/cdn.vox-cdn.com/uploads/chorus_image/image/29753575/146937350.0.jpg)
The goal here is to identify the most average starting pitcher based on the basics of pitching: strikeouts, walks, and home runs. I'm not aiming to do any serious mathematical analyses, or break the mold on how we identify which pitchers are better or worse than average. Quite the opposite actually. My goal here is to quickly and easily identify average pitchers in a way that anyone can replicate if they so choose.
More from our team sites
More from our team sites
The first step is identifying the sample size and how we will cull down the list. The first and most obvious decision is to limit this study to starting pitchers. This was done for a variety of reasons, including but not limited to their consistent workload, general consistency among starters, sample sizes for innings pitched, etc.
The next step is to determine the sample of starting pitchers with which to work. Typically we would use qualified starters, but I arbitrarily lowered the innings pitched requirement to just 100 (instead of 162) to expand our sample size. That leaves 139 pitchers ranging from Clayton Kershaw and his 6.5 fWAR to Jason Marquis and his -1.6 fWAR.
Now to start limiting this group even further, we'll simply start identifying pitchers that fall within arbitrary ranges of the league averages for a given stat. For example, the first data point we'll use is K/9, one of the fundamental components of FIP. The league average starter posted a 7.19 K/9 in 2013, and my "most average starter" should fall within a quarter of a strikeout of that. That gives a range from 6.94 K/9 to 7.44 K/9. That limits my 139 name list of starters who threw at least 100 innings to just 16 pitchers. Included in the list are, in descending order of strikeout rate:
Ricky Nolasco, Bud Norris, Jose Quintana, Ivan Nova, Dan Straily, Dallas Keuchel, David Price, Rick Porcello, Hyun-Jin Ryu, Kris Medlen, Yovani Gallardo, Aaron Harang, Tommy Milone, R.A. Dickey, Chris Archer, and Jeremy Hellickson
That group includes some pretty significant names like David Price and Hyun-Jin Ryu. The question is if any of those bigger names will still be in the running for most average starter of 2013 once walks and home runs are taken into account.
Walk rate is the next step in reducing the list even further, and we'll limit this group to pitchers within half a walk of the league average for last season: 2.83 BB/9. Looking for pitchers ranging from 2.33 to 3.33 BB/9 allows us to remove 7 names from the list of 16 above. Some of the notables to drop off include Nolasco, Price, Ryu, and Medlen. The resulting list is down to:
Norris, Quintana, Nova, Keuchel, Gallardo, Harang, Dickey, Archer, and Hellickson
Last season's league average home run rate was 1.01 home runs per nine innings, which is also the goal to reduce the list of pitchers even further. Looking for pitchers within .20 home runs of the league average rate brings the list down to just 5 pitchers. Our final five starters consist of Bud Norris, Jose Quintana, Dallas Keuchel, Yovani Gallardo, and Chris Archer. After looking at strikeouts, walks, and home runs the list of 139 starting pitchers who threw at least 100 innings in MLB last season is down to just 5 guys with a chance to be the most average.
Their stats, along with the league averages for starters are in the table below.
Any of the pitchers in that table have a justifiable case for being the most average pitcher in baseball, but I'll make the case for just one. My methodology here, keeping with the theme of this post, will be to keep it simple. My determining factor is to take the absolute value of the differences for each of the stats we used above. Doing so allows the easy elimination of Bud Norris, who barely fits into the criteria for each category, with a cumulative total 0.77 points higher/lower than the league averages. For perspective, the maximum possible deviation from the league average to be included in this list is 0.95 (0.50 for K/9 + 0.25 for BB/9 + 0.20 for HR/9).
Other easy eliminations are Jose Quintana and Yovani Gallardo whose differences from the league average were 0.53 and 0.59 respectively. Qunitana's difference comes on the back of a better than league average walk rate (0.31 points better), making up more than half of his score. Gallardo is in the opposite boat with a walk rate 0.49 points higher than league average despite being close in the other two categories.
My final group is down to Chris Archer and Dallas Keuchel. Ironically, these two pitchers show the two sides of luck here, as Keuchel's 4.05 FIP gives hope against his 4.90 ERA, whereas Archer's 3.22 ERA looks awfully inflated compared to his 4.07 FIP.
My choice is Dallas Keuchel, for the simple fact that his rate stats are all incredibly close to the league averages. In fact, Keuchel is within 0.10 points of the league average in all three stats highlighted in this post, despite having numbers that would make him a candidate for demotion to AAA.
Congratulations to Dallas Keuchel, as he's arguably the most average starting pitcher in baseball. At least as far as the 2013 season is concerned. This type of analysis isn't incredibly advanced, and it's not even particularly scientific. However, it does give some insight into which pitchers were most like the average major league starter last season. Even if they don't look the part at first glance.
...
All statistics courtesy of Fangraphs, Brooks Baseball, and Baseball-Reference.
Jeff Long is a writer at Beyond The Box Score and Baltimore Sports and Life. You can follow him on Twitter at @BSLJeffLong.