Marley,+C

 From this I hope to take away the affects of a crowd cheering on a team and how large a factor. I would also like to investigate the difficulties small market teams have. I have always wondered about the debates that go on radio talk shows about the difficulties small market teams have and would like to see if it is justified. Also in 2003 the NHL instituted a Salary Cap and I would like to find out if it is effective. I love all sports and would like to investigate if a persons playing level may differ from city to city. Population is choose because it effects the size of crowd as motivation as well as the finances of a team which should create a stronger correlation.
 * Brainstorm Ideas For Topic: **
 * Question: ** Does the Population of a City affect the Success of a Sports team in the City?
 * Question: ** Does the Population of a City affect the Success of a Sports team in the City?

Click Here For a Funny Sports Video


 * Hypothosis: ** I believe that the greater the population of a city the greater the success of the sports team in that city. I think this because generally the greater the population of a city the greater the funds to purchase better athletes. Furthermore the larger the population the more fans cheering on a team and the better the athletes will play. The officiating may also be affected by the large crowd cheering for one team. I also believe that a salary cap will not fully remove the advantage of having a large population because I believe that although the finances will be equalized between all teams, the fan support a media coverage will still compel the team to win more.
 * Variables: **
 * Independent:** The populations of people living in each city where the sports teams are located

This is the number of people living in a city where a major sports team from the NFL, NHL or, EPL(English Premier League, Europe). Each team will have a different populations but the population will affect the Dependent Variable. The population measurement will be taken from the cities bounderies, even though the fans and supporters will stretch beyond the cities bounderies, data could not be collected to account for the number of fans so to keep a consistent independent variable, the population will be the city not the metropolitan area.


 * Dependent:** The success of a sports team while playing at home venue

This success value will be a point system divided by the number of home games thus giving us points per game ratio. This is because some sports teams play many more home games then other sports teams so we have to make sure that the value is even for all teams. This ensures equality for a team with 41 games at home may get 50 points and a team with 9 games at home may get 20 points and the graph would show that the team with 41 games has more points but the team with 9 games actually did better at home. Lastly the points system will be 3 points for a win, 1 points for a tie, 2 points for an overtime win and this must be done because each sport has a different point system and this is the most fair point system and ensure every team will have the same point system. Winning percentage is not used because sports like hockey have many ties where as football as next to no ties which would skew the results

**TIme Frame:** 2005-Present. Any time past 2005 the populations may be different enough to skew the correlation. Populations are done from the most recent census from their respective countries


 * Region: ** The main target region is North America from the most popular Canadian sport of Hockey and the most popular American Sport of Football. To compare this data to world wide data the most popular European sport of soccer will be investigated as well

Raw data has been extracted to establish a correlation between the population of a city and the success at home of a sports team.


 * The Win - Loss Records of various sports teams
 * The population census


 * Sampling Technique: ** The population data was found through a census from each country and uses survey sampling. Every household was ordered to fill out a survey indicating the number of members living in the household. The win-loss records are the officail records and were found based on if a team wins or loses. This is not a sample


 * Bias: ﻿ ** There may be a response bias from participants not filling out surveys to indicate the number of people living in an area but the government will probably ensure the population of a city is very close to the actual population. There is a measurement bias due to the fact that population is found with people inside the cities boundaries when people living just outside the boundary still cheer for the sports team of the city. For example people in Hamilton will probably cheer on the Toronto Maple Leafs even though they do not live in Toronto. To fix this, one could use the metropolitan area of a city for population, but this data could not be located for each city and may not reflect the the fan support base because there is no way of knowing if the metropolitan area cheers for one team over another. To ensure equality, the population of each city was used not the metropolitan area. Populations were not divided by 2 if two of the same sport teams played within the same city such as the New York Rangers and the New York Islanders because there is no way to judge how the population is split.

NHL INFORMATION - Complete Table with Graphs from each year An example of a table is: This is the 2009 data table. For a complete list of all the data tables go to the complete information by clicking Home Success vs. Population An example of a graph for the NHL in 2009 is:
 * Results: **
 * HOCKEY**- There was minimum correlation between size of cityand success at home in the sport of hockey.
 *  ** NHL Hockey 2009 ** ||
 * Team || Wins || Ties || Overtime Wins || Total Points || Points per Game |||| Population ||
 * Washington || 30 || 6 || 0 || 96 || 2.341 || 599657 ||  ||
 * San Jose || 27 || 8 || 0 || 89 || 2.171 || 964,695 ||  ||
 * Chicago || 29 || 4 || 0 || 91 || 2.220 || 2,851,268 ||  ||
 * Phoenix || 29 || 2 || 0 || 89 || 2.171 || 1,593,659 ||  ||
 * Vancouver || 30 || 3 || 0 || 93 || 2.268 || 578,041 ||  ||
 * New Jersey || 27 || 4 || 0 || 85 || 2.073 || 278,154 ||  ||
 * Detroit || 25 || 6 || 0 || 81 || 1.976 || 910,921 ||  ||
 * Pittsburgh || 25 || 4 || 0 || 79 || 1.927 || 311,647 ||  ||
 * Los Angeles || 22 || 6 || 0 || 72 || 1.756 || 3,831,868 ||  ||
 * Nashville || 24 || 3 || 0 || 75 || 1.829 || 605,473 ||  ||
 * Buffalo || 25 || 6 || 0 || 81 || 1.976 || 270,240 ||  ||
 * Colorado || 24 || 3 || 0 || 75 || 1.829 || 610,345 ||  ||
 * Ottawa || 26 || 4 || 0 || 82 || 2.000 || 812,129 ||  ||
 * Boston || 18 || 6 || 0 || 60 || 1.463 || 645,169 ||  ||
 * St Louis || 18 || 5 || 0 || 59 || 1.439 || 356,587 ||  ||
 * Calgary || 20 || 4 || 0 || 64 || 1.561 || 988,193 ||  ||
 * Anaheim || 25 || 5 || 0 || 80 || 1.951 || 337,896 ||  ||
 * Philadelphia || 24 || 3 || 0 || 75 || 1.829 || 1,547,297 ||  ||
 * Montreal || 20 || 5 || 0 || 65 || 1.585 || 1,620,693 ||  ||
 * Dallas || 23 || 7 || 0 || 76 || 1.854 || 1,299,542 ||  ||
 * NY Rangers || 18 || 6 || 0 || 60 || 1.463 || 8,391,881 ||  ||
 * Minnesota || 25 || 4 || 0 || 79 || 1.927 || 385,378 ||  ||
 * Atlanta || 19 || 6 || 0 || 63 || 1.537 || 540,922 ||  ||
 * Carolina || 21 || 3 || 0 || 66 || 1.610 || 709,441 ||  ||
 * Tampa Bay || 21 || 6 || 0 || 69 || 1.683 || 343,890 ||  ||
 * NY Islanders || 23 || 4 || 0 || 73 || 1.780 || 8,391,881 ||  ||
 * Columbus || 20 || 9 || 0 || 69 || 1.683 || 190,414 ||  ||
 * Florida || 16 || 9 || 0 || 57 || 1.390 || 89,787 ||  ||
 * Toronto || 18 || 6 || 0 || 60 || 1.463 || 2,503,281 ||  ||
 * Edmonton || 18 || 4 || 0 || 58 || 1.415 || 730,372 ||  ||

