Wednesday, December 02, 2009
Official does not mean correct, when it comes to scorers
I love the effort put in here. When you see such wild inconsistency, with no second-level quality check, what does the data really mean? Increasing sample size does not reduce bias, if you are aggregating on the thing you suspect bias on.
To me, it’s extremely disappointing that the NHL does not care enough about its data recording that they don’t have a more robust system in place.
Glove-slap:Hawerchuk.


It’s a great article. Tango, this is the second article in row that you’ve really liked by this guy- you should learn his name (Sunny Mehta), so you can give him credit when the next opportunity rolls around.
The article supports my intuition that hockey is the opposite of baseball, in that watching the games will give you a better idea of how good teams (and, to a lesser degree, players) are relative to looking at the stats. eg, The casual baseball fan watches a pitcher throw nine innings of 2 hit ball with 0k and 4bb’s, and is misled into thinking the pitcher is highly skilled. Whereas, in hockey I think it’s easir to tell when a team outplayed another team, and not be misled by the goal totals in the game.