Comparing the Performance of US College Football Teams in the Web and on the Field

67 %
33 %
Information about Comparing the Performance of US College Football Teams in the Web and...

Published on July 14, 2009

Author: martinklein0815

Source: slideshare.net

Comparing the Performance of US College Football Teams in the Web and on the Field Martin Klein Olena Hunsicker Michael L. Nelson mklein@cs.odu.edu koval_olena@yahoo.com mln@cs.odu.edu Old Dominion University Hypertext 2009 Torino, Italy 06/30/2009

Naming Conventions 2

Naming Conventions Football 2

Naming Conventions Football Soccer 2

Motivation 3

Motivation • “Does Authority mean Quality?”[Amento00] 3

Motivation • “Does Authority mean Quality?”[Amento00] • Link-based web page metrics can be used to estimate experts’ assessment of quality 3

Motivation • “Does Authority mean Quality?”[Amento00] • Link-based web page metrics can be used to estimate experts’ assessment of quality • Lists compiled by experts are cool! 3

Motivation • “Does Authority mean Quality?”[Amento00] • Link-based web page metrics can be used to estimate experts’ assessment of quality • Lists compiled by experts are cool! • Companies, schools, people, places, etc 3

Motivation • “Does Authority mean Quality?”[Amento00] • Link-based web page metrics can be used to estimate experts’ assessment of quality • Lists compiled by experts are cool! • Companies, schools, people, places, etc • “Big 3” search engines play a central role in our lives 3

Motivation • “Does Authority mean Quality?”[Amento00] • Link-based web page metrics can be used to estimate experts’ assessment of quality • Lists compiled by experts are cool! • Companies, schools, people, places, etc • “Big 3” search engines play a central role in our lives • “If I can’t find it in the top 10 it doesn’t exist in the web” 3

Motivation • “Does Authority mean Quality?”[Amento00] • Link-based web page metrics can be used to estimate experts’ assessment of quality • Lists compiled by experts are cool! • Companies, schools, people, places, etc • “Big 3” search engines play a central role in our lives • “If I can’t find it in the top 10 it doesn’t exist in the web” • SEOs 3

Motivation • “Does Authority mean Quality?”[Amento00] • Link-based web page metrics can be used to estimate experts’ assessment of quality • Lists compiled by experts are cool! • Companies, schools, people, places, etc • “Big 3” search engines play a central role in our lives • “If I can’t find it in the top 10 it doesn’t exist in the web” • SEOs Do expert rankings of real-world entities correlate with search engine ranking of corresponding web resources? 3

Background • Expert ranking of real-world entities: • Collegiate football programs in the US • Associated Press (AP) poll • 65 sportswriters and broadcasters • USA Today Coaches poll • 63 college football head coaches • Published once a week, top 25 teams, 25-1 point system • “Big 3” search engines • Google, Yahoo and MSN Live (APIs) 4

US College Football Season 2008 • 2008 season began on August 28th 2008 • Concluded January 8th 2009 • 18 instances of poll data: • Final polls from 2007 season (as a baseline) • 2008 pre-season polls • once for each of the 16 weeks of the 2008 season 5

US College Football Season 2008 • 2008 season began on August 28th 2008 • Concluded January 8th 2009 • 18 instances of poll data: • Final polls from 2007 season (as a baseline) • 2008 pre-season polls • once for each of the 16 weeks of the 2008 season 5

Mapping Resources to URLs 6

Mapping Resources to URLs • Often impossible to distill the canonical URL for a football program 6

Mapping Resources to URLs • Often impossible to distill the canonical URL for a football program • e.g. Virginia Tech college football returned 6

Mapping Resources to URLs • Often impossible to distill the canonical URL for a football program • e.g. Virginia Tech college football returned 6

Mapping Resources to URLs • Often impossible to distill the canonical URL for a football program • e.g. Virginia Tech college football returned • Official school page 6

Mapping Resources to URLs • Often impossible to distill the canonical URL for a football program • e.g. Virginia Tech college football returned • Official school page • Commercial sports sites 6

Mapping Resources to URLs • Often impossible to distill the canonical URL for a football program • e.g. Virginia Tech college football returned • Official school page • Commercial sports sites • Wikipedia 6

Mapping Resources to URLs • Often impossible to distill the canonical URL for a football program • e.g. Virginia Tech college football returned • Official school page • Commercial sports sites • Wikipedia • Blogs, Fan sites, etc 6

Mapping Resources to URLs • Query 3 search engine APIs for representative URLs • Query: schoolname+College+Football • e.g.: Ohio+State+College+Football • Aggregate the top 8 representative URLs (n = 1 .. 8) • Temporal aspect in mind: • Repeat query and renew aggregation weekly 7

Ordinal Ranking of URLs from SE Queries 8

Ordinal Ranking of URLs from SE Queries We are not interested in computing search engine’s absolute ranking for a particular URL (PR values) rank(U RLA ) = 0.92 rank(U RLB ) = 0.73 rank(U RLC ) = 0.42 rank(U RLD ) = 0.13 8

