Michael Cave will draw on his experience as a college baseball player (and his 7.5mil row SQL database of baseball data) to walk the audience through the history of baseball analytics.
Almost 20 years ago, Moneyball sent waves through the baseball community by presenting a framework to value players analytically.
Baseball teams have continually advanced their technology in order to capture more data and gain a bigger analytical edge.
This presentation will discuss how far baseball has come, where baseball is now, and where baseball is going with data science.
3. Sports Analytics:
Where did it begin?
• Statistics has also been something that
major sports like to keep. Fans have an
obsession with how many goals, home
runs, points, etc their favorite player
has.
4. What are some
examples?
• Baseball: OPS+, wRC+, wOBA, DRS,
WAR
• Basketball: PER, Win Shares,
ws/48, VORP
• Football: QBR, Catch Probability
• Hockey: Corsi
Baseball was the first major sport to become
more mature in the analytical space.
Basketball is quickly following with
nba.stats.com
Football is following with Amazon NextGen
stats
Hockey is later to the party with analytics
such as Corsi.
5. Fundamental
Analytics
Corsi - advanced statistic used in the NHL to measure shot attempt differential while at even strength
play. This includes shots on goal, missed shots on goal, and blocked shot attempts towards the
opposition's net minus the same shot attempts directed at your own team's net
OPS+ - takes a player's on-base plus slugging percentage and normalizes the number across the entire
league. It accounts for external factors like ballparks. It then adjusts so a score of 100 is league
average, and 150 is 50 percent better than the league average.
eFG% - statistic that adjusts field goal percentage to account for the fact that three-point field
goals count for three points while field goals only count for two points
QBR - statistic created by ESPN in 2011 to measure the performance of quarterbacks in the NFL. It
incorporates all of a quarterback's contributions to winning, including how he impacts the game
on passes, rushes, sacks and penalties. Since QBR is built from the play level, it accounts for a team's
level of success or failure on every play to provide the proper context, then allocates credit to the
quarterback and his teammates to produce a clearer measure of quarterback efficiency
6. How has all of
this changed
the game?
Players are being valued
differently
Team construction
Salaries
7. So where
does Data
Science come
in?
PROJECTION SYSTEMS – A
LOT OF CLUSTERING AND
MIXED EFFECTS MODELING
MICROANALYTICS AKA “IN-
GAME” DECISIONING
PLAYER TRACKING AND IN
GAME ENTERTAINMENT
8. What is being
done post
Moneyball?
In baseball
sabermetrics have
gotten more
advanced when it
comes to increasing
the value of players
via biomechanical
feedback
In football advanced
analytics are showing
the value of certain
positions while
devaluing others.
In basketball
advanced analytics
are determining how
teams run their
offenses and where
shots are taken
In hockey advanced
analytics are showing
the value of simply
taking shots