One of the beautiful things about the so-called "blogger revolution" is that I think it has helped me become a much more educated fan of the game. Instead of being forced to read the same score-stat-quote template article a million times from mainstream media outlets, I can gain a lot of valuable information, and read things I wouldn't have necessarily known in a relatively short period of time.
The best example of this for me is the Big Ten Wonk. In five minutes, I can be up to date on everything of value that is going on in the Big Ten basketball world.
What does all this have to do with college hockey? Well, one of the things that I have found to be incredibly informative is Wonk's use of tempo-free stats. For the uninitiated, tempo-free stats use ratios and formulas to objectively compare player performance. The easiest examples of this would be free throw percentage, or yards per carry in football. They really give a much more accurate picture of what is going on by saying who is the most efficient rather than who has the best raw numbers.
So my challenge to you in the college hockey community, is how can we apply this to the sport of hockey? Is there any way to come up with a formula that more objectively compares players?
One way I was thinking was to compare a player's goals against the total goals in the league. That might help to compare performances of a player in the high-scoring BCHL as opposed to a more defensive league like the USHL. It probably wouldn't help a lot in college hockey though.
I realize it's difficult because college hockey stats are pretty limited. I'd love to hear whatever suggestions people had though, because I think we could come up with some interesting ways to analyze the game better.
Subscribe to:
Post Comments (Atom)
5 comments:
The conference issue seems to be the hardest thing to deal with. For most of the year I've seen three or four Mercyhurst guys at the top of the scoring sheet. But, they've only played five out of conference games and have only won a single game. How does that stack against the guys from UNO, CC, DU and UMN around him. One thought is to find out what the average player , with a minimum number of games played does across the country, and in each conference and try to boil it down to an objective number to work from.
I can't believe that I follow a link from The DECC is Stacked to this site and the first post I read is about stats. I love it!
I don't have anything valuable to contribute, but I will think about it...
For NCAA stats...regarding Mercyhurst (or whoever) having players at the top of the nation in scoring - how about somehow figuring strength-of-schedule into the equation?
It may have to be done in "groupings" of some sort, so as not to give a player on the team with the #1 SOS an advantage over a player on the team with the #2 SOS.
Not sure how you'd do it though (=.
After drinking copious amounts of beer, myself and a couple of friends realized, you're really going to need to break this down on a game by game basis, with the RPI or the SOS being very important
For example if a guy scores 15 goals on two teams with crappy schedules, but your team plays a really really strong schedule, that needs to be accounted for.
What if you weight each goal scored by how many goals the opposing team has allowed, or something along those lines?
Post a Comment