charge from WIRE

[beanumber]

Given your expertise and background, a review article from your group related to the topic of Statistical tools for sports analytics (e.g. baseball) would significantly contribute to WIREs Computational Statistics.

The article should critically present the current state-of-the-art research, include opposing viewpoints, and identify challenges and opportunities.

[beanumber]

Potential topics

• team strength
  • Bradley-Terry
  • Bayesian state space models
  • PageRank strength of schedule
• player ability
  • WAR: NFLWAR, openWAR, hockeyWAR
  • ELO ratings
  • tracking data
• situational / possession value
  • expected run value / expected points by field position / possession value /
  • win probability
• visualization
• competitions
  • BigDataBowl
  • Kaggle

Big Ideas in sports analytics:

• How good is this team?
  • In the past? In the present? In the future?
• How good was/is/will be this player? How valuable is this player to their team?
  • Past/present/future?
• What is the probability of this team winning at this point in the game?
• What is the impact of this rule/scoring system/ranking system/tournament?
• Where do you get the data to work on this stuff? What packages? How does the granularity of the data impact the kinds of analytics you can do?
• Are there any novel sports analytics methods that have been used in non-sports applications?

[beanumber]

• importance of visualization
• relevance to sports industry
• broad but shallow