The complete Graph can be seen here. There is very little correlation between points per game at home and population of a city. __Mean__ - 1.78 Points per Game which demonstrates that teams play better at home then on the road because the average amount of points per game will be 1.5 from 3(total points per game) / 2(teams each game). Therefor a team will earn about 0.56 points per game more at home then on the road __Standard Deviation__ - 0.2987 which means that the points per game deviates a little but it is fairley consistent meaning that there isn't a team that wins every game or loses every game __Correlation Coefficient__ - -0.1377 which is not very correlated a goes against my hypothesis. Weak Negative
 * Stats **

There is no data here to suggest that the larger the population the more succes a hockey team has at home.

Fan Base- City may have another sports team that is more popular and will focus on a different sport Salary Cap - Ensures that each team has the same amount of money to spend and allows small market teams to spend the same as large market teams. Takes out the extra money that can be spent by hockey teams with large populations within the city Players/Coaching - Good Coaching can help a team win as well as above average players. Although a high population city may bring in better players smaller cities could have stronger players on their team
 * Hidden Variables**

Soccer Hockey is played just in North America and in order to ensure fairness in the study Europe must be tested as well.



__Mean__ - 1.80 Points per Game still demonstrating the strength of a home team. A Premier League team earns 0.6 points per game more at home then on the road __Standard Deviation__- 0.4963. This is a high standard deviation meaning that there is a wide range and variance in the success of teams. Some teams such as Arsenal and Manchester United do extremly well. __Correlation Coefficient__ - 0.4093. Still not a very strong correlation that agrees with the hypothesis but does not proove it. Weak Positive
 * Stats **

Ladder System - The Premier League is a ladder system meaning the bottom three teams each year drop down to a lower league and the top 3 teams from the lower league move up. So there will be some very weak teams each year as well as they will not have the same fan base or finances even with a large population Fan Base - Most of the club teams are located within the same city such as London has 4 teams. So although the population is high, the fans and finances are split. Also a city may have another sports team that is very popular within Players/Coaching - Good Coaching can help a team win as well as above average players. Although a high population city may bring in better players smaller cities could have stronger players on their team
 * Hidden Variables: **

Football is the most popular team sport and will therefore be the best sport to test because of its popularity. It is normally the number one sport in each teames respective city Stats Mean - 1.73 Points Per Game. A NFL Team achieves about 46 points per game more at home then on the road Standard Deviation - 0.6901 which is an extremly high standard deviation. This is most likely due to the fact the NFL teams either win or lose and rarely tie. Also with such few games, the top teams rarely lose and the bottom teams rarely win. It is different then in most sports where the worst place team in football will not beat the first place team. Correlation Coefficient - 0.1092. Not a strong correlation which does not prove my hypothesis. Weak Positive

Hidden Variables Salary Cap - Each team can spend the same money so no advantage to high population teams

Players/Coaching - Good Coaching can help a team win as well as above average players. Although a high population city may bring in better players smaller cities could have stronger players on their team

Non - Salary Cap Hockey __Mean__ - 1.579 Points Per Game meaning that home ice was less important before the lockout __Standard Deviation__ - 0.351 indicating some variance between the teams but consistent with after the lockout and in the salary cap era __Correlation Constant__ - -0.1306 which goes against my hypothesis. Weak Positive
 * Stats **

A Salary Cap does not affect the succes at home indicating it may infact be useless. Still does not support my hypothesis and does not correlate population and success of team at home


 * Conclusion: **One can clearly see that there is little to no correlation between Population and success of a team at home. None of the graphs had a correlation coefficient that demonstrated a strong positive correlation indicating my hypothesis could not be varrified. Population incorperates both size of crowd and finances so from this one could assume that those have little effect on success but a full study would be needed to indicate this. The strongest correlation is in Premier League Soccer but the correlation is weak which does not proove my hypothesis. My hypothesis is therfore invalid and not supported by the data.


 * Bibliography: **