Ordinal Ranking of URLs from SE Queries We are not interested in computing search engine’s absolute ranking for a particular URL (PR values) rank(U RLA ) = 0.92 rank(U RLB ) = 0.73 rank(U RLC ) = 0.42 rank(U RLD ) = 0.13 BUT We are determining that a search engine ranks URLs in order rank(U RLA ) ≥ rank(U RLB ) ≥ rank(U RLC ) ≥ rank(U RLD ) 8

Ordinal Ranking of URLs from SE Queries We are not interested in computing search engine’s absolute ranking for a particular URL (PR values) rank(U RLA ) = 0.92 rank(U RLB ) = 0.73 rank(U RLC ) = 0.42 rank(U RLD ) = 0.13 BUT We are determining that a search engine ranks URLs in order rank(U RLA ) ≥ rank(U RLB ) ≥ rank(U RLC ) ≥ rank(U RLD ) distance(U RLA , U RLB ) = distance(U RLB , U RLC ) 8

Ordinal Ranking of URLs from SE Queries 9

Ordinal Ranking of URLs from SE Queries • Search engines enforce query restrictions (length, amount per day etc) 9

Ordinal Ranking of URLs from SE Queries • Search engines enforce query restrictions (length, amount per day etc) • Build unbiased and overlapping queries 9

Ordinal Ranking of URLs from SE Queries • Search engines enforce query restrictions (length, amount per day etc) • Build unbiased and overlapping queries • site and OR operators 9

Ordinal Ranking of URLs from SE Queries • Search engines enforce query restrictions (length, amount per day etc) • Build unbiased and overlapping queries • site and OR operators • Variation of strand sort 9

Ordinal Ranking of URLs from SE Queries • Search engines enforce query restrictions (length, amount per day etc) • Build unbiased and overlapping queries • site and OR operators • Variation of strand sort USC Georgia Ohio State Oklahoma Florida site:http://usctrojans.cstv.com/sports/m-footbl/usc-m-footbl-body.html OR site:http://uga.rivals.com/ OR site:http://sportsillustrated.cnn.com/football/ncaa/teams/ohiost/ OR site:http://www.soonersports.com/ OR site:http://www.gatorzone.com/ 9

Weighting Ranked URLs • If real-world resources are mapped to more than one URL (n > 1) • Need to accumulate ranking score • Determine one final overall school score • Assign weights per URL depending on their rank P W eight = 1 − T P - Position of URL in result set T - Total number of URLs in the list (n * number of teams) 10

Correlation Results Kendall Tau used to test for statistically significant (p<0.05) correlation 1.0 1.0 0.9 ● 0.9 0.8 0.8 ● ● ● ● 0.7 ● ● ● 0.7 ● ● ● ● ● 0.6 ● 0.6 ● 0.5 ● ● 0.5 0.4 ● ● ● Kendall Tau Kendall Tau ● 0.4 ● ● 0.3 ● ● ● ● 0.3 ● 0.2 ● ● 0.2 0.1 0 0.1 ● −0.1 0 −0.2 −0.1 −0.3 −0.2 Yahoo Yahoo −0.4 −0.3 ● Google ● Google −0.5 MSN MSN −0.4 P>0.05 P>0.05 ● 2007 W1 W3 W5 W7 W9 W11 W13 W15 2007 W1 W3 W5 W7 W9 W11 W13 W15 Time Intervals Time Intervals Top 10 AP Poll Top 10 USA Poll 11

Correlation Results Kendall Tau used to test for statistically significant (p<0.05) correlation 1.0 1.0 0.9 0.9 0.8 0.8 ● 0.7 ● ● 0.7 ● 0.6 ● ● ● 0.6 0.5 ● Kendall Tau Kendall Tau ● ● 0.5 0.4 ● ● ● 0.4 0.3 ● ● ● ● 0.2 ● 0.3 ● ● ● ● ● ● ● ● ● ● 0.1 ● ● 0.2 ● 0 ● Yahoo Yahoo 0.1 ● Google −0.1 ● Google MSN MSN 0 P>0.05 −0.2 P>0.05 2007 W1 W3 W5 W7 W9 W11 W13 W15 2007 W1 W3 W5 W7 W9 W11 W13 W15 Time Intervals Time Intervals Top 25 AP Poll Top 25 USA Poll 12

Correlation Results Kendall Tau used to test for statistically significant (p<0.05) correlation 1.0 1.0 0.9 0.9 0.8 0.8 ● 0.7 ● ● 0.7 ● 0.6 ● ● ● 0.6 0.5 ● Kendall Tau Kendall Tau ● ● 0.5 0.4 ● ● ● 0.4 0.3 ● ● ● ● 0.2 ● 0.3 ● ● ● ● ● ● ● ● ● ● 0.1 ● ● 0.2 ● 0 ● Yahoo Yahoo 0.1 ● Google −0.1 ● Google MSN MSN 0 P>0.05 −0.2 P>0.05 2007 W1 W3 W5 W7 W9 W11 W13 W15 2007 W1 W3 W5 W7 W9 W11 W13 W15 “Inertia” 25 USA Poll Time Intervals Time Intervals Top 25 AP Poll Top 12

