SaskRinkRat
Registered User
- Apr 1, 2010
- 502
- 0
As someone who dabbles a bit in hockey analytics, but spends a lot of time dealing with data and analytics in a professional (non-hockey) setting, I feel like one of the biggest barriers to advancing the use of hockey analytics is the lack of standard definitions for the metrics we pay most attention to. Or at least a lack of access to those definitions.
Sometimes the lack of definition is in the metric calculation itself (i.e., the math that goes into coming up with a QoC percentage for a player), but often it's in the underlying event observed on the ice that eventually winds up being counted (or excluded) from a metric (i.e., what specific play counts as a scoring chance, or what counts as a "carry in" zone entry vs. a dump in).
Are any of you aware of any work going on to create a sort of meta database for analytics? Perhaps we could get a discussion going here about what definitions are available, the degree to which they are accepted and applied with consistency, and whether there are areas where standardization is still required?
Sometimes the lack of definition is in the metric calculation itself (i.e., the math that goes into coming up with a QoC percentage for a player), but often it's in the underlying event observed on the ice that eventually winds up being counted (or excluded) from a metric (i.e., what specific play counts as a scoring chance, or what counts as a "carry in" zone entry vs. a dump in).
Are any of you aware of any work going on to create a sort of meta database for analytics? Perhaps we could get a discussion going here about what definitions are available, the degree to which they are accepted and applied with consistency, and whether there are areas where standardization is still required?