions on the diamond, so it would stand to reason that a team with better defensive players would benefit a particular pitcher, especially on balls in play that may or may not be turned into outs.
FIP focuses on the things that the pitcher can control, namely strikeouts, walks, hit batters and home runs allowed. The listed results do not involve anyone else on the field, so comparing pitchers across different teams and eras will be much more accurate, if you will.
For those of you who are statistically inclined, are not afraid of math and actually enjoy figuring things out by hand (that pretty much eliminates all of us), here is the formula
that is used to figure out FIP (borrowed from Fan Graphs);
FIP = ((13*HR)+(3*(BB+HBP))-(2*K))/IP + constant***
You can derive the constant yourself by using the following formula;
FIP Constant = lgERA – (((13*lgHR)+(3*(lgBB+lgHBP))-(2*lgK))/lgIP)
I know, I know……make it stop!
The main thing to remember here is that you can find FIP on most statistical websites, so you don’t have to actually figure it out yourself (thankfully). But, if you look at the statistic, you should understand how it was computed, right?
This formula allows the reader to get a better glimpse into how effective a pitcher may have been, despite how good or bad the team around him was and that is valuable. For example, a ground ball pitcher with an excellent infield defense will be more effective then the similar pitcher who has a less then stellar group of defenders playing behind him. But, if you focus on the parameters that the pitcher can control, the other variables fall away and the picture comes into focus.
A pitcher’s FIP is expressed similarly to Earned Run Average (ERA) and in some cases, the FIP and ERA will be fairly close together. While ERA is important, remember that FIP focuses on the things the pitcher can control. So basically, you are looking for a difference between FIP and ERA. A large difference in the two may be a reflection of the quality (or lack thereof) of the team around the pitcher in question (i.e. is a pitcher over or underrated).
That analysis could help predict a pitcher’s future performance, especially when trades and/or free agency are at play and pitchers move from one team to another.
Here is a chart that attempts to explain FIP results in terms that are easily understood;
Excellent less than 3.20
Great 3.20 to 3.50
Above Average 3.50 to 3.80
Average 3.80 to 4.20
Below Average 4.20 to 4.40
Poor 4.40 to 4.70
Awful 4.70 to 5.00
***There isn’t a word for FIP values above 5.00, or at least one fit to use on this blog.
So, what about our pitching staff?
Using only the top four pitchers in terms of innings pitched (which is extremely telling in a different way), you come up with the following list;
Pitcher IP ERA FIP
deGrom 201.33 3.53 3.50
Gsellman 119.67 5.19 4.89
Montero 119.00 5.52 4.37
Lugo 101.33 4.71 3.95
Vargas 179.00 4.16 4.67
Obviously, JDG was our best pitcher last year and it wasn’t close. His numbers are pretty close, which suggest he is effective and the outs he generates are less susceptible to other players on the diamond. What is fascinating is the difference between ERA and FIP for the other three
folks on this list! Clearly, the Mets were not a good team last year and if you were relying on your defense to get consistent outs, you were in trouble. RM and RG didn’t pitch well in 2017, but come on!
If RG, RM and SL are second, third and fourth in 2018 innings pitched, I think we all know where the season will be headed.
I also tossed Jason Vargas’ line in the mix, which was crafted in Kansas City (a very good defensive team that is fundamentally sound). Not surprisingly, his ERA was better then his FIP, which reflects the quality of the defensive team behind him. Let’s hope that the 2018 Mets are better in the field then the 2017 version, or things may get ugly for JV.
So, much like any statistic, FIP can assist with player evaluations, but should not be used as the only factor in an evaluation. However, it can be a valuable predictive tool for future seasons, especially when you have a larger body of work to analyze.
I have fallen in love with the new statistics in baseball, but I am glad that someone else does the mater and comes up with the final number.
ReplyDeleteThe mathematical formula for this one looks like the one used to put a lander on the moon.
What would have been interesting would have been to, if available, also add the actual numbers that went into the formula that showed Jake's calculation.
ReplyDeleteI'm just asking for my old friend FIP Wilson, mind you.
Personally, I am happy with ERA and WHIP as my favorite go-to stats.
Gents, I don’t know where to park this, so I’ll turn it over to y’all. I am seeing Noah Syndergaard as another Joe Namath.
ReplyDeletehttps://www.theplayerstribune.com/en-us/articles/noah-syndergaard-mets-2018-mlb-opening-day
I remember when I was about 12 or 13 I had created some ways to evaluate pitchers and hitters that took into account history of such output at various ballparks and a team's won-lost record. Little did I know that my early attempts at ranking the effectiveness of players would someday become part of the normal statistical criteria used to make such judgments. I remember using the pitching one in particular to "prove" to a non-Mets fan how much better Tom Seaver was than most of his contemporaries when you factored in how few runs were scored for him :)
ReplyDeleteNot to get too deep into the weeds on this, but a ballpark adjustment would seem important to a stat like this, unless you're only comparing players on the same team, as you're doing above. You would think a pitcher who plays half the year in Yankee Stadium or Minute Maid Park is probably going to give up more homers than one who pitches half the year in CitiField or AT&T Park, no? So is there any kind of ballpark adjustment to level the playing field for a bandbox ballpark vs. a huge one?
ReplyDeleteMy beef with these "weighted" stats (13xHR, 3xBB...) is there seems to be no rationale for the coefficients other than to produce the result you are looking for. [e.g. If SAT scores don't correlate with Freshman GPA, add AxHeight + BxWeight and divide by C--choosing A,B & C to maximize correlation]
ReplyDeleteIn this case, if FIP & ERA are relatively close they are designed to say the same thing. If they are widely different, neither says much of anything.