The Jose Bautista of the Ivy League

Multiple influences led to the composition of this brief post — and I fully anticipate that the end result will appeal to approximately X readers, where X is an integer less than 1. That said, as my fourth-grade teacher at Broken Ground School in Concord, NH, once told me during an actual student-teacher conference, “Carson, you’re not that special.” Which is to say: there’s a possibility that at least one other person will derive some pleasure from what follows and perhaps, not unlike that same bright star upon which Fievel Mousekewitz and his sister Tanya both wished in 1986 animated musical An American Tail, the current dispatch will allow us to feel less alone in a world populated by talking felines who extort small immigrant mice in return for quote-unquote protection.

Earlier today, my colleague Jeff Zimmerman — a person who, I sense, very much anticipates the return of domestic baseball — asked if I had plans to do any scouting this spring/summer in the the northeast. The short answer is “No” — not because I don’t intend to transport my dumb body to actual games (I do), but rather because, even were I to acquire both a radar gun and a lifetime supply of moisture-wicking polos, I am a mere impostor in this regard.

Zimmerman’s question did compel me to take note, however, of how I live within an hour’s drive of Dartmouth College and how, moreover, the athletic conference to which Dartmouth belongs, the Ivy league, has produced a non-zero quantity of major leaguers (including relevant current ones like Kyle Hendricks, Will Venable, and Chris Young, for example). “There might,” I thought to myself, “there might be some merit in attending some of those Ivy League games.”

“How,” I continued to think, though, “how, in the absence of a readily available scouting report, how might I identify those Ivy League hitters of greatest promise?” It’s the sort of question I’ve typically answered by means of a tool called SCOUT+ (an example of which is present here), which metric combines regressed home-run, walk, and strikeout rates in a FIP-like equation to produce a result not unlike wRC+, where 100 is league average and above 100 is above average. The idea of SCOUT+ is to remove some of the vagaries of luck — in particular the sort that occur with batted-ball data over small samples.

A minor flaw in that methodology, however, is how hitters really do possess signature batted-ball averages. They require at least 1000 plate appearances to identify, those “true talent” BABIPs, but they exist. This is why, for example, Steamer projects Mike Trout to produce a .346 BABIP in 2015; Brian McCann, just a .253 mark.

So, like I say, that’s one likely flaw inherent to the approach I’ve used in the past. An even greater one probably, though, was made clear by means of Chris Mitchell’s recent work on KATOH, a model he’s developed for forecasting a hitter’s future major-league performance with minor-league stats. While Mitchell’s work contains within it a number of valuable observations, the most relevant among them to the current work is that the walk rate of a hitter in the low minors offers almost no predictive value of future value. Age, strikeout rate, isolated power, and even BABIP — they’re all of greater significance.

With a view, then, to acquainting myself more closely with the batters of the Ivy League while also attempting to integrate Mitchell’s work into a better estimate of offensive talent, what I’ve spent too long doing this morning/afternoon is devising another index hitting stat (like wRC+) — except, in this case, one that utilizes not regressed home-run, walk, and strikeout rates, but rather (to better reflect Mitchell’s findings) regressed strikeout rate, isolated power, and BABIP.

I’ll discuss some particulars below. In the meantime, here are the top-10 hitters from the 2014 Ivy League season by what I’ll temporarily call KATOH+. Note that xK%, xISO, and xBABIP denote expected strikeout rate, isolated power, and batting average on balls in play, respectively.

Some assorted observations:

• The relevant league averages for the Ivy League in 2014 were as follows: 17.3% strikeout rate, .097 isolated power, .310 BABIP.

• At the major-league level, the combination of those three variables (K%, ISO, and BABIP) correlates very strongly with wRC+, producing a 0.88 r-squared.

• Note that that expected BABIPs in the leaderboard above almost all fall in the low-.300 range. This is due to the extreme influence of regression on that metric and the relative paucity of plate appearances produced by Ivy League hitters over the course of the season. The full range of xBABIPs in the Ivy League last year was .295 to .322, or roughly 10-15 points on either side of the league average.

• Among all Ivy League hitters, Columbia’s Gus Craig produced the best offensive line by this methodology, producing an expected strikeout rate (8.9%) of roughly half the league average and expected ISO (.208) roughly twice the league average. As suggested by the title of this post, this combination of contact and power roughly equivalent to Jose Bautista’s in the majors. Which is not to say that Craig is Bautista’s equal, but which is to say that as Bautista is to Major League Baseball, so Craig is to the Ivy League.

• Here’s more information concern Craig. He’s from Oregon. He generally plays right field. He’s entering his senior season. He was omitted from Baseball America’s preview of the 2014 Ivy League season last February.

• Finally, among those players who exhibited a promising offensive line and also occupied a place towards the more challenging end of the defensive spectrum, U Penn’s Ryan Mincher is most notable. He started all 41 of the school’s games at shortstop last year in just his sophomore season while also supplying the promising offensive line noted above.

• The complete stats for the Ivy League are available here.