Your cart

Spring Training: Fresh Grass And Analytics On The Field

By Jay Laramore, analytic training consultant, SAS

Baseball fans in withdrawal since the end of the World Series last October are gearing up for their next fix. The boys of summer will soon begin trudging across the freshly cut, emerald green fields and donning their mitts and gloves for another year’s spring training. This is a ritual I know well from my college baseball career.

Get The Latest Baseball Tech News In Your Inbox!

As today’s players practice their fastballs, fielding and hitting, managers and front-office executives will increasingly embark on another rite of spring to help secure their teams’ advantage on and off the field – they’ll be using analytics.

A Long History of Analytics in Baseball

Baseball’s use of analytics spans more than 45 years. The Society for American Baseball Research (SABR) has performed scientific and historical research, dubbed sabermetrics, on baseball since 1971. Bill James, widely considered the godfather of modern statistical baseball research, has published more than two dozen books on the topic since 1977. And most people are familiar with the 2003 book Moneyball: The Art of Winning an Unfair Game by Michael Lewis (later made into a movie starring Brad Pitt) that detailed how the Oakland Athletics used analytics to build a playoff-caliber team on a shoestring budget.

Turning Insight into Advantage

The idea behind analytics is that even the smallest insight into a player’s tendencies can deliver a huge advantage on the field. As a former college starting pitcher, I experienced this firsthand. One Saturday my team faced off against a small college in Virginia, whom we had every intention of annihilating.

After pitching two and one-third innings, I was dumbfounded when I was pulled for giving up my eighth run. After the game, the opposing coach pulled me aside and explained that when I was about to throw a fastball, my hand began down by my side. But if I was planning an off-speed pitch, my hand began in my glove. Every batter I faced that afternoon had known exactly what pitch I was about to throw.

The Advantages of Modern Analytics

If one passing observation by an opposing coach can have a decisive impact on the outcome of a game, imagine what can happen when you incorporate years of historical baseball data on every player across all 30 Major League Baseball teams. And even as more data becomes available, the way teams extract insight is rapidly evolving. Simply reporting what happened is no longer a primary focus; teams are using data mining, predictive modeling and optimization tools to make more accurate predictions.

One place we’re seeing the impact of modern analytics is in the defensive “shift.” The shift moves players from their original positions in the infield to locations where the current batter is most likely to hit the ball. For example, if you know a left-handed batter is most likely to pull the ball (i.e., hit it to the side from which he hits), the shortstop will slide over to the pitcher’s left to fill the gap between first and second.

But not all teams benefit equally from this tactic. The more accurately teams predict where a player is most likely to hit the ball, the better this strategy works. For example, in the 2014 season, the Houston Astros used the shift 1,562 times and saved 44 net hits. In contrast, the Miami Marlins deployed the shift 309 times and saved negative-three net hits, while the Tampa Bay Rays shifted 1,028 times and only saved four net hits. The shift delivers a competitive advantage for teams that can use big data and analytics to their advantage.

Better Predictions

Teams today increasingly have the opportunity to take advantage of more detailed data to improve their predictions. Historically, teams have collected data on a per-inning or per-at-bat basis. But the new PITCHf/x system collects data on every pitch thrown in Major League Baseball since 2006.

Data collected includes velocity, movement, release point, spin and pitch location. This data can be used to analyze not only a pitcher’s patterns, but also which pitch sequences and locations are most successful against different hitters.

Possible uses for this data include:

  •         Predicting batter success based on pitch type/sequence/location for use in scouting.
  •         Analyzing a pitcher’s velocity, pitch movement and release point during a game to determine when he is growing tired.
  •         Using pitch type and pitch location metrics to determine where to pitch when the shift is used.
  •         Determining successful pitch sequences and types to predict how to get the batter out.
  •         Using a pitcher’s stats to calculate how likely he is to be successful against a particular team.

Team Optimization

Analytics can even help teams better gauge how to field the best team within their budget. Many techniques and forecasting tools are available to determine value and project future production. These tools can answer questions such as:

  •         How much is a particular player worth?
  •         Which free agents are underpriced relative to their market value?
  •         What trade is in the team’s best interest?

Using analytics techniques to answer these questions allows teams to objectively look at a player’s current and future ability as it relates to their expected contract value and compare that to all other available players in the same position.

The Future

Despite the long history of analytics in baseball, we’re still in the pioneering stage of employing these techniques. As today’s analytics trailblazers refine their baseball analysis techniques over time, incorporating strategies based on data mining, predictive modeling and optimization, they stand to benefit tremendously in terms of improved performance in the field. In the future, teams who have refused to employ analytics will be at an increasing disadvantage.

 

 

Jay Laramore is an analytical training consultant with SAS, where he teaches customers how to get the most out of their analytics software. He is a former standout first baseman and designated hitter with the Roanoke College Maroons, where he started all 27 games of the 2009 season. That year, Laramore led the team in HRs and RBIs while leading the team in batting average at .560. Earlier in his college career he was a top starting pitcher, and he is now a key player at leading analytics provider SAS. He can be reached at jay.laramore@sas.com.

 

 

 

 

← Back to SportTechie