Potential Starting Pitcher K% Decliners
Yesterday, I took a look at the potential pitcher strikeout percentage surgers based on the regression equation I developed and shared with all of you last week. Today, I will look at the opposite side of the coin — those pitchers whose expected strikeout percentages are significantly lower than their actual strikeout percentages.
| L/Str | S/Str | F/Str | K% | xK% | Diff | |
|---|---|---|---|---|---|---|
| Yu Darvish | 27% | 25% | 25% | 34.8% | 30.4% | 4.4% |
| Felix Hernandez | 26% | 19% | 27% | 26.6% | 22.3% | 4.3% |
| Tyler Lyons | 30% | 10% | 25% | 15.8% | 11.8% | 4.0% |
| Jerome Williams | 23% | 16% | 27% | 18.2% | 14.5% | 3.7% |
| A.J. Burnett | 31% | 17% | 27% | 28.7% | 25.2% | 3.5% |
| Jordan Zimmermann | 25% | 13% | 27% | 15.8% | 12.4% | 3.4% |
| Adam Wainwright | 29% | 17% | 25% | 24.4% | 21.0% | 3.4% |
| Edwin Jackson | 23% | 17% | 28% | 20.1% | 16.9% | 3.2% |
| Max Scherzer | 29% | 20% | 28% | 31.4% | 28.2% | 3.2% |
| Eric Stults | 29% | 11% | 28% | 18.1% | 15.0% | 3.1% |
| Unweighted Population Avg | 28% | 15% | 27% | 19.1% | 18.9% |
Before we complain that this formula doesn’t work because Yu Darvish is atop the list, let’s just marvel at that S/Str rate. That leads all of baseball among starting pitchers…and easily too. Next highest is Jeff Samardzija and Anibal Sanchez at just 22%. Anywho, a pitcher is unlikely to post a K% this high without some luck. Sure, his stuff is ridic, but his current mark ranks second among all qualified starting pitchers since 2000! Of course, even an expected regression won’t change the fact that he’s likely still a top five starter (perhaps better? top three? top one?) the rest of the way.
Felix Hernandez ranks second because of his lack of called strikes and just league average rate of foul strikes. Still, I am amazed at how dominant he has remained despite his fastball velocity nearly declining in a straight line since 2007. Even though his velocity is at the lowest of his career, his SwStk% is currently at its highest. His xK% is right around what he’s posted since 2009, so it isn’t too hard to believe.
Woah. Tyler Lyons hasn’t exactly been a strikeout machine since making his debut, and yet xK% suggests that he should be even worse! That’s not a good sign and suggests that he’ll struggle to even earn NL-Only league value. More likely, he’ll lose his rotation spot in short order and return to the minors.
It’s hard to figure out what A.J. Burnett is doing differently that has led to the huge strikeout percentage spike. His pitch mix is the same as it has always been and his velocity is nearly identical to last season. xK% suggests some of that jump has been a result of good fortune, but even his expected mark is well above years past. It’s truly hard to believe the 36-year-old could remain this dominant all year.
Wow, Jordan Zimmermann! That’s what happens when you induced a below league average rate of both looking and swinging strikes. Pretty shocking given his outwardly good stuff and strong fastball velocity. I wondered if maybe Zimmermann has consistently underperformed his xK% and this has not been the case. In every previous season, he has actually slightly outperformed. While I have to imagine both his xK% and K% will rebound, his ERA/SIERA differential makes him a strong sell high candidate in my view.
Could you believe if Edwin Jackson’s strikeout percentage plummeted to the level xK% is hinting at? Though he barely has any redeeming qualities at the moment, at least he has contributed positive value in strikeouts for those fantasy owners still brave enough to keep him active. But what if those strikeouts disappeared? Oh boy, Chicago, we have a problem. If he didn’t appear here, I would have suggested that he makes for an interesting buy low candidate for those owners who like to embrace risk given his 3.86 SIERA. But maybe he’ll just be worthless in all leagues all season long.
Here are all of the pitchers whose xK% is lower than his K%.
Mike Podhorzer is the 2015 Fantasy Sports Writers Association Baseball Writer of the Year and three-time Tout Wars champion. He is the author of the eBook Projecting X 2.0: How to Forecast Baseball Player Performance, which teaches you how to project players yourself. Follow Mike on X@MikePodhorzer and contact him via email.
I like this idea. I suspect there is an artificial bias introduced in your process. I think this bias should be “easy” to fix. Have you noticed that your potential surgers are mostly bad pitchers and your potential decliners are mostly bad pitchers?
You’re correlating a rate from fractions of a whole. Therefore your inputs are not independent of one another. For example, if the next pitch thrown is a strike looking the rate of strike swinging correspondingly declines. You can get independence by correlating counting stats against one another and then transforming back into rate stats. Correlate total strikeouts against strikes looking/swinging/fowling, normalize with strike type per batter faced.
that is, decliners are mostly good pitchers.