Fluoxetine is the NCAA
- Thread starter HRTheCard
- Start date
They clearly don't realize we played our first two games on the road.
"The NET (NCAA Evaluation Tool) is the NCAA’s alternative to the RPI and it is being used for the first time this season. It is the primary metric the selection committee will use to choose and seed teams for the tournament next spring.
The NET is unique in that, unlike the RPI, it will take into account margin of victory. It also relies on game results, strength of schedule, game location, net offensive and defensive efficiency, and the quality of wins and losses."
So Kentucky gets a play in game right now? I'm hoping this shapes up a little better once the season has progressed.
It makes sense to me, honestly. We haven't been impressive at all. Take preseason rankings out of it. I'm absolutely positive there are 60 plus teams with a better resume than us right now.
Big Ten has been slaying. 4 of the top 8 teams in the Net Rankings. They also have 7 of the top 25 teams in the new AP Poll. ACC/Big Ten Challenge should be a good one.
Seems odd UNC isn't higher then tbh. Margin of victory has been high, game locations, and offensive efficiency should be good. The quality of teams played hasn't been great yet but hasn't been awful. Of course, there is that defensive efficiency which is probably ugly. This is way different than every other metric out there, which seems like an NCAA thing to do.
The NET will model the RPI in the sense that early in the season when there isn't enough data to connect everything, there will be massive fluctuation and wildly varying results compared to more tested models such as Pomeroy that use preseason predictors to stabilize the early season ratings.
However, unlike the RPI, with all of the other factors involved, it should produce a more cohesive rating late in the season when selections are made.
Time will tell if it's more accurate than Pomeroy and/or Sagarin's formula, but I think we can be sure it will be better than the archaic RPI in a couple months.
However, unlike the RPI, with all of the other factors involved, it should produce a more cohesive rating late in the season when selections are made.
Time will tell if it's more accurate than Pomeroy and/or Sagarin's formula, but I think we can be sure it will be better than the archaic RPI in a couple months.
I assume that this tool, like most stats-based tools, is nigh-useless this early into the year and will improve over time. It doesn't appear to have a preseason ranking component (or prior season ranking) like most systems (i.e. KenPom), so it will be more volatile than most projection systems.
Nate Silver's critique seems spot-on. Why go to the drawing board when you can rely on other, smarter analysis that withstood scrutiny over time?
Nate Silver's critique seems spot-on. Why go to the drawing board when you can rely on other, smarter analysis that withstood scrutiny over time?
@JVDBeak'em weekly aggregate rankings about to get real weird.
Last edited:
Nahh. This thing is doomed. It is faulty from the get-go. They aren't factoring in strength of the opponent when looking at the efficiency margins, just the raw efficiency margin. So beating Central Connecticut State by 40 gets you more credit than beating Duke by 8.
It's a complete embarrassment by the NCAA. Pure buffoonery.
I think the margin of victory scale doesn’t go any higher than 10. So if you beat Kentucky by 34 it’s the same as beating Lasalle by 10. Still dumb.
Garbage in, garbage out. Anything this early in the season that doesn't rely on starting rankings isn't going to make any sense.
Oh, I remember that. So this is those rankings. I could be wrong, but IIRC it's 10-15. Exactly right about Duke beating UK by all those points, it caps out.
I looked for that last thread. We were talking about ppl pulling back b/c after a point in time, it wouldn't matter.
This their newest excuse to give Duke the easiest road to the Final Four again? Sounds about right.
The margin of victory component is capped at 10, yes. But I'm talking about the net efficiency margin. Which essentially measures margin of victory.
Offensive efficiency per possession minus defensive efficiency per possession. Except the morons aren't factoring in strength of the opponent. So when you beat Southeastern Mo State by 45, your net efficiency gets an enormous boost.
KenPom and all the others control for the relative strength of the opponent. That's why they look much more appropriate, even early on.
Basically KenPom on steroids. It looks bad early; Loyola Marymount 1oth; OSU 1st. But by mid-conference, it will look much different.
Similar threads
