Sabermetrics: Baseball Analytics and the Science of Winning

Though the term “sabermetrics” has only been around since the early 1980s, the practice of using statistics to gain a winning edge in baseball goes back almost two centuries. As the tools of sabermetrics continue to advance, teams like the New York Yankees and Mets are becoming more adept at using analytical tools to dissect the science of winning.

By the Numbers: What is Sabermetrics?

Sabermetrics is a science of sport. It is the empirical analysis of baseball through statistics, used to predict the performance of players, giving teams a winning edge.

With the help of sabermetrics, teams can:

  • Forecast results by making predictions based on previous data
  • Analyze on-field performance by recording and evaluating important aspects of play
  • Assist in decision-making by offering objective insights into players’ performance, matchups, and scouting prospects

Who’s a better pitcher? Where should our outfielders play? Which player is a better value for our team? Who was the greatest second baseman of all time?Thanks to sabermetrics, all of these questions can now have objective answers.

Beyond the Box Score: A Brief History of Sabermetrics

Just as the origins of baseball are difficult to pin down, so too is the start of sabermetrics. The term itself was coined in 1980 by renowned baseball analyst Bill James. Named in honor of the Society for American Baseball Research (SABR) of which James is a member, he called sabermetrics “the search for objective knowledge about baseball.” However, that’s not where the story begins. From simple scorekeeping to the more complex statistics that define the game today, statistics have long been important in baseball.

Let’s take a closer look at key milestones in the making of sabermetrics.

A Brief History of Sabermetrics

Sabermetrics is a science of sport. It is the empirical analysis of baseball through statistics, used to predict the performance of players, giving teams a winning edge.

1845

The first baseball box score, which tallied only batters’ runs and outs, appears in the New york Morning News

1859

The New York Clipper publishes sportswriter Henry Chadwick’s box score, which, in addition to runs and outs, also includes hits, assists, and errors.²

1872

Fan Hervie Alden Dobson writes to Chadwick proposing a new measure, “hits per time at bat,” which would go on to become the standard gauge of batting-the batting average.³

1911

After joining the staff of Baseball Magazine, sportswriter Ferdinand Cole Lane devises his own values for singles, doubles, triples, and home runs in an attempt to overcome the inadequacy of simply using batting averages as a performance indicator.&sup4;

1947

Hired by Brooklyn Dodgers President Branch Rickey, Allen Roth becomes the first full-time statistician working for a major league team.&sup5; Roth goes on to calculate on-base percentages, batting averages with runners in scoring positions, performance in ball-strike counts, and more.&sup6;

1968

Earl Weaver joins the Baltimore Orioles as a manger. His devotion to statistics leads him to create a combination of note cards and loose-leaf note books tracking head-to-head matchups between pitchers and hitters.&sup7;

1971

The Society for American Baseball Research forms in Cooperstown, N.y., to promote the exploration of baseball’s history and statistics.&sup8;

1977

Bill james self-publishes his first edition of Baseball Abstract, an annual collection of statistical insights. It was one of the first books of its kind to reach a mass audience.&sup9;

1981

STATS Inc., a sports data technology company, is founded and begins development of Edge 1,000 software meant to assist teams in keeping their own statistics.&sup9;

2003

Journalist Michael Lewis publishes Moneyball: The Art of Winning an Unfair Game (on which the 2011 film is based), detailing Oakland Athletics General Manager Billy Beane’s use of sabermetrics to compensate for the team’s relatively small revenue.¹&sup0;

Sabermetrics in Action

Now that we know what sabermetrics is and how it came to be, we can demonstrate baseball analytics in action with a side-by-side comparison of two players.

The value of sabermetrics is that it cuts through the hype to reveal the realities of the game. Let’s take a look at two of New York’s most talked about players, the Met’s Yoenis Cespedes and the Yankees’ Alex Rodriguez.

By more traditional metrics such as Batting Average (Hits / At Bats), Home runs, and OnBase Percentage (Walks+Hits)/(At Bats+Walks+Sacrifice Flies), these two players’ offensive abilities would be considered fairly comparable.

2015 Stats

Yoenis Cespedes (Mets)

  • Batting Average (BA): .287
  • Home Runs (HR): 35
  • On-Base Percentage (OBP): .328

Alex Rodriguez (Yankees)

  • Batting Average (BA): .250
  • Home Runs (HR): 33
  • On-Base Percentage (OBP): .356

Using more complex sabermetrics such as Total Bases [Singles + (2 X Doubles) + (3 X Triples) + (4 X Home Runs)], Slugging Percentage (TB / At Bats), and On-Base Plus Slugging (OBP + SLG), however, the statistics clearly show that the Mets are getting a significantly better value out of Yoenis Cespedes than the Yankees are getting out of Alex Rodriguez.

