2016 NBA championship win probabilities

My championship probabilities come from combining rankings that use margin of victory in games and closing point spreads in the markets. In these type of ensemble rankings, the aggregation of different methods tends to cancel out the small mistakes of either method, leading to better predictions.

These numbers come from before the start of the 2016 playoffs.

1. Golden State, 58.0%.
2. San Antonio, 20.1%.
3. Cleveland, 10.7%.
4. Oklahoma City, 5.0%.
5. Toronto, 2.0%.
6. Atlanta, 1.6%.
7. Boston, 0.7%.
8. Los Angeles Clippers, 0.5%.
9. Charlotte, 0.5%.
10. Indiana, 0.4%.
11. Miami, 0.4%.
12. Portland, 0.1%.
13. Detroit, 0.1%.
14. Houston, 0.0%.
15. Dallas, 0.0%.
16. Memphis, 0.0%.

These win probabilities come from the futures markets.

1. Golden State, 53.5%.
2. San Antonio, 16.7%.
3. Cleveland, 16.0%.
4. Oklahoma City, 5.0%.
5. Los Angeles Clippers, 2.0%.
6. Toronto, 1.6%.
7. Miami, 1.1%.
8. Boston, 1.0%.
9. Atlanta, 0.8%.
10. Charlotte, 0.6%.
11. Portland, 0.5%.
12. Indiana, 0.4%.
13. Houston, 0.3%.
14. Detroit, 0.3%.
15. Dallas, 0.2%.
16. Memphis, 0.1%.

There was confusion on Twitter last week about these odds, so let me explain how I came up with them.

Before the playoffs started, I grabbed the futures odds for each team to win the NBA title. These odds translate to a probability for each team.

I find the sum of probabilities for all teams, and then I divide each team’s probability by this normalization factor. This gives an implied probability that factors in the large vig taken by the house (normalization factor was 1.25).

The futures markets like LeBron James and Cavaliers more than the numbers do. This comes at the expense of the Spurs.