n-Values for Correlation Yahoo Yahoo ● Google ● Google MSN MSN P>0.05 P>0.05 ● ● 8 8 7 7 ● ● ● ● ● 6 6 n n ● ● ● ● ● 5 5 ● ● ● ● ● 4 4 ● ● ● ● ● ● ● 3 3 ● ● ● ● ● 2 ● 2 ● ● 1 1 2007 W1 W3 W5 W7 W9 W11 W13 W15 2007 W1 W3 W5 W7 W9 W11 W13 W15 Time Intervals Time Intervals Top 10 AP Poll Top 10 USA Poll 13

n-Values for Correlation Yahoo Yahoo ● Google ● Google MSN MSN P>0.05 P>0.05 ● ● 8 8 ● ● ● ● ● 7 7 ● ● ● ● 6 6 n n ● ● 5 5 ● ● ● 4 4 ● ● ● ● ● ● ● ● ● ● 3 3 ● ● ● ● 2 ● 2 ● 1 1 2007 W1 W3 W5 W7 W9 W11 W13 W15 2007 W1 W3 W5 W7 W9 W11 W13 W15 Time Intervals Time Intervals Top 25 AP Poll Top 25 USA Poll 14

n-Values for Correlation Yahoo Yahoo ● Google ● Google MSN MSN P>0.05 P>0.05 ● ● 8 8 ● ● ● ● ● 7 7 ● ● ● ● 6 6 n n ● ● 5 5 ● ● ● 4 4 ● ● ● ● ● ● ● ● ● ● 3 3 ● ● ● ● 2 ● 2 ● 1 1 2007 W1 W3 W5 W7 W9 W11 W13 W15 2007 W1 W3 W5 W7 W9 W11 W13 W15 Time Intervals Time Intervals Top 25 AP Poll n=2..6 Top 25 USA Poll 14

Correlation of Overlapping URLs Over Time • 12 schools occur in all AP polls throughout the season • Given the “inertia”, by how much does the web trail? • Can we measure a “delayed correlation”? • Declare AP ranking for each week as separate “truth values” • Compute correlation between truth values and search engine ranking • Expect to see in increased correlation in the weeks following the truth value USC Georgia Ohio State Oklahoma Florida Missouri Texas Texas Tech Alabama BYU Penn State Utah 15

Correlation of Overlapping URLs Over Time 1.0 1.0 1.0 1.0 1.0 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0.5 0.5 0.5 0.5 0.5 ● ● ● ● ● ● ● ● ● correlation correlation correlation correlation correlation ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0.0 0.0 0.0 0.0 0.0 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● −0.5 −0.5 −0.5 −0.5 −0.5 1 3 5 7 9 11 13 15 1 3 5 7 9 11 13 15 1 3 5 7 9 11 13 15 1 3 5 7 9 11 13 15 1 3 5 7 9 11 13 15 delta in weeks delta in weeks delta in weeks delta in weeks delta in weeks 1.0 1.0 1.0 1.0 1.0 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0.5 0.5 0.5 0.5 0.5 ● ● ● ● ● ● ● ● correlation correlation correlation correlation correlation ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●

Add a comment

Related pages

NCAA College Football Teams - ESPN - ESPN: The Worldwide ...

College Football Teams. Conferences ... College Football; College Basketball; Soccer; NASCAR; ... Contact Us; Member Services;
Read more

NFL - Head to Head Stats - National Football League

© 2015 NFL Enterprises LLC. NFL and the NFL shield design are registered trademarks of the National Football League.The team names, logos and uniform ...
Read more

Chart: Which NCAA Football Teams Outplay Their Recruit ...

But given there's more than 120 teams in FBS college football, ... When college football teams fail to ... better on the field while teams in the red ...
Read more

NCAA College Football Teams, Scores, Stats, News ...

Get the latest NCAA college football news ... and are now one of just two division teams unbeaten in ... he's a handful on the football field.
Read more

NFL Teams - National Football League

Contact Us; FAQ | Rule Book; Public Relations; Jobs ... NFL and the NFL shield design are registered trademarks of the National Football League.The team ...
Read more

College athletics in the United States - Wikipedia, the ...

... college football bowl eligibility ... The scale of college sports in the US is measured by ... Colleges would still be able to field "club teams ...
Read more

NCAA College Football Teams - CBSSports.com

College football teams listed by conference at CBSSports.com. College ... Driving Performance; GMC Perfect ... Nike College Championship Drive Gear Get ...
Read more

Compare Colleges & Universities - Best College Comparison ...

Compare schools by rankings, tuition, ... Compare Colleges & Universities. ... and a list of all the available fields of study..
Read more

The 7 Major Differences Between NFL and College Football

College football has it’s ... it gives the losing college team more time ... the college games are more open than the NFL games. The NFL’s field of ...
Read more

BBC Sport - Football - Teams

Find out the latest news stories, results and fixtures, photo galleries, videos and audio, player statistics and team facts for your favourite football ...
Read more