Looking for regression among quick-starting pitchers

Alfredo Simon highlights the list of pitching regression candidates.(USATSI)
Alfredo Simon highlights the list of pitching regression candidates. (USATSI)

If the season ended Tuesday, Charlie Blackmon's .379 batting average would be the highest in baseball since Larry Walker matched that number in the halcyon days of the Blake Street Bombers in 1999.

Even more impressively, Aaron Harang's 0.85 ERA would have matched Tim Keefe's all-time best mark, set back in 1880; Rutherford B. Hayes served as the 19th President of the United States this same season.

I shouldn't have to belabor the point further; weird things happen in the first month of the season. When dealing with small sample sizes, nearly any outcome is within the realm of possibility, though most players tend to revert back to their true talent level before long. Harang provided an excellent example of that principle in action Wednesday night, as the Marlins drubbed him for nine runs to raise his ERA to a much more believable 2.97.

Small-sample size anomalies tend to happen more often with pitchers than hitters just based on the sheer volatility of the position. Even in small samples, we can look at certain numbers that can give us a clue of how well a pitcher has actually thrown the ball, independent of his ERA.

As Chris Cwik noted last week in his ongoing series on advanced stats, xFIP is one stat that can cut through the noise of ERA. It attempts to give a pitcher credit for those things he can control (strikeouts, home runs, etc.), while removing the influence of stone-handed infielders or a particularly stiff wind blowing out to right. When looking for possible regression candidates, looking at pitchers with an ERA significantly below their xFIP is as good a place to start as any:

Alfredo Simon 1.60 4.27 +2.67
Julio Teheran 1.47 4.11 +2.64
Adam Wainwright 1.20


Yovani Gallardo 1.91


Justin Verlander 2.48 4.37


Mark Buehrle 2.16 4.04 +1.88
John Danks 3.48 5.32 +1.84
A.J. Burnett 2.15


Jason Vargas 2.40


Shelby Miller 2.86 4.59


Earlier today, Adam Aizer named Shelby Miller as someone he would be willing to take off the hands of a frustrated owner. Based on these numbers, you might want to take him up on the offer, especially if he really is offering a resurgent Scott Kazmir, whose xFIP is more than 1.5 runs lower than Miller's.

Of course, that list shouldn't be mistaken for a cut-and-dry "Sell-high list," even if it would tend to suggest the players named are due for some regression. As Aizer noted, Miller is probably better than he has pitched so far this season, and could easily put together a long string of solid performances, where his performance catches up to his stuff. If you were a believer in Miller's potential and youth five weeks ago, your faith shouldn't be shaken too bad.

And you obviously won't be moving someone like Adam Wainwright, who doesn't have all that far to fall, no matter what xFIP tells us. We know Adam Wainwright is one of the most dependable starters in the game, and will still be so even if a few more hits start to fall in as his luck evens out. You would be a fool to move him. 

But, if you've been enjoying the strong starts of guys like Danks, Vargas or Alfredo Simon, you certainly may want to look into moving them before the bad times creep in. For less-than-elite players pitching over their heads, the fall back to reality can reach terminal velocity with little notice.

Just ask anyone who had Aaron Harang in the lineup for Week 5.

Fantasy Writer

Though he can be found covering three different sports depending on the time of year, there is one unifying theme in how Chris Towers approaches sports; "Where's the evidence?" It doesn't matter how outlandish... Full Bio

Show Comments Hide Comments
Our Latest Stories