--- "Dr. Bob Hamilton" <[log in to unmask]> wrote:

. . . 

> Despite increasing public unease over violence in hockey, a statistical analysis of NHL data by
> university professors shows that on-ice fighting is a good strategy for team success.
> 

I guess I'm not being rude being the grammar/logic police here, since this is a quote from an
article not a Hockey-Ler, but that sentence is a non-sequitur.  If there is increasing unease over
violence in hockey, it has nothing to do with whether or not it's a good strategy for success.

> An intensive numbers-crunching of five years of statistics shows major penalties -- those most
> often issued for fighting -- increases the total points of the offending player's team and
> decreases the number of goals scored by their opponents.
> 
> Hockey goons must be careful, however. Only major penalties help win games; minor penalties
> lower a team's success, according to the analysis.
> 
> The team of professors at the Department of Economics and Business at Colorado College in
> Colorado Springs and the School of Business at the University of Sioux Falls analyzed data for
> all NHL teams from the 1999-2000 season through to 2003-2004.<<

The link didn't work for me; I'll try again later.  I'd be curious to know whether or not they
corrected for whether the penalties were matching majors or not.  If they didn't, I'd guess that
the sample size for non-matching majors is small, perhaps too small to be statistically
meaningful.  Also, if they didn't correct for matching majors, then I don't think that the
statistics are meaningful, because being shorthanded for five minutes is very, very, different
than having your goon and their goon (who typically aren't scorers anyway) sitting in the penalty
box instead of on the bench.  Also, for matching majors, then the effect on one team would exactly
counteract the effect on the other team; if a player from team A and a player from team B are off
with matching majors and team A scores a goal, then there is a goal against team B.  On the other
hand, it looks like they compared goals scored by each team, which is different than net goals,
which would balance out in matching penalties.

> >>The professors have measured the precise effect of a major penalty: For each penalty minute
> served, a team accrued 0.07 points and decreased their opponent's scoring by 0.24
> goals.

That second number looks fishy to me.  A five minute major decreases the opponent's scoring by
more than a goal?!?!



 
____________________________________________________________________________________
Have a burning question?  
Go to www.Answers.yahoo.com and get answers from real people who know.