mfosterftw
9/17/2008, 10:01 AM
I was looking around some sites and found this interesting article from the Journal of Sports Economics.
Update: I've been reminded of the need to seek permission before posting articles such as this one in their entirety. As such, I'm reducing it to the abstract, conclusions, and references.
Evidence of Television Exposure Effects in AP Top 25 College Football Rankings
JOURNAL OF SPORTS ECONOMICS, Vol. 8 No. 4, August 2007
NOEL D. CAMPBELL - University of Central Arkansas
TAMMY M. ROGERS - North Georgia College & State University
R. ZACHARY FINNEY - University of South Alabama
Abstract
A potential source of bias in the Associated Press (AP) Top 25 football rankings is television exposure. Using the 2003-2004 and 2004-2005 college football seasons, the authors observe, all else equal, that AP voters change the ranking of teams differently on the basis of television exposure: The more often a team is televised, relative to the total number of own- and opponent-televised games, the greater the change in the number of AP votes that team receives, even after accounting for own and opponent’s on-field performance.
Keywords: college football rankings; sports rankings; rankings bias; sports television exposure
CONCLUSION
The obvious weakness of our results is that they derive from two seasons’ worth of data. Having noted this caveat, we find evidence of television exposure bias in the AP Top 25 college football rankings. Although not a dominating effect, television exposure is significant and potentially decisive given that relatively small differences in AP points can move a team up or down several places in the rankings. This becomes especially significant for strong teams in the “midmajor” conferences seeking access to the highly lucrative BCS bowls. BCS rules at the time of this writing state,
Any Division I-A independent or team from Conference USA, the Mid-American, Mountain West, Sun Belt or Western Athletic Conferences will earn a guaranteed slot in one of the BCS games should that team be ranked sixth or higher in the final BCS Standings—unless more than two teams meet this criterion. (“Bowl Championship Series,” n.d.)
In this BCS index, AP points constituted one third of a team’s ranking, with the ESPN Coaches Poll and an average of six computer polls (while dropping the highest and lowest two rankings) constituting the other two thirds. Compared with teams from the BCS-founding conferences, teams from the midmajors typically get less television exposure. Therefore, even for similar performance, AP voters were biased against midmajor teams, making it more difficult for them to get into BCS bowls.
However, the AP dropped out of the BCS poll in 2005. We believe that our results retain relevance, even after the AP withdrew from participation in the BCS. In July 2005, the BCS announced that it had replaced the AP poll with the Harris Interactive College Football Poll. The Harris Interactive poll would be included in the BCS rankings in the same fashion as AP poll. The Harris Interactive poll, like the AP poll before it, is composed of voting panelists expressing their beliefs about the relative strengths of different football teams. Although we have no evidence to date, because of the similarities in construction between the AP poll and the Harris Interactive poll, we believe that the Harris Interactive poll will exhibit the same television exposure bias we observed in the AP poll. Thus, we believe that the position of teams from the midmajors has not changed: even with similar performance, midmajor teams will face greater difficulty in obtaining invitations to BCS
bowl games.
REFERENCES
Borland, M. V., Goff, B. L., & Pulsinelli, R. W. (1992). College athletics: Financial burden or boon? In G. Scully (Ed.), Advances in the economics of sport (pp. 215-235). Greenwich, CT: JAI.
Bowl championship series. (n.d.). Retrieved October 4, 2004, from http://www.bcsfootball.org
Gildea,W. (1998, January 5). Ranking the teams is not as easy as π. The Washington Post, pp. D1, D4.
Goff, B. (1996). An assessment of path dependence in collective decisions: Evidence from football polls. Applied Economics, 28, 291-297.
Lebovic, J., & Sigelman, L. (2001). The forecasting accuracy and determinants of football rankings. International Journal of Forecasting, 17, 105-120.
McCormick, R., & Tinsley, M. (1988). Athletics versus academics? Evidence from SAT scores. Journal of Political Economy, 95, 1103-1116.
Noel D. Campbell is an associate professor of business administration (economics) at The University of Central Arkansas. He earned a PhD in economics from George Mason University. His research interests include lotteries, merit-based financial aid, and state-level public finance. With R. Zachary Finney, he recently published “Mitigating the Combined Distributional Consequences of the Georgia Lottery for Education and the HOPE Scholarship” in Social Sciences Quarterly.
Tammy M. Rogers is an assistant professor of business administration (finance) at North Georgia College & State University. She earned a PhD in finance from Texas Tech University. Her research interests include the regulation of banks and other financial institutions, financial markets, and corporate finance. With Melissa Williams, she recently published “Financial Institution Stock Reaction to the Suspension of the Thirty-Year Treasury Bond” in International Journal of Business Disciplines.
R. Zachary Finney is an assistant professor of marketing at the University of South Alabama. He earned a PhD in marketing from the University of Alabama. His research interests include strategic management, strategic marketing, and topics in the resource-based theory of the firm. With Noel D. Campbell, he recently published “Strategies and Resources: Pathways to Success” in Journal of Business Research.
