Tuesday, April 15, 2014

NHL Pay and Performance

Yesterday I blogged about competitive balance in the NHL, and today I want to extend that to take a look at pay and performance in the NHL during the regular season.  Recently a new website has emerged which focuses on the payroll cap in the NHL, called capgeek.com.  It is excellent and I know of no better website for salary and team payroll information on the internet.  So, I am going to use their information to estimate the relationship between regular season payroll and regular season performance from the 2009/10 NHL regular season to the 2013/14 regular season.

To do that, I will run a linear regression on relative payroll and team points (team performance measure).  For those interested, here is a step-by-step guide as to how to do this yourself.  You may be wondering why I use relative payroll and why I am focusing on the statistical measure called r-squared.  I answer (or link) to those directly below.

Why relative payroll?  This is for statistical purposes, but simply stated - the average of team payrolls are rising over the time period and the average of team performance is remaining relatively constant.  So if you just take a look at payroll and performance you are comparing a variable that is increasing to one that is constant (or stationary) and you will get a lower statistical relationship that what is truly taking place.  In fact the correlation coefficient (called r) between payroll and performance falls dramatically over this time period as opposed to the coefficient of determination (r-squared).  For this time period, for payroll and performance r = 0.034 while for relative payroll and performance r = 0.235, and thus for payroll and performance r-squared = 0.001 and for relative payroll and performance r-squared = 0.055.

Why do I use r-squared?  I have answered that here.

So, what is the relationship between team relative payroll and team regular season performance?  For the the 2009/10 NHL season to the 2013/14 NHL regular season that I have data from capgeek, relative payroll is statistically significant (at the 99% confidence level) with respect to team points. Yet, in terms of the common variation between the two variables, there is not much with the R-squared = 0.055.  If I include payments of long-term injured reserve and also include bonuses, the r-squared actually decreases.  Thus relative payroll "explains" less than 6% of NHL team regular season performance since 2009/10.

Hence, I conclude that NHL regular season team payrolls are not a good indicator of NHL regular season performance.

Monday, April 14, 2014

NHL Competitive Balance

With the 2013-14 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 probabilty 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.  For transparency, I will also report for each season this probability estimate.

OK, with the measurement details noted, here are the Noll-Scully competitive balance numbers for the last NHL season.  The first table used the binomial measure and the second table uses the trinomial measure.    The first column of numbers uses a sample standard deviation and the second column uses the population standard deviation. 


Binomial Distribution Sample Population
Standard Deviation of Winning Percent = 0.093023 0.091459
Average of Winning Percent = 0.5624 0.5624
Square root of Games Played = 9.055385 9.055385



Noll-Scully Measure of Comp. Balance = 1.497792 1.472617









Trinomial Distribution Sample Population
Richardson EEJ 2000

Idealized Standard Deviation= 0.047192 0.047192
Probability of a Tie = 0.269512 0.269512
Number of Games Played = 82 82
Noll-Scully Measure of Comp. Balance = 1.971147 1.938017

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.

Wednesday, April 2, 2014

World Cup and Economic Impact

Moody's forecasts that the World Cup will not have a large economic impact.  While there will be lots of tourists generating economic activity, there will also be a decline in economic activity that normally takes place to offset the gains.  Sports economists have found this same result numerous times here in the US.

Monday, March 17, 2014

Anti-trust Lawsuit Against NCAA

Sports labor lawyer files class action lawsuit against NCAA.

Army Changes their Head Football Coach

Army has fired head football coach Rich Ellerson, and hired Jeff Monken who was the head football coach at Georgia Southern.  Given head coach Ellerson's departure, here is an analysis of the Army Black Knights using the Complex Invasion College Football Production Model.  First here is a snapshot of the Black Knights over the last five years during head football coach Ellerson's tenure.  As you can see, the Black Knights were a team on the rise at the beginning and have been on a downward trend in terms of on-field production over the last few years.


2013
At the time of Ellerson's departure, not all the post-season bowl games were completed, so I will take a look at the Black Knights as of the end of the regular season (which is after the last game of the regular season Army-Navy).  Pre-game analysis was done using the Complex Invasion College Football Production Model.  Turning to the team, the Black Knights finished the regular season at 3-9 playing against an "easier" strength of schedule (SOS) as compare to the "league".  Army's best game was a victory over #90 ranked Louisiana Tech and their worst game was a defeat to #112 ranked Hawaii.  Army finished as the #102 ranked team out of 125 in terms of total production, with the #110 ranked offense and the #74 ranked defense.  As you can see from the chart above, Army has faced a downward trend in terms of team production rank from 2010 and continued losses to Navy, made changing head coaches a way of trying to turn around the program.

2012
The Black Knights finished the regular season at 2-10 playing against an "average" SOS as compared to the "league".   Army's best game was a victory over #81 ranked Air Force (other was a win over #100 Boston College) and their worst defeat was to FCS Stony Brook.  Army finished as the #107 ranked team with the #93 ranked offense and the #105 ranked defense.  It looks as if I did not write up an analysis of the Army-Navy game this year, but the trend of Army losing to Navy continued.

2011
Army finished the regular season at 3-9 and out of the bowl picture.  Army played against an "average" strength of schedule and Army's best game was a win over #65 Northwestern and Army's worst game was a loss to #105 Ball State.   In terms of overall production Army finished as the #76 ranked team with the #67 ranked offense and the #73 ranked defense.  Army lost their game to Navy.  Here is an analysis of the Army-Navy game.

2010
Ellerson's second year resulted in the Black Knights finishing the regular season at 6-6 and becoming bowl eligible.  Army defeated #40 SMU in the Armed Forces Bowl to finish overall at 7-6.  The Black Knights played against an "average" SOS as compared to the "league" overall.  Army's best game was their defeat of #40 SMU and their worst game was a loss to #93 ranked Rutgers.  Overall Army was the #57 ranked team in total production with the #68 ranked offense and the #38 ranked defense.  Army lost to Navy, and here is an analysis of this year's Army-Navy game.

2009
In Ellerson's first year at the helm of the Black Knights, Army finished the regular season at 5-7 and out of post-season bowl contention.  The Black Knights played against an "easier" SOS as compared to the league as a whole (meaning that their SOS was between one and two standard deviations above the "league" average SOS).  Another way of thinking about this is that Army played eight of their twelve games against teams in the bottom 25% of the league (including one of those games against FCS VMI and they went 5-3 against the bottom 25%.  The Black Knights best game was against #88 ranked Vanderbilt and their worst was a one point loss to #116 Tulane.  Overall the Black Knights were the #106 ranked team in total production with the #118 ranked offense (out of 120).  On the bright side Army had the #26 ranked defense.  Given their terrible offense, this is even more impressive.  Probably of most interest to Army fan's, Army lost to Navy.

Analysis of 2013 NCAA FBS Head Coach Changes
Vanderbilt and James Franklin
Penn State and Bill O'Brien
UMass and Charley Molnar
Boise State and Chris Petersen
Texas and Mack Brown
Washington and Steve Sarkisian
Wake Forest and Jim Grobe
Wyoming and Dave Christensen
Eastern Michigan and Ron English
Florida Atlantic and Carl Pelini
Miami of Ohio and Don Treadwell
UConn and Paul Pasqualoni
USC and Lane Kiffen