ERA and why 0.00 ERA over three wins does not matter.

The next few posts will be sabermetric focused and provide some insight into why some stats are great for the layman, but are highly misleading. First up, ERA and what are some better ways to analyze a pitcher’s talent.

From time to time, especially when talking about my favorite player, Clayton Kershaw, I’ll gush about or bemoan a pitcher’s ERA. It just happens. Deep down I know ERA is a bad stat, but when it’s the one pitching stat that will forever be seared into baseball lore, it’s hard not to discuss Kershaw’s historically low ERA. And while Earned Run Average provides a glimpse into a pitcher’s effectiveness or ineffectiveness, it is through a heavily blurred lens. To explain why ERA is flawed, we have to examine what an earned run is.

An earned run is any run for which the pitcher is held accountable. For example, if Kershaw allowed Batter A to reach base on a hit and that Batter A came around to score, it is an earned run against Kershaw. Makes sense, right? It seems intuitive. But in a different scenario, say Kershaw allowed Batter A to reach base and then he was pulled from the game and was replaced by Brandon League. If League allowed Batter B to get a hit and Batter A scored, the earned run still goes against Kershaw because he allowed Batter A to get on base. You might be thinking, “But it shouldn’t count against League because it’s not his fault Batter A is on base.” And you’re right. But League was the reason why Batter A scored. The other problem with ERA is that some pitchers with a high ERA are backed by a lousy defense. Balls that get by slow or inept infielders, or drop in because the outfields are lazy schlops go against the pitcher who is normally lights-out. On a more analytic level, ERA does not account for league differences – American League pitchers are more likely to have higher ERAs since they face a designated hitter – nor does it consider differences in parks, the umpire, the scorer, and just random luck in general. So ERA doesn’t limit the amount of variables that can affect a pitcher in a single game. 

If you’re one who would just like to look at a stat and not have to put much thought into it, K/BB is a great ratio to measure a pitchers efficiency. It simply looks at a pitcher’s strikeout rate over walk rate. Kershaw has a 8.6/1.9 K/BB. So for every one walk, he has about 4.6 strikeouts. Brandon League has a 4.66/2.56 K/BB, or 1.82 strikeouts for every walk. Another quick stat to look at is WHIP, which is Walks + Hits per Innings Pitched. This stat will tell you how many walks and hits a pitcher allows per inning. A WHIP rate of 1.32 is about average, with 1.00 being excellent and 1.60 being awful. This season, Kershaw is sitting on a nice 0.86 WHIP rate. League is at 1.42

If you simply cannot let go of ERA, then I suggest you take a look at FIP. FIP stands for Fielding Independent Pitching. FIP looks at strikeouts, walks, home runs, hit-by-pitch pitches.  Yes, some variables that affect FIP are the umpire, the park, and some luck, but it does a great job at examining over which events a pitcher has control. FIP is scaled to ERA, so a person interested in this stat can easily understand the stat at first glance. A variant of FIP is xFIP and it is a slightly more accurate calculation of FIP – it doesn’t assume that home runs are the fault of the pitcher, so instead it looks at fly-ball to ground-ball ratio instead of home runs allowed. To have an idea of what is a good FIP or xFIP and what is awful: an average FIP/xFIP is 4.00, with excellent being at 2.90 and awful being 5.00. Clayton Kershaw has a 2.47/2.97 FIP/xFIP this season. League has a 4.78/4.17 FIP/xFIP.

Now, if you’re suspect of ERA, but K/BB, WHIP, FIP, and xFIP haven’t convinced you, perhaps tERA is a better stat. Again, like FIP and xFIP, it is scaled to ERA so looking at it is intuitive for any baseball fan. tERA does not differentiate between earned and unearned runs. How a ball is hit can determine how a ball is fielded or not fielded. Since how a ball is hit is determined by the pitcher, we can extend the logic and conclude that how a ball is fielded is also determined by the pitcher. A groundball is expected to be an out more so than a line drive is. A fly ball is more likely to lead to runs scored than a ground ball is. With that said, a fly-ball pitcher is more likely to have a higher expected run to expected out ratio. With that ratio, tERA can be determined and we can gauge how hard a pitcher is being hit and what exactly a pitcher is doing. For reference, an average tERA is 4.20, with 3.20 being excellent and 5.50 being awful. This season, Clayton Kershaw is at a 2.78 tERA. League, poor poor League is at a 4.86 tERA.

So there you have it. The next time you hear that Brandon League has found his groove with three wins and a 0.00 ERA over those three wins, take a look at his peripheral stats and let them guide you to a more accurate discussion.

TL;DR: Kershaw good, League bad.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s