View Single Post
Old
03-05-2013, 06:32 PM
  #135
Mathletic
Registered User
 
Mathletic's Avatar
 
Join Date: Feb 2002
Location: St-Augustin, Québec
Country: Canada
Posts: 12,419
vCash: 500
Quote:
Originally Posted by Jagorim Jarg View Post
These things may be measurable to an extent but not in a meaningful way, and even if the data were processed, no human has a mind that works fast enough to make an in-game decision based on the metrics that would be at play for the extremely rare data that has been procured based on that exact moment of puck control and relative positioning.
ok? But then I guess I don't understand your point. You're telling me that things like velocity, player distance and whatnot are intangibles. I tell you they are measurables. But then you tell me that they are meaningless. So why would a human mind account for velocity, player distance, etc. and then say it's meaningless?

As for in game decisions. Models aren't built on the actual game going on. You'd want to have your data ready and enter the new information in. Then your model would tell you the optimized solution. For example, say a team wants to match a certain line against your top line, then you could use that information to optimize matchups, time on ice and whatnot.

Same for baseball, say a manager makes a pitching change, do you keep your actual batter or do you replace him? It takes a couple clicks to figure out what the best solution is.

Mathletic is online now   Reply With Quote