Wednesday, December 24, 2008

More picking on Bedard

I'm feeling Grinchly, so why not pick on Bedard a little more?  Lately, it's like shooting fish in a barrel, but I have enough ammo and I like seafood so what the hell....  From today's Journal Sentinel.

According to the Elias Sports Bureau, only one other team - the 1984 Cleveland Browns - had more losses by four points or fewer than the current Packers. They had eight in a 5-11 campaign that saw coach Sam Rutigliano replaced by Marty Schottenheimer after a 1-7 start.

With one game to play, the Packers are tied with three other teams with seven losses by four points or fewer in a season: the 2001 Carolina Panthers, 1994 Houston Oilers and '93 New England Patriots. 

So what does Bedard think this means?

The Packers surely will learn from all these close games and put the knowledge to good use next season, won't they?

If history is any predictor, don't bet on it.

The four other teams the Packers are grouped with for close-game futility improved a collective 19 games the next season (13-51 to 32-32) but only one posted a winning record. The '93 Patriots were 5-11 in Bill Parcells' first season as coach and went 10-6 the next to make the playoffs as a wild-card entry.

Bedard concludes---from this overwhelming evidence----that the "Patriots are the exception rather than the rule."   The big problem here is that Bedard is working with a sample size of 4.   You cannot extrapolate "rules" and "exceptions" based on a sample size of 4.

But even if you could... Bedard's (very limited data) shows that each team improved the following season by a minimum of three games (and as many as six games).  Moreover, each of the teams that had at least five wins in the season in which it lost seven close games (the Browns and Patriots) went to the playoffs the following season.  (The Browns went to the playoffs at 8-8).

But enough of that silliness.  You cannot deduce anything based on a sample size of four.  But there is a great metric for predicting how the Packers will do next season: Pythagorean wins.  If you have never heard Pythagorean wins, Football Outsiders has a good primer here.  In one sentence, you can calculate how many games a team "should have" won based upon the teams points scored and points allowed.  Pythogorean wins are also (supposedly) the single best indicator for how a team will perform following season.  Teams that underachieve below their Pythagorean projection tend to bounce back the following season.  Teams that overachieve tend to slip the following season.

Anyhow, using this methodology, the Packers "should be" 8-7 right now,  and probably will finish the season with a Pythagorean number of 9.

Fun fact from Football Outsiders: "Right now, the Packers are on pace to underachieve their Pythagorean projection by .213, which would be more than any team since 1981."

Football Outsiders also has its own special metric called DVOA ("defense adjusted value over average").  "DVOA breaks down every single play of the NFL season to see how much success offensive players achieved in each specific situation compared to the league average in that situation, adjusted for the strength of the opponent."  Got it?

Anyhow, this will surprise you: the Packers have the 11th highest DVOA rating in the NFL.  Ahead of the Bears, Vikings, Patriots, and Cowboys, among many others.

In sum, notwithstanding Bedard's painstaking research based upon sample size of four, from a numbers and statistics standpoint, the Packers look like a team due for a bounce in 2009.

Merry Christmas!

Ho, ho, ho!

No comments:

Post a Comment