Monday, 23 April 2012

NHL playoffs

I'm not a big fan of hockey, but since it's playoff season I'm watching a few games.

Given the recent loss of the first-place Canucks I checked some stats and noticed just how weakly regular season play is related to Stanley Cup victories.

Consider past President's Trophy winners (for the best regular league points record): Of the 26 winners, only seven won the Stanley Cup (27%). Were the Cup doled out randomly among playoff teams they'd win about 6% of the time. Then I did some quick number crunching. It turns out a conference-winning team survives the playoff's first round 76% of the time (since 1982, when the modern four-round system was implemented). Not the worst odds ever, but perhaps revealing how lower ranking teams win so often; one in four times a conference leader is upset by an eighth-place team. The odds of both first-placers making it through are about 57% (but the odds of neither making it through are 6%, which might happen this year for the first-ever time). I'd like to compare this trend to other sports' playoffs... later.

2 comments:

  1. Funny that you interpret it that way. This hockey writer notes that of all the teams, the PT winner has the best chance, and the higher up you finish, the more likely you are to advance. So I'm not sure why you take the same numbers and say the relationship is weak.

    http://blogs.thescore.com/nhl/2011/04/02/the-presidents-trophy-curse-2/

    ReplyDelete
    Replies
    1. Neat stuff. I think my definition of 'weak' is probably wrong (interesting that the 4th place has about equal merit as first: 4 cups vs 5). I was going off stats like this one:

      "there have been 56 series matching the No. 1 team in a conference against the No. 8 team, and No. 2 versus No. 7. Of those 56 series, 20 have ended in victory for the underdog".

      (http://www.nytimes.com/2009/04/12/sports/hockey/12slapshot.html?_r=2&ref=todayspaper)

      Compare to tennis, where first round upsets are far less common: http://ca.sports.yahoo.com/tennis/news?slug=ycn-8532289

      Unfair comparison? Probably. Apples and oranges. No curse obviously. But if the 'best overall' team is crowned victor 1 in 4 times, could the correlation not be stronger?

      Delete