Short answer: Yes.

Slightly longer answer: Yes he is.

Long answer: After the jump.

On a gamethread the other day, I was thinking out loud and wondered if Max Scherzer was oddly inconsistent in his starts. I mean, we've been told for the past two-plus years that the guy's either fantastic or a horror-show -- so I thought it was about time to crunch some numbers.

It's not a perfect measure of a pitcher's performance in a start, but Game Score tries to measure, in a quick-and-dirty kind of way, how good (or bad) a given start is. Points are assigned for this, taken away for that, and we can debate the merits of the statistic until we're polka-dotted in the face. But hey, it's freely available and people are used to seeing it.

For those of you not familiar with (traditional, math-y) statistics, there are a couple of definitions that need to be given here:

mean
add a list of numbers up and divide by the number of numbers you have

median
take the same list of numbers, arrange then from smallest to biggest, and find the middle-number in the list

standard deviation
attempts to measure how "spread-out" the data are; one SD on either side of the mean should encompass 68.2% of the total data

Assumption: If a pitcher is to be called "inconsistent," the SD of their Game Scores should be bigger than average.

And now, the numbers. Data are from baseball-reference.com (mostly) and ESPN (a little bit). I decided to get data from 2010-12 (the years Scherzer has been with Detroit) for four "good" pitchers who have been in the American League: Justin Verlander, plus 3 of the top 4 vote-getters for the 2010 AL Cy Young. (I skipped Jon Lester because I wanted two righties and two lefties; not sure if it makes a difference or not, but thought I'd be on the safe side). I also wanted to get some "bad" pitchers, so I took the AL league leader in losses, plus the highest ERA amongst eligible pitchers in the AL, in the years 2010-12. Finally, my "average" pitchers are the middle two pitchers ERA, from best to worst, in the AL in each of those seasons. Again, not a perfect methodology, but you gotta start somewhere. (And, n is just the number of starts; relief outings were ignored.)

Table 1: Cumulative Data from 2010-12 (incl. games on June 22, 2012)

So, he's "more inconsistent" than those four very good pitchers since 2010. Let's see how he stacks up against the good (the four in the above table combined), the mediocre, and the bad.

Table 2: Scherzer vs. Groups of Pitchers

Going by SD, "inconsistency" isn't something particular to good pitchers, or bad pitchers, or somewhere in between. Obviously the mean Game Scores tell a story, but there doesn't seem to be a trend with SDs.

Is Max "consistently inconsistent?" His year-by-year totals suggest that he is. So he's got that going for him. Which is nice.

Table 3: Year-by-Year Game Score Statistics

For funsies, I threw in Kid Rick. Felix Hernandez is the only pitcher on this list who's anywhere close.

So, there you have it: Max Scherzer is more inconsistent, by this measure, than a bunch of pitchers. Given the right automated analytical tools (or enough time to do this in a brute-force kind of way), one could crunch similar numbers for every AL starting pitcher. But, using groups of pitchers of varied abilities lends this method some statistical power.

_________________________________________________________________________
* 2010: J Guthrie, J Vargas; 2011: I Nova, M Pineda; 2012: J Beckett, R Colon
** 2010: K Millwood, J Bonderman; 2011: J Guthrie, B Penny; 2012: J Arrieta, G Floyd