Update: I've been reminded of the need to seek permission before posting articles such as this one in their entirety. As such, I'm reducing it to the abstract, conclusions, and references.
Evidence of Television Exposure Effects in AP Top 25 College Football Rankings
JOURNAL OF SPORTS ECONOMICS, Vol. 8 No. 4, August 2007
NOEL D. CAMPBELL - University of Central Arkansas
TAMMY M. ROGERS - North Georgia College & State University
R. ZACHARY FINNEY - University of South Alabama
Abstract
A potential source of bias in the Associated Press (AP) Top 25 football rankings is television exposure. Using the 2003-2004 and 2004-2005 college football seasons, the authors observe, all else equal, that AP voters change the ranking of teams differently on the basis of television exposure: The more often a team is televised, relative to the total number of own- and opponent-televised games, the greater the change in the number of AP votes that team receives, even after accounting for own and opponent’s on-field performance.
Keywords: college football rankings; sports rankings; rankings bias; sports television exposure
CONCLUSION
The obvious weakness of our results is that they derive from two seasons’ worth of data. Having noted this caveat, we find evidence of television exposure bias in the AP Top 25 college football rankings. Although not a dominating effect, television exposure is significant and potentially decisive given that relatively small differences in AP points can move a team up or down several places in the rankings. This becomes especially significant for strong teams in the “midmajor” conferences seeking access to the highly lucrative BCS bowls. BCS rules at the time of this writing state,
Any Division I-A independent or team from Conference USA, the Mid-American, Mountain West, Sun Belt or Western Athletic Conferences will earn a guaranteed slot in one of the BCS games should that team be ranked sixth or higher in the final BCS Standings—unless more than two teams meet this criterion. (“Bowl Championship Series,” n.d.)
In this BCS index, AP points constituted one third of a team’s ranking, with the ESPN Coaches Poll and an average of six computer polls (while dropping the highest and lowest two rankings) constituting the other two thirds. Compared with teams from the BCS-founding conferences, teams from the midmajors typically get less television exposure. Therefore, even for similar performance, AP voters were biased against midmajor teams, making it more difficult for them to get into BCS bowls.
However, the AP dropped out of the BCS poll in 2005. We believe that our results retain relevance, even after the AP withdrew from participation in the BCS. In July 2005, the BCS announced that it had replaced the AP poll with the Harris Interactive College Football Poll. The Harris Interactive poll would be included in the BCS rankings in the same fashion as AP poll. The Harris Interactive poll, like the AP poll before it, is composed of voting panelists expressing their beliefs about the relative strengths of different football teams. Although we have no evidence to date, because of the similarities in construction between the AP poll and the Harris Interactive poll, we believe that the Harris Interactive poll will exhibit the same television exposure bias we observed in the AP poll. Thus, we believe that the position of teams from the midmajors has not changed: even with similar performance, midmajor teams will face greater difficulty in obtaining invitations to BCS
bowl games.
REFERENCES
Borland, M. V., Goff, B. L., & Pulsinelli, R. W. (1992). College athletics: Financial burden or boon? In G. Scully (Ed.), Advances in the economics of sport (pp. 215-235). Greenwich, CT: JAI.
Bowl championship series. (n.d.). Retrieved October 4, 2004, from http://www.bcsfootball.org
Gildea,W. (1998, January 5). Ranking the teams is not as easy as π. The Washington Post, pp. D1, D4.
Goff, B. (1996). An assessment of path dependence in collective decisions: Evidence from football polls. Applied Economics, 28, 291-297.
Lebovic, J., & Sigelman, L. (2001). The forecasting accuracy and determinants of football rankings. International Journal of Forecasting, 17, 105-120.
McCormick, R., & Tinsley, M. (1988). Athletics versus academics? Evidence from SAT scores. Journal of Political Economy, 95, 1103-1116.
Noel D. Campbell is an associate professor of business administration (economics) at The University of Central Arkansas. He earned a PhD in economics from George Mason University. His research interests include lotteries, merit-based financial aid, and state-level public finance. With R. Zachary Finney, he recently published “Mitigating the Combined Distributional Consequences of the Georgia Lottery for Education and the HOPE Scholarship” in Social Sciences Quarterly.
Tammy M. Rogers is an assistant professor of business administration (finance) at North Georgia College & State University. She earned a PhD in finance from Texas Tech University. Her research interests include the regulation of banks and other financial institutions, financial markets, and corporate finance. With Melissa Williams, she recently published “Financial Institution Stock Reaction to the Suspension of the Thirty-Year Treasury Bond” in International Journal of Business Disciplines.
R. Zachary Finney is an assistant professor of marketing at the University of South Alabama. He earned a PhD in marketing from the University of Alabama. His research interests include strategic management, strategic marketing, and topics in the resource-based theory of the firm. With Noel D. Campbell, he recently published “Strategies and Resources: Pathways to Success” in Journal of Business Research.