Sunday, April 11, 2004
Runs, Not Victories, Keep Fans in Stands
The Fort Worth Star-Telegram has analyzed MLB's attendance over the past five seasons. The paper examined eight factors: team payroll, winning percentage, playoff appearances, home runs, runs scored, pitchers' earned run averages, ticket prices and Team Marketing Report's Fan Cost Index. The paper's conclusions:
• Every increase of 100 runs scored brought in 273,160 fans.
• Every $10 million increase in payroll brought in 130,000 fans.
• Every $10 increase in the cost of attending a game brought in 51,372 fans.
(Yes, that last sentence is correct, though the way it's phrased obscures that the causation arrow points the other way. Teams that draw more fans have higher prices because the demand for their product is higher.)
Winning percentage did not have a significant effect, probably because the study didn't control for the new-stadium effect. Cincinnati, Detroit, Milwaukee, and Pittsburgh, in particular, combined terrible teams with new facilities, attracting many more fans than were likely to have attended in the old ballpark.
Update: Prof. J.C. Bradbury, proprietor of the Sabernomics blog, has identified what the Star-Telegram did wrong. Since winning percentage is so closely correlated to the ratio of runs scored to runs allowed (Bill James's "Pythagorean Formula"), by including both winning percentage and runs scored/ERA in its model, the paper effectively counted the same factors twice. When runs scored and ERA are dropped from the model, an improvement of 0.100 in winning percentage can be expected to increase attendance by 100,000 fans/season.
|
The Fort Worth Star-Telegram has analyzed MLB's attendance over the past five seasons. The paper examined eight factors: team payroll, winning percentage, playoff appearances, home runs, runs scored, pitchers' earned run averages, ticket prices and Team Marketing Report's Fan Cost Index. The paper's conclusions:
• Every increase of 100 runs scored brought in 273,160 fans.
• Every $10 million increase in payroll brought in 130,000 fans.
• Every $10 increase in the cost of attending a game brought in 51,372 fans.
(Yes, that last sentence is correct, though the way it's phrased obscures that the causation arrow points the other way. Teams that draw more fans have higher prices because the demand for their product is higher.)
Winning percentage did not have a significant effect, probably because the study didn't control for the new-stadium effect. Cincinnati, Detroit, Milwaukee, and Pittsburgh, in particular, combined terrible teams with new facilities, attracting many more fans than were likely to have attended in the old ballpark.
Update: Prof. J.C. Bradbury, proprietor of the Sabernomics blog, has identified what the Star-Telegram did wrong. Since winning percentage is so closely correlated to the ratio of runs scored to runs allowed (Bill James's "Pythagorean Formula"), by including both winning percentage and runs scored/ERA in its model, the paper effectively counted the same factors twice. When runs scored and ERA are dropped from the model, an improvement of 0.100 in winning percentage can be expected to increase attendance by 100,000 fans/season.
|
