My original intent for this highly anticipated (you were anticipating it, right?) sequel was to use the the findings from Part I to identify players who might have under- or over-performed in 2011, and I will still do that (feel free to skip the next five paragraphs if that is all you are looking for).
However, after posting part I, I got some great feedback, both in the comments and via email and decided to re-do some of the analysis, so I am going to start by recapping a couple of the changes I made and what I found.
The most common question I received was about multicollinearity as a result of having O-Contact%, Z-Contact%, Contact%, O-Swing%, Z-Swing%, Swing% and Zone% all in a single regression. Since Contact% or Swing% can be defined as equations made up of the other variables, multicollinearity is definitely an issue. My stance on this is that when dealing with predictive equations, multicollinearity is not always an issue (it matters a lot more if you are trying to determine the impact any individual independent variable has on the dependent variable). That said, I re-ran the regression taking out Contact% and Swing% and had EXTREMELY similar results.
However, as I compared the data across multiple years, I found something interesting – a couple hitters kept cropping up at the top of the list of guys who saw too few first-pitch strikes. Specifically, Brian McCann and Albert Pujols. I figured there was a decent chance the impact here was based on the talent level of those players – wouldn’t we expect pitchers to avoid going after those two (among other very good hitters) while being more aggressive against weak hitters? So I added the previous season’s wRC+ to the independent variables as a proxy for the “fear” a pitcher feels facing a given hitter.
The results matched what we saw last week – the nine hitters who were the biggest outliers in terms of seeing fewer first-pitch strikes than we expected in 2010 saw their wRC+ dropped by 9.1 on average in 2011. The 11 biggest outliers on the other end saw their wRC+ increase by 7.2 on average.
Based on that, I am going to move forward using this new equation for ExF-Strike%, using wRC+ from the previous year, O-Contact%, Z-Contact%, O-Swing%, Z-Swing%, Zone% and SwStrk% as the key variables.
Now I want to look at 2011 result. We’ll start with a couple over achievers (guys who saw too few first-pitch strikes):
Vernon Wells: Wells is a popular target among baseball fans, which is bound to happen when you have a huge contract and a tiny impact on the field. And he has a reputation for having good years in even years (see 2006, 2008, 2010) while being down in odd years (2007, 2009, 2011), suggesting a bounce back in 2012 (if you believe this is a predictive theory). But in 2011 his F-Strike% was 56.9% as opposed to a predicted 58.9%. If he sees the same impact as the 2010 outliers did, his wRC+ could drop even lower – and when you are starting from an abysmal 77, there is not much room to fall. But if you are counting on a bounce back from Wells in 2012, you might want to temper your expectations.
Johnny Damon: The former Royals/A’s/Red Sox/Yankees/Tigers/Rays OF has not been a below average offensive player (by wRC+) since 2011, and has managed to stave off a complete collapse through his age 37 season. But this year his ExF-Strike% was 56.4% compared to his actual F-Strike% of 54.1%. And that does not bode well moving forward. If he sees the same drop in wRC+ as the 2010 outliers did, he will fall to a wRC+ of 100 for the first time since 2003 and could easily fall below that mark. Damon turns 38 a week from Saturday (Happy Birthday, Johnny!), and while he will attract far more attention in drafts than the average 38 year old outfielder, this may be the beginning of the end.
On the other end of the spectrum, we have:
Corey Hart: Hart has been up and down in his career – 2006, 2008, and 2009 were not particularly impressive, while 2007 and his last two seasons have been quite good. And it seems Hart may be in a position to at least maintain his 2011 production. 65.5% of first pitches to Hart were strikes, while his ExF-Strike% was just 56.3%. If that F-Strike% seems REALLY high…it is. Hart had the 5th highest F-Strike% in baseball this year – it is also significantly higher than the 58.3% and 58.0% he faced in 2009 and 2010. Hart saw his offensive value jump in 2011, and I would not be surprised to see that stick.
Colby Rasmus: Rasmus saw over 60% first pitch strikes during a rough 2009 campaign. That dropped to 58.4% in 2010 and his performance jumped. In 2011, his production plummeted as his F-Strike% jumped again, to 63.1%, but his ExF-Strike% was only 56.0%. Obviously there was more to his huge swing in offensive performance than just more first pitch strikes, but if his F-Strike% falls back to where we would expect it to be, it will do a lot to help him make it look like the Blue Jays got themselves a steal in this 25 year old outfielder with tremendous upside.