AlanMacNeill
Super Rockchucker
Registered: Sep 2011
Location:
Posts: 1064 |
It's easy, the "analysts" are going to, once again, forget about the rule of small sample size and make massive changes to the system based on a single failure.
News Flash: just because you are the single best nation in the sport does NOT guarantee you golds, or even medals. The other nations try too, and even if you are enough better to be a 66% favorite, you're going to lose one out of three.
The biggest mistake to make is overreacting. Look at the US, after 2014, they ditched Shuster, threw him to the curb, declared him a failure, and made him persona non grata.
It took them a year to win their way back into the Program (and even then, only half-heartedly). the Analytics guys swore up and down McCormick was better, or Fenson, or any one of the other folks on the inside based on "the numbers"
Look who has a gold now?
Analytics is great, when matched with common sense and an understanding that just because putting your first stone in the house increases your win expectancy 0.03% doesn't mean you should do it every single end.
But, that's Analytics maturity Level 5 (seriously, this is my field, it's what I do for a living, Analytics is serious business for business). Curling is only barely at maturity level 1 (an agreement on what should be tracked).
The rest of the trip involves agreeing on how to track it (our stats system is *almost* there, but relies too much on human judgement), change in organizational behavior as a result of analytic results (some teams are looking at it), and then a focused revision and refinement of the analytics. You gotta go through all of that before you are actually a Mature system.
Then, the final stage is realizing that it's just another input, and is worthless without judgement and thought. That part takes the longest, in my experience.
Report this post to a moderator | IP: Logged
|