As i was getting ready to write another piece on pitching WAR and why it should be based on ERA, not FIP, after seeing the numbers of Corey Knebel and Roberto Osuna propelled me to do so, i had an epiphany.
One of the hardest things to do in baseball is to accurately separate pitcher and defense, what do we give the pitcher credit for, what do we give the defense credit for.
FIP was created to do exactly that, take away sequencing, luck, and defense, giving you a measurement of the pitcher's performance, and over the majority of them it does that pretty well, once in a while you get a pitcher who masters the ability of outpitching their FIP but even those guys tend to even out over the course of their careers.
For instance, a guy like Matt Cain during his prime well known for doing that finished it off with a 3.68 ERA, 3.92 FIP.
The point is, FIP is pretty accurate, and over the long haul, a pitcher's ERA through the ups and downs of the defense around him, the fluctuation in performance during high leverage situations and the lucky-unlucky Ferris wheel all pitchers go through, tends to balance itself out towards what his performance earned him.
As you all have seen my thought process in stating that the proper way to calculate WAR was through ERA came from the idea that WAR was created to measure added value, and since luck affects results, results determine value and WAR is added value, therefore luck must affect WAR, and FIP is designed to strip away the luck factor.
But as I was writing this article, it became clear to me that i could not blame a pitcher for his poor defense and in regards to the two other factors, luck and sequencing do matter, but to this day no one has been able to give an objective, logical explanation of what really affects it. The majority of pitchers tend to fluctuate on those two and eventually settle on what their ability should give them, plus I have a proper way to include them and not just dismiss their value, will get to it later.
It goes beyond simply realizing that it wasn't all the pitcher's fault, that was already clear, what came to me was that as soon as the act of:
Windup, delivery, pitch, batter's reaction, swing or take.
Is completed, the pitcher's job has ended, obviously he has fielding duties but as the pitcher it has, at that point he added whatever value he did to his team, it's done, and then what happens afterwards it should be considered as the defense giving value away, not subtracting it from the pitcher. For example:
Chris Archer throws a fastball right on the corner, the batter gets a piece of it and hits a Flyball into left field with a catch probability of 85%. Corey Dickerson doesn't get to it, becoming a double, the ball should've been caught, later in the inning, that hitter comes around to scores on a couple of sacrifice flies, that run goes into Chris Archer's box score, but when in fact as soon as the ball left the bat, Archer finished his job and gave his value on that play to his team.
ERA based WAR is flawed because it benefits pitchers with a better defense, but it has its merits given it values luck and sequencing which aren't recommended to predict future value but are a part of the added value.
You can't deny a guy who performed better when pitching with 2nd and 3rd and doing it so gave up fewer runs added more value, whether that's sustainable or not is a whole different story.
FIP based WAR is also flawed given it strips away luck and sequencing which are two factors that affect added value, therefore going against the principle behind WAR, which is, measuring how much value did a player give his team.
But also has its merits, stripping away the defense element, basing the performance solely on what the pitcher did, which is all he should be responsible for, also discounting luck and sequencing thereby making it easier to predict future performance and future WAR.
My belief is you should never dismiss something that has merits and positive sides, like both of these versions, do.
You should look for ways to use both, so i propose the following.
Use ERA based Pitching WAR to represent added value, i said pitchers shouldn't be responsible for their defense, but we are not at a point in which we completely understand defense, much less, separate it from luck and sequencing, to get the perfect WAR version.
Use FIP based WAR as the equivalent of xwOBA, have an expected Wins Above Replacement.
What is xwOBA really, but what a hitter's wOBA would look like under normal circumstances, same goes for xWAR, which dismisses luck and sequencing, numbers that change added value but fluctuate and tend to regress to average and defense, which is not the responsibility of the pitcher, basically giving you what his WAR would be like, under normal circumstances.
As always feel free to comment, suggest, subscribe and anything else you can think of, any questions you may have, I'm here to answer them.
