ADVERTISEMENT

BACATOLOGY: NCAA ANALYSIS 3/7......NEW UPDATE 3/12 1PM PAGE 37

Man I respect you but the only advantage the current approach has over literally just letting a computer algorithm select the teams is the current system is more entertaining.

Watching humans try to parse a big data set manually when we have tools to do that for us (i.e. data science or “science” as you put it) is laughable and basically just shows how much the world has passed the old guys by.

Alas, they are still in control.

For now.
When the "science" bases anything on expected outcome vs. actual outcome it is a joke.
Computers are only as good as the program and info put in. Being penalized when you win by six but were expected to win by 12 is ridiculous.
Oh. And "**** you" youngster.
 
Duke how do you not get up for this … Coach K last ever ACC game, in NYC in the ACC Championship against the 7 seed … all the money dumped into that program, if you’re a bball player you do not want for anything … just pathetic on the court
 
When the "science" bases anything on expected outcome vs. actual outcome it is a joke.
Computers are only as good as the program and info put in. Being penalized when you win by six but were expected to win by 12 is ridiculous.
Oh. And "**** you" youngster.
Love your last sentence. Some of us youngsters need their ass whooped. You go pops! Haha
 
"And 1!" as shot rims out and Tech takes it up for another easy basket for their own And 1 😂
 
Dayton is our destiny. I believe it
This game either was thrown or something is really off here.

Any of the B1G quarterfinalists would have beaten this Duke team soundly. Soundly like by twice what they after losing by now.

No effort.
 
This game either was thrown or something is really off here.

Any of the B1G quarterfinalists would have beaten this Duke team soundly. Soundly like by twice what they after losing by now.

No effort.
Va tech looked awesome and running it up
 
Cmon Fluox. You seem to be too good of a quants guy to post something like that.

Sure computers can analyze a huge data set more efficiently than humans ever could. But that analysis is only as good as the algorithms selected, and the algorithms are only as good as the assumptions built into them.

The "old guys" might have the experience and expertise to understand the limitations of the algorithms and assumptions.
Your second sentence is a great point, and I certainly would take the committee method over just using the current NET for example. If you were actually going to go that route you would need someone to design a transparent system that gives the right incentives.
It would be nice if the NCAA (and other ranking services like KenPom or Sagarin) published the calculations that they used. That would allow others to critique the algorithms, similar to a peer review in scientific journals, and help clarify the limitations of the rankings. Maybe we then get a better NET, or maybe we get the same NET while understanding the limitations.
It seems the Selection Committee has already concluded that the NET has limitations at the fringes (or at least one fringe), because they didn't put Houston in their top 16 despite a top 5 NET.

That isn't evidence that the world has passed them by. It is evidence that they have the experience to recognize that the NET is limited by the algorithm and assumptions built into it.
I don't know how to unsplit these but yes I agree pretty much 100% with it.
When the "science" bases anything on expected outcome vs. actual outcome it is a joke.
Computers are only as good as the program and info put in. Being penalized when you win by six but were expected to win by 12 is ridiculous.
I'm saying a computer algorithm could do a better job. The NET is absolutely not that algorithm, and neither are Bart, Kenpom, or other predictive models. Kenpom is good for what it's for which is predicting which teams will win games going forward, but it's not designed for selecting a tournament field.
Oh. And "**** you" youngster.
:P
 
This game either was thrown or something is really off here.

Any of the B1G quarterfinalists would have beaten this Duke team soundly. Soundly like by twice what they after losing by now.

No effort.
Duke was expected to win by 5 and they lost by 15, that's bad but not really that weird.
 
ADVERTISEMENT
ADVERTISEMENT