2015 stats

Yoenis Cespedes (Mets)

  • Batting Average (BA): .287
  • Home Runs (HR): 35
  • On-Base Percentage (OBP): .328
  • Total Bases (TB): 343
  • Slugging Percentage (SLG): .542
  • On-Base Plus Slugging (OPS): .870
  • Annual Salary: $3.7 million

Alex Rodriguez (Yankees)

  • Batting Average (BA): .250
  • Home Runs (HR): 33
  • On-Base Percentage (OBP): .356
  • Total Bases (TB): 254
  • Slugging Percentage (SLG): .486
  • On-Base Plus Slugging (OPS): .842
  • Annual Salary: $22 million

Numbers are based on regular season 2015 stats.
All data is from baseball-reference.com.

Tools Of The Trade: Software that Plays Hardball

Sabermetrics is made possible in part because every baseball game generates a ton of recorded data. But how is it processed? “Sabermetricians,” as experts of sabermetrics are sometimes known, use a variety of tools to record games and glean insight from this data. These tools can be as simple as formulas used to derive statistics and as complex as high-definition cameras coupled with recognition software to analyze plays.

Each sabermetric tool has its uses and drawbacks, but some are more commonly used than others. Here are a few of the tools you’ll find in the modern sabermetrician’s toolbox:

FIELDf/x

FIELDf/x digitally records the position of all players and hit balls in real time using live camera feeds and object-recognition software.

Calculates the difficulty of a catch, probability of a particular player making a catch, and the true defensive range (plays that might be made).

Part of the baseball product suite of Sportvision, a company providing television-viewing enhancements.¹¹

BaseRuns Estimator

The BaseRuns Estimator estimates the number of runs a team should score given their offensive statistics and the number of runs a hitter or pitcher creates or allows.¹²

  • Invented by sabermetrician David Smyth, this is just one of many run estimator formulas.
  • Runs Scored=Baserunners x Percentage of Runners Who Score + Home Runs
  • Percentage of Runners Who Score is estimated according to hitting performance.

Statcast

Statcast measures everything from the perceived velocity of a pitch (derived from the velocity at the exact release point) to a hit’s hang time, distance, and projected landing point.

Uses high-resolution cameras and radar equipment to track the location and movement of the ball and every player on the field at all 30 MLB parks.

Incorporated into every MLB Network Showcase game by the league.¹³

Who’s Keeping Score? Careers in Sabermetrics

The rise of sabermetrics has revolutionized not only how managers and coaches make decisions on the field, but also in how teams hire talent for front office positions. Historically, jobs such as general managers and assistants were reserved for former players. Thanks to sabermetrics, these front office positions have now opened up to those able to analyze sabermetrics data and help implement those insights in operations and scouting departments. Unsurprisingly, the industry is piquing the interest of many. In 2013, undergraduate students at Syracuse University’s Falk College of Sport and Human Dynamics formed the Baseball Statistics & Sabermetrics Club. And when EdX first launched a free sabermetrics 101 course, more than 13,000 people signed up in 2014. The interest in the business analysis of baseball continues to evolve.

For those interested in a career in sabermetrics or who want to become experts in discovering meaningful patterns in data, it is important to do a bit of research on business analytics and data science occupations to determine if sabermetrics is a suitable career path. Ideal candidates for sabermetrics jobs not only need an understanding of sabermetrics (and an interest in baseball), but also need formal training in statistics possessed by analytics professionals, data scientists, and MBA graduates.

Resources:

  1. http://sports.espn.go.com/mlb/columns/story?id=1835745
  2. http://www.npr.org/templates/story/story.php?storyId=106891539
  3. http://books.google.com/books?id=TFNK7be-KGcC&pg=PA281&dq=Hervie+Alden+Dobson+Baseball&source=bl&ots=Cd4NEjZJOu&sig=aal_xzrcQINfg5riwczzpZhNLc&hl=en&sa=X&ved=OCEYQ6AEwBm
  4. https://britannica.com/sports/sabermetrics#ref1182353
  5. http://www.nytimes.com/1992/03/05/sports/alan-roth-74-dies-baseball-statistician.html
  6. http://sports.espn.go.com/mlb/colums/story?columnist=schwarz_alan&id=1835745
  7. http://sabr.org/bioporj/person/0cfc37e3
  8. http://sabr.org/about
  9. http://www.newyorker.com/magazine/2003/07/14/the-professor-of-baseall
  10. http://sports.espn.go.com/mlb/columns/story?columnist=scwarz_alan&id=id=1835745
  11. http://www.sportvision.com/baseball/fieldfx%C2%AE
  12. http://www.hardballtimes.com/the-great-run-estimator-shoutout-part-1/
  13. http://m.mlb.com/news/article/119234412/statcast-primer-baseball-will-never-be-the-same