For two years in a row, my attendance prediction model has come extremely close to predicting the actual attendance at the team's win level.
In 2009's prediction, my model overshot actual attendance by 1.15%. In 2010, it overshot actual attendance by 0.85%.
The model has been updated to include data from 2010.
Again, the model predicts an increase in attendance. At last year's win level -- 90 wins -- the model predicts an average attendance of 33,645 per home game. To fall below last year's attendance level, the model says that the Rangers would have to win fewer than 77 games.
Coming off a World Series appearance, it will be interesting to see how the model holds up for 2011.
Regression Notes
The standard error is down from last year's 2,602 attendees per game to 2,560. The R-square and Adjusted R-Square values are nearly identical to the previous year's -- all three years have been right around 0.90 for both values.
Thanks to the accuracy of last year's prediction, the t Stat and P-value numbers for all three independent variables improved. The growth factor variable (inflation) is still the least significant of the three with a t Stat of 1.415, but again, removing it from the calculations results in larger errors.
Playoff Probability
There were no significant changes to the playoff probabilities for each win level in the AL West. The 90% barrier is crossed at 95 wins, and the 50% barrier is crossed at 91 wins.
I'm keeping it short-and-sweet this time to avoid repeating what I've said in the past. If you'd like to read my previous articles, which are good if you'd like to read about how I constructed my model, check out the links below:
Texas Rangers Win-Curve Part I: Wins vs Attendance
Texas Rangers Win-Curve Part II: Playoff Probability
2010 Texas Rangers: Wins, Attendance, and Playoffs
If you're fascinated by this stuff and haven't read Vince Gennaro's book Diamond Dollars, I strongly encourage you to take a look at it.