Much of the content on this blog falls under the header of “Research”–I’ve been using numbers in an attempt to solve many unanswered questions.  Click here for my most recent articles in that category.

Before I began writing, I published a number of tennis mini-studies on my personal blog, The Summer of Jeff.  Here’s a directory of some of those studies:

  • My main focus has been the predictiveness of ATP rankings.  I started with a simple algorithm based on ATP rankings.  I’ve looked at giving players credit for the quality of their opponents and the systematic underrating of wild cards.
  • Closely related is home court advantage: 12
  • Using some of that research, I generated forecasts for a few grand slams before the event started: 2009 US Open2010 US Open2011 Australian
  • Probability of various outcomes, at the game, set, and tiebreak level.  I published my source code.  Also: how theory differs from practice.
  • Based on the above, win probability and volatility graphs: 1, 2
  • Is there an advantage to serving first?
  • Server dominance by surface.
  • Who plays and wins the most tiebreaks?
  • Who wins the most points, and who wins matches while losing more than half of total points?
  • Rating players by peak ranking.
  • Variance in draw quality in ATP 250s.
  • The bias resulting from seeded draws.
  • Results for different types of wild cards.
  • Prevalence and success of lefties: 1, 2

Leave a Reply

Your email address will not be published. Required fields are marked *