Thursday, February 4, 2016

Strength of Schedule for Carolina and Denver


Strength of Schedule for Carolina and Denver


The Carolina Panthers had one of the great regular season records of all time.   However, their regular season schedule appeared to be a bit weak compared to other teams including their Super Bowl Opponent the Denver Broncos.  Let’s test the hypothesis that the Panthers had a weak regular season schedule.

Question:  The table below has regular-season won/loss records for the opponent of the Carolina Panthers and the Denver Broncos.  

Is the total win percentage of the opposition to Carolina significantly different from the total win percentage of the opposition to the Denver Broncos?

Discuss the limitations of this test and other things a statistician might do to assess the strength of schedule for the two Super Bowl Teams.


Strength of Schedule
Carolina
Denver
Opposition
Wins
Losses
Wins
Losses
Jacksonville
5
11
Baltimore
5
11
Houston
9
7
Kansas City
10
5
New Orleans
7
9
Detroit
7
9
Tampa Bay
6
10
Minnesota
11
5
Seattle
10
6
Oakland
7
9
Philadelphia
7
9
Cleveland
3
13
Indianapolis
8
8
Green Bay
10
6
Green Bay
10
6
Indianapolis
8
8
Tennessee
3
13
Kansas City
10
5
Washington
9
7
Chicago
6
10
Dallas
4
12
New England
12
4
New Orleans
7
9
San Diego
4
12
Atlanta
8
8
Oakland
7
9
New York Giants
6
10
Pittsburgh
10
6
Atlanta
8
8
Cincinnati
12
4
Tampa Bay
6
10
San Diego
4
12
Total
113
143
Total
126
128


Analysis:   The aggregate win/loss records for the regular-season opponents of the two Super Bowl teams are presented in the table below.

Total Regular Season Wins and Losses for Opponents of Carolina and Denver
Carolina Opposition
Denver Opposition
Total
Wins
113
126
239
Losses
143
128
271
Win Proportion
0.4414
0.4961
0.4686



The difference in the won-loss record for the two teams appears non-trivial but is it significant?

The answer to this question is No!



t-test on  Difference in Regular Win Percentages for the Opponents
of the Two Super Bowl Teams
Diff
-0.0272
SE
0.022096746
t-stat
-1.232
Critical Value for .05 significance level
-2.3


Discussion:  If we added playoff games of the two teams to the sample we would increase sample size and decrease standard error by around 15%.   This would make the test results a bit closer but still not significant.

 I also very quickly looked at median wins for the opponents of the two teams.  Found it was 7.0 for Carolina and 7.5 for Denver.   This seems trivial.

Carolinas schedule does seem a bit weak but their record was impressive and you can only beat the teams that you play.

I do sense that Denver is the more tested team having played and beaten New England twice once in the regular season and the second time in the AFC championship game.   The counter argument is that Carolina beat Seattle twice but I Seattle had some hiccups this year.

I don’t believe Carolina has played against a defense as accomplished as Denver’s defense.   It might be useful to study defensive strength of the two teams opponents and attempt to determine whether Carolina can beat a truly great defense.

I don’t expect Denver to score much but I think Denver’s defense will also keep the score low.



Go Back to the home-field advantage hypothesis testing problems
http://www.dailymathproblem.com/p/home-field-hypothesis-testing-problems.html








No comments:

Post a Comment