• Welcome to OGBoards 10.0, keep in mind that we will be making LOTS of changes to smooth out the experience here and make it as close as possible functionally to the old software, but feel free to drop suggestions or requests in the Tech Support subforum!

Ken Pomery/Tennessee Rank Question (NWT)

Wakeforest22890 said:
It's just another data point. They did have a couple of scrappy wins against Notre Dame and Clemson in overtime sandwiching that State game where they showed they had courage, intestinal fortitude, and are clearly the top of team who has lots of success in the postseason though! That is until they lost to UVA by three, losing their resolve to win close games.
Click to expand...

Notre Dame and Clemson: Not in the tournament.

UVA and NC State: In the tournament.
 
TN played THREE teams that were ranked when they played them. They lost to all of them and none were down to the wire. The spreads were 9, 8, 26, 9, 7.

If you can't play within five points of anyone who is ranked when you play them, it is irrational to say they are a Top 15 team.
 
BiffTannen said:
Notre Dame and Clemson: Not in the tournament.

UVA and NC State: In the tournament.
Click to expand...

Well Clemson is 20 spots better on KenPom than State.

Or how about Pitt holding off UNC, a solid six seed, down the stretch in a game that got closer than it needed to be!?! Surely that showed a lot of courage and provides a lot of predictive value: far more than the fact that they beat the absolute piss out of them for almost the entirety of the game right?!
 
TexasDeac10 said:
This is some millenial BULLSHIT but the Harvard undergrads have a little sports analytics club (Harvard Sports Analytics Collective). They post things to their blog every now and then at www.harvardsportsanalysis.org.

Anyway, they've done an NCAA bracket model every year based on what they call Survival Analysis. Its basic premise is that postseason games are fundamentally different than regular season games, and then it does some statty stuff to account for that. Here's the big 2012 post: http://harvardsportsanalysis.wordpr...t-a-new-model-for-ncaa-tournament-prediction/

They also did one for 2013: http://harvardsportsanalysis.wordpr...-fittest-predicting-the-2013-ncaa-tournament/

Does it work? From 2012:

screen-shot-2013-02-19-at-5-33-26-pm.png
Click to expand...

I assume the stats weighted Tyler's will very heavily in 2009.
 
RJKarl said:
TN played THREE teams that were ranked when they played them. They lost to all of them and none were down to the wire. The spreads were 9, 8, 26, 9, 7.

If you can't play within five points of anyone who is ranked when you play them, it is irrational to say they are a Top 15 team.
Click to expand...

We've had this discussion a hundred times, but the games that don't get included in this calculus are those against similar or inferior opponents. If you're supposed to beat a team by five and you beat them by 30, you've far exceeded what was likely to occur. Of course nobody ever talks about this. There is no stat that says "wins by more than they were supposed to win by" even though this would probably be a more useful tool than looking at the raw margin in games against an arbitrary set number of teams voted on by media and coaches who just value recent wins over overall body of work while doing these rankings.
 
RJKarl said:
TN played THREE teams that were ranked when they played them. They lost to all of them and none were down to the wire. The spreads were 9, 8, 26, 9, 7.

If you can't play within five points of anyone who is ranked when you play them, it is irrational to say they are a Top 15 team.
Click to expand...

They were down by 2 points to Florida with 1:21 left two days ago. I'm not saying that makes them a top 15 team, but you, with all of your basketball knowledge know that spreads of close games usually open up in the last minute or so and can change completely on a single shot or turnover, followed by a parade to the free throw line.

In other words, it's just luck, and it's due to change. Which is why you'll have UT in the Sweet Sixteen if you're smart.
 
Wakeforest22890 said:
Well Clemson is 20 spots better on KenPom than State.

Or how about Pitt holding off UNC, a solid six seed, down the stretch in a game that got closer than it needed to be!?! Surely that showed a lot of courage and provides a lot of predictive value: far more than the fact that they beat the absolute piss out of them for almost the entirety of the game right?!
Click to expand...

Pitt shows me that you don't want to bet on them and probably don't want to bet against them, regardless of any prediction model.
 
TexasDeac10 said:
This is some millenial BULLSHIT but the Harvard undergrads have a little sports analytics club (Harvard Sports Analytics Collective). They post things to their blog every now and then at www.harvardsportsanalysis.org.

Anyway, they've done an NCAA bracket model every year based on what they call Survival Analysis. Its basic premise is that postseason games are fundamentally different than regular season games, and then it does some statty stuff to account for that. Here's the big 2012 post: http://harvardsportsanalysis.wordpr...t-a-new-model-for-ncaa-tournament-prediction/

They also did one for 2013: http://harvardsportsanalysis.wordpr...-fittest-predicting-the-2013-ncaa-tournament/

Does it work? From 2012:

screen-shot-2013-02-19-at-5-33-26-pm.png
Click to expand...

I was concerned that they made up their own little thing and called it survival analysis instead of actually using survival analysis. Interesting stuff. I haven't seen survival analysis used for sports analysis before.
 
UT parlays the Jim Grobe theory: If we lose by a little we actually win by a lot and we will all be happy. Losses are not really losses in bizzaro world.
 
Hey!

Michigan won three straight close games before losing to Michigan State by 14 yesterday. The Wolverines are a stone cold statistical lock to lose in their first close game.
 
If you can't beat anyone good, you aren't good. It's about that simple.

Theories are great to talk about in a bar. Results count in the real world. The results are they lost big to every good team they played. The results are they lost to multiple bad teams as well.

There is no rational way to say such a team could ever be in the Top 15.
 
DeacsATS said:
I'm pretty sure Ron Wellman used advanced statistics in seeding the NCAA tourney. That's what he means when he says, "eye test."

In fact, I bet some of these very same arguments were had amongst the selection committee.

Oh, to be a fly on that wall.
Click to expand...

I really hope one of the committee members was a rj clone. Would have driven Wellman (and everyone else) crazy.
 
RJKarl said:
If you can't beat anyone good, you aren't good. It's about that simple.

Theories are great to talk about in a bar. Results count in the real world. The results are they lost big to every good team they played. The results are they lost to multiple bad teams as well.

There is no rational way to say such a team could ever be in the Top 15.
Click to expand...

So is Virginia not good because they weren't ranked at the time Tennessee played them?
 
qcdeac said:
So is Virginia not good because they weren't ranked at the time Tennessee played them?
Click to expand...

At that time UVA they weren't that good. In fact, the UT game was their wake up call.

It's similar to having played Michigan State when they were down 2-3 starters versus playing them now. A team isn't always as good at one point in the season as they are at another.
 
RJKarl said:
If you can't beat anyone good, you aren't good. It's about that simple.

Theories are great to talk about in a bar. Results count in the real world. The results are they lost big to every good team they played. The results are they lost to multiple bad teams as well.

There is no rational way to say such a team could ever be in the Top 15.
Click to expand...

That's it. I just moved Tennessee into the regional finals.

BTW, how many teams in the field do you think hang with Florida into the final minute and a half. Or beat them by 7 points in the first half?
 
Tennessee's biggest problem is they have to contend with Iowa in the play-in game. That game is likely to be better than almost every single game we see on Thursday or Friday.
 
You must log in or register to reply here.
Back
Top