Thanks to my recent work on NBA Win Probability, I now have a more robust infrastructure for storing and sharing win probability data. So, it was easy enough to tweak my code to provide the same level of detail for tennis. The result:
Win Probability Graphs for the Australian Open
A few callouts:
- My data source is very unofficial, so I make no claims as to the accuracy of each individual data point
- For ease of updating, I am using the following generic probabilities for serve and return:
- Point Probability when Serving: 62.5%
- Point Probability when Returning: 37.5%
- For my US Open graphs, I had taken the extra step of calibrating the point probabilities such that the initial match probability was consistent with the pre-match betting odds. But given the number of matches, it's a step I unfortunately don't have the time for right now.
- I'll do my best to keep these updated daily, but can't make any promises on timeliness
The Most Exciting Match of the Australian Open
As a reminder, the Excitement Index, found at the top right of each graph, measures how far the win probability graph traveled over the course of the match. Because tiebreakers are not allowed on the final set in the Australian Open, you can end up with some rather large values for the Excitement Index. Case in point:
As you can see from the graph, the fifth set was a marathon, with Simon ultimately winning by a score of 16-14. Interesting to note: Despite the length of the match, the cumulative point differential amounted to just one point, in favor of the loser, Brands.
The Biggest Comeback
The comeback factor, shown below the Excitement Index, measures how big of a comeback the winner had to mount. The factor itself is the victorious player's odds of winning at their lowest point. As of January 18, the biggest comeback was Galina Voskoboeva's victory over Irina-Camelia Begu. At her lowest point, Voskoboeva was down 5-1 in the third set and facing Begu's serve (a win probability of 1.4%, or 70-1 odds).