View Single Post
Old
02-24-2013, 11:42 AM
  #484
Miller Time
Registered User
 
Miller Time's Avatar
 
Join Date: Sep 2004
Posts: 9,279
vCash: 500
Quote:
Originally Posted by Talks to Goalposts View Post
My good fellow, if you are looking at how to Vancouver Canucks run things for examples for a coaches knack over analysis I believe you may be quite mistaken in terms of what you saw. The Nucks organization is famous for leaning on analysis to come up with better ways of doing things and one of their most easily observable conclusions is to run out the Sedins whenever they see an offensive opportunity to exploit. What you witnessed likely could be the fruit of the methodologies I embrace and you deride.
i was more pointing out a specific situation, rather than an organizational direction.

the bolded part is the distinction I think that gets lost in the focus on analytics.

I don't think anyone needs a complex statistical analysis to figure out that you want to run your best offensive players whenever you see an offensive opportunity to exploit...

but even if a team did need advanced stats to come to that conclusion, deciding WHEN those situations to exploit occur, and making the judgement of sending out those players even if it means double shifting, are both decisions that are best made by the observer, not by following some predictive model based on previous data collection (though I'm sure you could find some stat junkie that could show numbers pointing to specific minutes of a given period where the specific opposing team/players are at their historically most likely to give up a goal... the numbers could be calculated, but I'd still prefer the expert/knowledgable coach go with his gut, then check out a stat sheet to make those decisions.)


Quote:
Originally Posted by Talks to Goalposts View Post
Now for Diaz, he may be "in sync" in the defensive zone with Price. But there is a key problem with that in such things rarely lasting. Plus, as those discussing why he should not be pilloried for the New York game have so well pointed out, he has no control over the actions of his line mates. In that case is it not more prudent to rely on a player that can better control the balance of events and specifically, reduce the amount of offensive zone time the opposition has to work with? Mistakes by a 5 man unit in the defensive zone are inevitable in the long run in a parity league to favour the player with the proven track record of limiting the times in which your squad is most vulnerable to the opposition's talents and your own side's lapses?
they may be "rarely" lasting, but it is precisely the players/coaches/managers who are best able to take advantage of those "rare" exceptions, and succeed because of it.

being able to read the human beings in front of them is what best translates to success. That's not to say stats aren't useful, and any team would be stupid not to take advantage of the leaps and bounds being made in that area to help direct their decision making... but ultimately the real time/ever changing/unstable realities of human interaction are still best computed & interpreted in the moment by the expert brain.

The pats (as I understand it) are, like the nucks a team that relies heavily on advanced stats and playing the odds for most of their decision making... yet back in the early 2000's, when they decided to roll with a young, unproven kid, instead of re-inserting the established, successful, veteran... many people questioned them (both on the "code" that injury shouldn't cost a vet starter his job, and on the numbers which pointed heavily towards Blesoe as their best bet for success...)
a decade later Brady is still proving that his coaches gut instinct was correct.

Miller Time is offline