Pineda and Severino and Buchholz and Fiers, Oh My

Let’s begin our discussion of a foursome of American League starting pitchers saddled with inflated ERAs by presenting two tables first:

Pitching Metric YoY Correlations

Metric	YoY Correlation 2002-2012
WHIP	0.430
ERA	0.373
LOB%	0.238
BABIP	0.235
HR/FB	-0.029

SOURCE: http://www.fangraphs.com/blogs/basic-pitching-metric-correlation-1955-2012-2002-2012/

Pitching Metric Stabilization Points

Metric	Stabilization Point
HR/FB	400 fly balls
BABIP	2,000 balls in play

SOURCE: http://www.fangraphs.com/library/principles/sample-size/

Let’s understand something. The majority of starting pitchers have started four or five games so far, which is about 10% to 15% of a typical pitcher’s starts for the entire season. From a full season to the next full season, spanning 30 to 33 or so starts, ERA and WHIP do not correlate strongly, while BABIP’s correlation is weak, and there’s literally no correlation in HR/FB rate. So why oh why are we even bothering to look at the ERA and WHIP column at all at this point in the season?

I get it, it’s difficult to focus on anything else, especially if the pitcher in question is destroying your team’s ratios and you’re pulling your hair out as a result. But let’s not let emotion get in the way of sound analysis. My offseason analysis on pitchers focuses primarily on the underlying skills, largely ignoring the surface results, such as ERA and WHIP. So you better believe that just 10% to 15% into the season, those surface results should be, like, 99% ignored. Decisions should not be made based on those numbers.

The trio of BABIP, HR/FB rate, and LOB% greatly impact a pitcher’s ERA and WHIP marks, especially over small sample sizes. These “luck” metrics gyrate significantly from start to start, causing ERAs to yo-yo up and down. It’s silly to allow those marks to affect your pitcher evaluations.

Moving along, we find the stabilization points of two metrics, or the point at which each becomes reliable. You know how many innings it takes for a pitcher to reach 400 fly balls? Obviously, it’s dependent on his strikeout and fly ball rates. But let’s use Jose Quintana as an example, as his strikeout and fly ball rates have landed around the league average in 2014 and 2015. In those two seasons over 406.2 innings, he still didn’t hit the 400 fly ball barrier, as he finished those two years with just 383.

Essentially, that tells us that on average, a pitcher is going to allow one fly ball an inning. To reach 400 of them, you need 400 innings. The Major League leader in innings pitched is Chris Sale. He has thrown just 38 innings. The leader in fly balls allowed is Jered Weaver, at 42, or about 10% of what we need before we should start truly caring about HR/FB rate.

If we thought that HR/FB rate is a long ways away from stabilizing, how about BABIP?! The MLB leader in balls in play allowed is Marcus Stroman at 107. That’s just about 5% of the way to our stabilization point. You know what that means? BABIP means nothing now.

Phew. So this turned into a far longer diatribe than I expected, which leaves less time for a deep dive into the pitchers in the article’s title. We’ll instead summarize the foursome and we could discuss further in the comments, if you so desire.

April Busts Summary

Name	ERA	SIERA	WHIP	K%	BB%	BABIP	HR/FB	LOB%	FBv
Michael Pineda	6.95	3.11	1.59	26.7%	5.0%	0.377	26.9%	68.7%	92.7
Luis Severino	6.86	3.63	1.78	13.5%	3.4%	0.417	18.2%	62.1%	95.6
Clay Buchholz	6.33	4.66	1.50	18.1%	9.6%	0.306	14.8%	66.9%	91.2
Mike Fiers	5.73	3.81	1.36	18.1%	3.2%	0.309	28.6%	74.1%	89.2

I’ve included each pitcher’s ERA and SIERA, WHIP, important underlying skills, the luck metric triumvirate, and fastball velocity to investigate the possibility of injury.

We find here that despite the sky high ERA, Michael Pineda has been his fantastic usual self. Unfortunately, balls are finding the holes and/or the defense isn’t making the plays, and fly balls are leaving the park with regularity. His fastball is good, while his strikeout and walk skills remain elite. In his entire career, he has allowed just 463 fly balls, which is just above our threshold for reliability. So take his current HR/FB rate with a grain of salt.

And the BABIP? Ummm yeah, he’s only allowed 1,213 balls in play over his career, a far cry from the 2,000 we’re looking for to validate his BABIP skills. His career BABIP is .289, so unless you believe that his skills went from elite (.250 BABIP through 2014), to terrible (.337 BABIP 2014-2015), seemingly overnight, I think it’s fair to either blame a weak defense or chalk it up to some poor fortune. Or both.

Pineda isn’t the only Yankee suffering from a high ERA early in the season. Luis Severino came with big expectations, thanks to a mid-90s fastball and excellent changeup. Oddly, the swings and misses have disappeared this year, but the combination of his ground ball rate and pinpoint control has nonetheless led to a solid SIERA.

It’s worth nothing that Severino’s LD% allowed is third highest in baseball. So that would certainly help explain the high BABIP. But you know how many balls in play it takes for line drive rate to become reliable? 650. Severino is at…73, or 11%. So that’s rather meaningless from a predictive standpoint, though like I said, does do a good job from an explanatory perspective. Oh, and check his fastball velocity. The lack of strikeouts isn’t because his fastball has lost oomph.

We have come to expect health problems from Clay Buchholz and his performance record has been extremely erratic. This year, his skills aren’t so hot, but his fastball velocity is fine, so there’s no injury concerns just yet. Unlike the two Yankees above, he’s hurting from just a bit of poor fortune in all three luck metrics, which combined, have done a lot of damage to his ERA. He’s not exactly a great target, but owners should hold tight and hope the health cooperates.

Mike Fiers has posted the fourth highest HR/FB rate in baseball. You can’t blame the home park, as Miller Park is even more home run friendly than the friendly confines of Minute Maid. His strikeout rate is down, but his SwStk% sits right where it always has. With Scott Feldman pitching as poorly as his skills suggested would happen, it’s looking like him, not Fiers, is now most at risk of moving to the bullpen when Lance McCullers returns. It makes Fiers a nice acquisition target as he should come cheap.