Today I am going to look at NBA payroll and NBA performance. Over the last few years, NBA payroll has not been a very good predictor of NBA regular season performance. Let's see if that trend holds up for the last two NBA regular seasons. Thus I have grabbed data from the internet on NBA payrolls for 2013/14 regular season and the 2014/15 regular season and NBA performance and run a regression on how well relative payroll is related to regular season performance.
What I find is that over the last two NBA regular seasons is that relative payroll is positive and statistically significant, but only explains about 12.6% of regular season winning percentage.
Monday, April 20, 2015
Saturday, April 18, 2015
2014-2015 NBA Competitive Balance
At the end of the regular season last year I blogged about competitive balance in the NBA since the 99/00 season. Now I will update this given the end of this year's regular season. For those interested here is a guide to calculating competitive balance using the Noll-Scully method on your own. The only difference here is that I will give both the sample and population measures of the Noll-Scully for this season.
After downloading the data from basketball reference, I found that the NBA is still highly uncompetitive as compared to other sports leagues, with a Noll-Scully of 2.97 (sample) and 2.92 (population). Either way, the NBA is still very uncompetitive relative to other sports leagues.
After downloading the data from basketball reference, I found that the NBA is still highly uncompetitive as compared to other sports leagues, with a Noll-Scully of 2.97 (sample) and 2.92 (population). Either way, the NBA is still very uncompetitive relative to other sports leagues.
Friday, April 17, 2015
2015 NBA Attendance Analysis
Now that the NBA regular season is in the books, let's take a look at how home attendance has changed (or not changed) over the last two seasons using the NBA attendance numbers from ESPN. So specifically I want to know if average regular season home attendance has changed from the 2013/14 season to the 2014/15 season statistically. I choose to use a t-test (assuming unequal variance) in NBA teams average home regular season attendance. What I find is exactly what I found previously, and that is that NBA average home regular season attendance has not changed in a statistical sense. In this case, the value of the t-test for NBA average home regular season attendance between the two seasons is 0.215, which is greater than generally accepted confidence level of 0.05. Hence, I conclude that NBA average home regular season attendance in the 2014/15 season is statistically no different than in the 2013/14 season.
Tuesday, April 14, 2015
2015 NHL Attendance Analysis
Now that the NHL regular season is over, let's take a look at NHL team
attendance and see how fan
attendance compares over the last two NHL seasons and see if anything is different from earlier analysis I have done for NHL average home regular season attendance.
First, I downloaded the NHL attendance data from ESPN for the last two NHL regular seasons, sorted the data and then calculated the t-test for the last two NHL regular seasons and found that there is no statistical difference between the home regular season attendance in the two seasons since the partial lockout.
For those curious, the t-test was 0.406 comparing the lockout season and the 2013/14 NHL season using a two sample equal (and unequal) variance measure and the t-test was 0.445 comparing the 2013/14 and 2014/15 NHL season using a two sample equal (and unequal) variance measure.
First, I downloaded the NHL attendance data from ESPN for the last two NHL regular seasons, sorted the data and then calculated the t-test for the last two NHL regular seasons and found that there is no statistical difference between the home regular season attendance in the two seasons since the partial lockout.
For those curious, the t-test was 0.406 comparing the lockout season and the 2013/14 NHL season using a two sample equal (and unequal) variance measure and the t-test was 0.445 comparing the 2013/14 and 2014/15 NHL season using a two sample equal (and unequal) variance measure.
Monday, April 13, 2015
2014-15 NHL Competitive Balance
With the 2014-15 NHL regular season in the books, let's take a look at how competitively balanced the season was (using the Noll-Scully measure of competitive balance) and compare this with recent seasons.
There are two ways of measuring competitive balance in hockey since unlike baseball or basketball, hockey games can end up tied at the end of regulation. So I will report both the binomial and the trinomial Noll-Scully measure. Additionally, there are two ways of reporting both the binomial and trinomial Noll-Scully measure: one using the standard deviation of a sample and the other using the standard deviation of the population. Again, I will report both.
Additionally, I will have to compute (for the trinomial distribution) the probability of a tie under equal playing strength. In the past, I used Richardson's estimate from Stanley Cup playoff games. In this case I will change and just assume that the probability of games that go into overtime occurs among teams with equal playing strength. Feel free to quibble with this, as this is simplification of the estimated probability.
OK, with the measurement details noted, here are the Noll-Scully competitive balance numbers for the recent NHL season. Using the sample the Noll-Scully under the binomial distribution was 1.5634 and under the trinomial distribution was 2.0286. For the population the Noll-Scully under the binomial distribution was 1.5372 and under the trinomial distribution was 1.9945.
Compared to recent years, this year was less competitive balanced, but overall the level of competitive balance in the NHL is still rather similar to recent historical numbers.
There are two ways of measuring competitive balance in hockey since unlike baseball or basketball, hockey games can end up tied at the end of regulation. So I will report both the binomial and the trinomial Noll-Scully measure. Additionally, there are two ways of reporting both the binomial and trinomial Noll-Scully measure: one using the standard deviation of a sample and the other using the standard deviation of the population. Again, I will report both.
Additionally, I will have to compute (for the trinomial distribution) the probability of a tie under equal playing strength. In the past, I used Richardson's estimate from Stanley Cup playoff games. In this case I will change and just assume that the probability of games that go into overtime occurs among teams with equal playing strength. Feel free to quibble with this, as this is simplification of the estimated probability.
OK, with the measurement details noted, here are the Noll-Scully competitive balance numbers for the recent NHL season. Using the sample the Noll-Scully under the binomial distribution was 1.5634 and under the trinomial distribution was 2.0286. For the population the Noll-Scully under the binomial distribution was 1.5372 and under the trinomial distribution was 1.9945.
Compared to recent years, this year was less competitive balanced, but overall the level of competitive balance in the NHL is still rather similar to recent historical numbers.
Subscribe to:
Posts (Atom)