In Search of the Averagest Player in the League

Here’s a little insight into my writing process. When I turn on my computer in the morning, my mind completely devoid of ideas apart from the knowledge that Meg is going to message me in a couple hours asking if I plan on working today, the first thing I do is look at our leaderboards. Maybe just seeing a name will jog something loose, or maybe I’ll learn about someone doing something exceptionally good or bad.

It’s fun to write about the extremities of baseball, and fun to read about them. It’s why we fight over who gets to write about Aaron Judge, or Paul Skenes, or the White Sox. We aim to please.

But I also have a soft spot in my heart for the unremarkable. My very first week on this job, I wrote an ode to Cal Quantrill, declaring him “the averagest pitcher north of the Rio Grande.” Well I’ve been noodling on averageness. Who’s the anti-Judge or anti-Skenes? The anti-Jose Altuve? Who is the least remarkable player in baseball?

I decided to approach this issue from the perspective of what a player — a position player, for the purposes of this exercise, so that I might repeat it for pitchers if there’s a slow news day later in the season — is expected to accomplish on the field. Well, he has to hit, he has to play defense, and he has to run the bases. Good news: We have a number to measure each of those. There’s wRC+ for offense, and then the WAR components for baserunning and defense.

And maybe just because I had the idea of clutchness on my mind from earlier in the week, I decided to control for the idea that an average hitter might be preternaturally good or bad in big moments by including a fourth number: win probability added.

Now, all four of those stats are conveniently scaled to league average: wRC+ to 100, the other three to zero. But getting to 100, or zero, would be more indicative of a player who’s made no impact than a player who has made an average impact. So I set a playing time threshold — 200 plate appearances — which ensured not only a certain level of participation, but also a certain level of quality. We don’t want the conception of “average” to be skewed too far by Quad-A guys who Moonlight Graham it for a week before being sent back down.

Which means while the mean and median for all of those stats is close to the scaled average, it’s not exactly the average. The mean wRC+ is all the way up at 102, thanks in part to Judge, with his 216 wRC+, breaking the curve. It’s not a perfect curve, at least for wRC+, but it’s still recognizable as one.

Having calculated the mean and median for each of the four stats, we can get to work defining “average.” A good place to start is to find players who are within one standard deviation of the median in all four categories.

Unfortunately, that limits the sample from 303 all the way down to… 90. Which I should’ve anticipated. Not only is 74 to 128 a massive range for wRC+ (any parameter that gives you both Pete Alonso and Vidal Bruján is probably overly broad), but definitionally more than two-thirds of a normal sample is going to end up being within a standard deviation of average. If there were no correlation at all between these four stats, you’d expect to end up with more than 20% of the population being within a standard deviation of the norm in all four. As it stands, those 90 names comprise almost 30% of the players with 200 or more PA this year.

That’s too broad a definition of “normal.” Let’s narrow it down to half a standard deviation. That turned up 17 names. But limiting it to one-third of a standard deviation each direction for all four stats? That narrowed the field to three names.

So, aspects of this group make a lot of sense. I like that we’ve got a second baseman, a corner outfielder, and a guy who plays both second base and corner outfield. If you asked me what the most average position in baseball was, I’d say either second base or right field. Biographically, the only way to create a more generic-sounding ballplayer than a guy named Colt Keith from Mississippi is to have a guy named Jesús Sánchez from the Dominican Republic.

Team-wise, I think we could do a little better in terms of seeking out “average.” Feels like a list of truly average players would include at least one Brewer or Guardian, but this list of three is solid.

Nevertheless, I’m not sure I trust a list of 99th-percentile average guys that has Schneider on it. His results might be average, but he’s a short guy (albeit with very tight pants) and a mustache and glasses that make him one of the most distinctive-looking players in baseball. Plus he’s got a very particular, arguably extreme, offensive approach. Should I be focusing more on physical appearance, then?

So I narrowed down the field using a Stathead search, using height, weight, and age. Now, two immediate caveats off the top. First, the Stathead search only returns an integer for age. Some players are not actually the age at which they’re credited as playing this season. For instance, Manny Machado is in his age-31 season, but because he was born six days after the seasonal age cutoff, he’s actually already 32. Second, some of these guys are not actually as tall or as heavy as their listed dimensions. There are several MLB players — I won’t name names — who are playing under a listed height that wouldn’t fly as a fib in a Tinder profile, and who haven’t been weighed since three Batmans ago. So take all that with a grain of salt.

Unfortunately, adding in biographical information doesn’t narrow the field much more quickly. There were 41 players who ended up within a standard deviation of the median in all four statistical categories and all three biographical categories. Once again, it was necessary to cut the range by half. This time, doing so cut the sample to three at only half a standard deviation from the median.

There we go. Austin Hays is the name you’d come up with if “Colt Keith” were on the tip of your tongue but you couldn’t quite remember him. Peña is more conspicuous than you’d like from an avatar of the forgettable — the man is a Gold Glove winner and World Series MVP — but I do like that he splits the difference between being born in the Dominican Republic and having been drafted out of an American college.

And then there’s Sánchez again. Seems to me that, as the sole survivor of the great phenotypic culling (Keith and Schneider are both too young; Schneider is additionally too small), Sánchez is the slam dunk answer to “Who is the averagest player in the league?”

An outfielder whom the Marlins have been trying to develop and/or trade for half a decade seems like a pretty well-trodden biographical path. And yet, there are distinctive things about Sánchez’s game. He hits the ball on the ground a lot and has an above-average strikeout rate. Shouldn’t those qualities factor into a player’s averageness?

So, having found a perfect spot to take a knee, run out the clock, and file my story, I decided to run another play. I went back to my spreadsheet and added eight new categories, capturing each player’s strikeout and walk rates, plus their batted ball distribution both horizontally (Pull%, Cent%, Opp%) and vertically (GB%, LD%, FB%).

Adding everything together, only two players are within a standard deviation of the median in all 15, yes, 15 categories: Connor Wong and Dominic Smith.

And you know what? That doesn’t sit right. It feels overdetermined, with too many parameters with too broad a range. Particularly because of how close Sánchez came to making the one standard deviation cutoff in all 15 categories.

(You know how I know there are too many categories here? This chart is now too wide to fit on the page without adding a scroll bar.)

Sánchez missed in three of 15 categories: He hits too many balls to the opposite field and too few to pull, and his groundball rate was too high by less than a percentage point. These feel like trivial quibbles. When you tell your grandkids about Jesús Sánchez, the averagest ballplayer who ever lived, are they going to snipe back about how he’s too much of a spray hitter?

I think not.

So I return to my original conclusion: Jesús Sánchez is the most average position player in the league. Time to get him the least distinctive trophy in the shop.