Here's a Thought: Kevin Jepsen, Jose Arredondo, and Deceiving ERAs

Nathaniel StoltzSenior Analyst IAugust 6, 2009

ANAHEIM, CA - APRIL 08:  Jose Arredondo #66 of the Los Angeles Angels of Anaheim throws a pitch against the Oakland Athletics at Angel Stadium on April 8, 2009 in Anaheim, California. The Athletics defeated the Angels 6-4.  (Photo by Jeff Gross/Getty Images)

As an A's fan, I'm always particularly interested in what the rival Los Angeles Angels are doing.

I also often look through the FIP-ERA difference leaderboard to see who's getting lucky or unlucky.

For those of you who don't know, FIP (Fielding Independent Pitching) is a statistic that basically says what a pitcher's ERA would be if he had average defensive luck behind him.

Anyway, two pitchers with very high FIP-ERA differences are Angels relievers Kevin Jepsen and Jose Arredondo. 

Arredondo has an ERA of 5.55, but FIP shows him to be a very effective pitcher, with a 2.33 mark.

Jepsen's ERA is 6.66, but his FIP is a sparkling 2.99.

What happens when a pitcher has a dramatic split like this is that you get two very different opinions.

Traditional, old-school guys who look at just ERA say these two pitchers should be sent to the minors for their poor performance.

Video Play Button
Videos you might like

Stats guys will look at the FIP and say Arredondo and Jepsen are excellent relievers.

Who is right? Is it A) the old school or B) the new school?

The correct answer is C) we don't have enough information.

ERA and FIP are both incomplete metrics. If you go exclusively by one metric or the other, you'll often make incorrect predictions.

Let's look at Arredondo.

On the most superficial level, he looks like a bad pitcher with the 5.55 ERA and 1.64 WHIP.

On a second, deeper level, he looks merely unlucky, with the 2.37 FIP, .402 BABIP, and 62.5 percent strand rate.

But looking to that second level isn't enough to determine Arredondo's 2009 level of performance.

Stats guys might argue "The .402 BABIP must come down! It's all bad luck!"

And it should come about .380.

The average BABIP is .300, and a lot of sabermetric analysts mistakenly assume that everyone's will come near that level.

That's simply not true. Line drives usually fall for hits, ground balls sometimes do, and fly balls rarely do. Depending on the amount of each hit a pitcher allows, the BABIP he deserves could change dramatically.

A fairly good, if inexact, formula to calculate deserved BABIP is line drive percentage +.12.

Twenty-five percent of hits off Arredondo were line drives in 2009. Combine that with far more grounders than flies allowed, and his BABIP should be in the .370-.390 range.

Batters are smoking the ball of the righty this year, and that's his own fault, not the defense's.

So yes, .402 is a bit high, but it's not way out of line like the "it regresses to .300" conventional wisdom would have you believe.

And sure, the strand rate is bad luck.

Taking all that into account, it's reasonable to project Arredondo's true ability in high-fours ERA range. That's in between his ERA and FIP, but far closer to the ERA.

Arredondo is a great example of where some of the more advanced stats simply don't capture what's actually going on.

Now let's look at Jepsen.

On the superficial level, he's got a 6.66 ERA, 1.75 WHIP, and .332 batting average against.

On the second, deeper levl, the rookie righty has a .405 BABIP and just a 55.1 percent strand rate, putting his FIP at 2.99.

Examining his batted-ball splits, we see Jepsen allows 17.4 percent liners, 64 percent grounders, and 18.6 percent flies. Using the LD% +.12 formula, we get a deserved BABIP of .294. That should go up a bit because of the extreme GB/FB split, but only to the .310ish range.

Jepsen's high BABIP, unlike Arredondo's, is pretty much all a fluke. The ghastly strand rate likely is as well.

There's no real statistical evidence that shows Jepsen to be anything other than the 2.99 ERA pitcher his FIP claims him to be.

The only indictment in his stat line seems to be toward the Angels defense.

So, while ERA shows Jepsen and Arredondo to be terrible, and FIP shows them to be great, in reality, it looks like Arredondo is poor and Jepsen is very good.

I have to give the Angels credit for leaving Jepsen in the majors while sending Arredondo down. Obviously, they saw through the statistical mess as well and picked the right pitcher to stay in an L.A. uniform.

The lesson here: Look at the big picture, not just one stat that claims to show all of a pitcher's value by itself.