How MLB Teams, Draft Analysts Navigate Tricky College Baseball Run Environment

Image credit: ARLINGTON, TX - MAY 24: Kansas outfielder Mike Koszewski (2) watches a game winning home run ball for Oklahoma go over the fence during the 2024 Phillips 66 Big 12 Baseball Championship game between Oklahoma and Kansas on May 24, 2024, at Globe Life Field in Arlington, TX. (Photo by David Buono/Icon Sportswire)

Over the past 25 years, the data revolution has transformed the MLB draft. The statistical tools the Oakland A’s used in their 2002 Moneyball draft are as quaint and outdated nowadays as flannel uniforms and Ballantine beer.

The analytical tools teams use to evaluate college and high school players now would have seemed like science fiction just two decades ago. Teams measure how hard hitters hit every ball they put in play. They see how a hitter performs against different pitches, different locations, different quality of pitchers or a thousand other data points, all of which can be sliced or compared to the rest of the draft pool in a thousand other ways.

And nowadays, a scout can watch every pitch, or every defensive play a shortstop made all season, or throughout his college career, with just a couple of clicks of a mouse.

But that massive amount of data hasn’t made the job any easier as teams try to adjust and comprehend one of the weirdest environments college baseball has ever seen.

In an age where it feels like we’re getting closer and closer to measuring all of the game’s minutia, the college run environments of the last two seasons have become a tricky problem without an easy fix.

The Reality Of The College Run Environment

Offense in college baseball is currently at all-time highs. Last year, Division I baseball broke the all-time home run record, as team’s averaged 1.14 home runs per game. As the NCAA tournament begins, that record seems assured of being broken again, as teams are averaging 1.16 home runs per game this year. Just a decade ago, teams averaged 0.39 home runs per game. Runs per game are near records as well–teams average 6.85 runs per game this year, which would be the sixth highest mark of all-time.

By itself, a changing offensive environment is a perfect example of how using data can help humans better process information. As humans, we have a tendency to anchor to nice round numbers: hitting 20 home runs in a college season seems impressive. But hitting 20 home runs in 2024, when hitters are averaging a home run every 34.7 plate appearances isn’t the same as it was in 2014, when hitters averaged a home run every 98 plate appearances.

Whether it’s counting stats (home runs), rate stats (batting average, on-base percentage or strikeout rate) or more granular data (exit velocity, pitch movement) analytical models allow analysts to easily compare players to each other and to “average” even when what is average changes dramatically.

But that only works if everyone is working from the same baseline. And MLB teams are not as confident that they can fully account for everything that is going on in college baseball currently.

Is It The Bats Or The Balls?

No one disputes that offense has skyrocketed in college baseball. What everyone wants to know is why. If the spike in home runs and offense is because the current baseball is way “hotter” than baseballs were five or 10 years ago, that’s a factor that teams can account for. As long as everyone is using the same type of baseball, this is a universal factor.

When it comes to the bats, that’s where everything gets more complicated. Some bats are viewed by college teams to be “hotter” than other bats, even while remaining within the realm of meeting BBCOR standards. That’s a factor that is dramatically different than pro ball, where there are very few differences between wood bats.

But as Baseball America reported last year, multiple coaches and players believe that some players are managing to sneak into games non-compliant bats that perform far beyond the BBCOR standards.

The NCAA altered bat testing this year, going from testing once a weekend to testing every day, but as long as it’s a sticker-based system with few if any punishments for using a non-compliant bat, the risks of being caught are quite low.

And the benefits could be dramatic. In addition to helping his team win more games, a hitter could transform his draft status if an illegal bat can add just a few mph of exit velocity. Exit velocity plays a significant role in many team’s draft models.This is the struggle front offices face when evaluating potential draft picks.

“Is it the bats? Is it the balls? Both? We don’t really have any ability to say this player is using an illegal bat and it’s impacting his performance this much,” said an analyst for an NL team.

Analysts add another step to their evaluation process, as noticeable spikes in exit velocity and home run production need further detective work. That often becomes the job of an area scout, and is an additional way an area scout can provide a massive edge for a team.

“The biggest issue is we can perhaps get information on a particular player using a hot bat, but how do we know to what degree the bat is aiding his production? We don’t know how the sweet spot plays on a juiced Marucci vs a juiced DeMarini. So even if we get the information there’s no formal way to adjust that,” said an analyst with an American League team.

Playing The Percentages

We spoke with analysts from five different front offices to get a clearer idea of how front offices are navigating the offensive environment in college baseball. While each analyst voiced concerns around the potential impact of juiced bats and balls, the environment itself wasn’t a difficult task to tackle.

One analyst with a 2023 playoff team showed confidence in his organization’s process and ability to spot signal from noise.

“Performance numbers and how parks play is generally handled pretty well by current year park factors in a performance model. Xstats (expected statistics) based on exit velocity (EV) and launch angle data help account for any juiced ball concerns. Now as far as hot bats and exaggerated EV data, for me, I like to look at it as where a player sits within the population of college players rather than looking for a specific EV number. Assuming that the general talent level has remained similar, one standard deviation above average is still as valuable as one standard deviation above average was five years ago.”

This was a consistent response from each analyst we spoke with. While each team might have slight variations of how they achieve this, each analyst described a process of standardizing the field and weighing factors of performance, exit velocity, launch angle as well as contact/approach data to remove noise. Statistical tools adjust for quality of competition, park factors and a multitude of other “noise” that could explain why one .330/.420/.550 hitter is actually more impressive than one who’s hitting .380/.520/.700.

“We really just look at the percentages” is how one analyst put it. “We standardize everything to the SEC level and then we factor in park factors, batted ball data etc. This allows us to sort of look at everything on an even playing field. You don’t get caught up focusing on the home run total or the 90th percentile EV, instead you look at where he ranks percentage-wise against the field.”

So if everyone’s on a similar playing field, it doesn’t matter if a hitter is hitting the ball consistently 110 mph in a college environment where 110 mph exit velocities are top tier or 118 mph where 118 mph is top tier.

Where your skills rank across the larger pool of draft prospects is still the measuring stick. Certainly we’ll see some players falter as professionals, products of tampered bats and potentially juiced balls. But teams are confident they’ll be able to spot the signal from the noise.