Eight of a kind, three of a kind

I like ground ball pitchers. But I also expect them to throw strikes and miss bats. And, if it’s at all possible, suppress home runs on whatever fly balls and line drives, they do give up.

Show me a pitcher who is good at all these things, and I’ll want him on my team. Show me a guy who isn’t good at any, and I’ll shun the man.

Is this a healthy bias? Let’s take the case of 11 pitchers I should love or hate.

Setting the bar

To make either one of my little clubs here, you have to make the grade (or miss it) twice. I’m using only 2008 and 2009 data, although I could go back further and make use of non-PITCHf/x data. And I will, but let’s start with a mix of batted ball and PITCHf/x based criteria.

League averages and criteria

criteria 2008 2009
whiff rate (misses/swings) 0.191 0.197
in wide zone rate (using a two foot wide plate) 0.524 0.526
ground ball rate (per batted ball) 44% 43%
home run rate (per fly balls plus line drives) 8% 8%

The height of the “wide” zone is set for each batter, individually. I aggregate all the PITCHf/x operators’ top and bottom zone values across all available plate appearances for a given hitter.

Making the mark

Not many pitchers make the cut in a single season, let alone two consecutive seasons—31 in 2009, 24 in 2008, with just eight repeats.

Here are the magic eight, along with their two-year stat lines and run values (rv100 is based on actual outcomes, rv100E on expected outcomes derived from batted ball type). Negative run values are better for pitchers, and indicate runs saved per 100 pitches relative to league average. Run expectancies are based on count-by-count wOBA, without regard to base- or game-state.

2008-2009 performance – above average

pitcher # Pitches Whiff IWZ GB% HR/FL% rv100 rv100E
C.C. Sabathia 7099 0.256 0.535 45% 6% -1.11 -0.67
Ryan Madson 2366 0.251 0.539 50% 7% -0.84 -0.66
Tony Pena 2200 0.251 0.551 46% 6% -0.28 -0.14
Felix Hernandez 6554 0.217 0.548 52% 6% -0.87 -0.47
Ubaldo Jimenez 6658 0.215 0.531 54% 5% -0.64 -0.20
Sean Green 2470 0.214 0.561 65% 5% -0.09 -0.02
Josh Johnson 4580 0.211 0.540 49% 6% -0.93 -0.44
Jonathan Broxton 2315 0.295 0.549 51% 5% -1.76 -1.27

Green and Pena stick out like sore thumbs; otherwise my bias seems healthy. My “dream” four-man rotation appears to be Sabathia, Hernandez, Jimenez and Johnson. I’ve got Madson setting-up for Broxton in the bullpen. I’ll have to live with Pena in the seventh and Green in the sixth, but I could do far worse. Talking about doing worse…

Missing the mark

The opposite trick, being below average in each category, is easier to pull off in a single season. Exactly 50 pitchers did the trick in 2008 and 50 did it in 2009. Repeating, however, is much tougher—just three pitchers pulled that off. I’m not using any kind of innings or pitch total cut-off—but all of the repeat offenders (and high-performers) have reasonable pitch counts, at the least.

2008-2009 performance – below average

pitcher # Pitches Whiff IWZ GB% HR/FL% rv100 rv100E
Dave Bush 4403 0.167 0.519 38% 10% 0.10 0.20
Collin Balester 1840 0.166 0.507 38% 11% 1.01 0.57
Wade LeBlanc 1105 0.186 0.489 38% 10% 0.31 0.48

Okay, so LeBlanc pushes the sample size issue, but even when he doubled his workload, and cut his ERA in half, he still missed the mark on all four counts. Balester’s line, and his performance for Washington, seems to make him my nightmare fifth starter, edging out Bush.

Health check

Shocker, my bias is healthy. Next step is to apply it further, down to the minor leagues. Since PITCHf/x won’t be available there, I’ll be using a slightly different set of metrics:
{exp:list_maker}K per 9 IP
BB+HBP per 9 IP (excluding IBB)
HR/FL% {/exp:list_maker}
I have reservations about the last item, as I have another bias—pitchers can control fly ball frequency, but not outcomes. Putting that notion aside, maybe I’ll find some pet prospects or under-the-radar big leaguers that either fit my tastes or offend my sensibilities. Meanwhile, I’ll contemplate becoming a Sean Green or Tony Pena fan.

References & Resources
PITCHf/x data from MLBAM and SportVision
Batted ball data from MLBAM
Run values by the author

newest oldest most voted

Awesome read.  Definitely looking forward to the minor league feature you are working on.  May I add a caveat?  Perhaps set the age cut date at 25/26 so we aren’t looking at 30+ year old journeyman who can outclass Triple-A ball but pitch at a replacement level in the bigs (I’m looking at you Ramon Ortiz!).


So… how does Marc Rzepczynski look under those circumstances?