NCAA Tourney Talk

[calfan]

Doc,

I have been thinking about a list of schools for a possible new conference.
I would start with what we already have in the MVC, add some other good schools in the area. See what you think of this list.

WSU, Creighton, SIU, St. Louis University, Bradley, Missouri State, Illinois State, New Mexico, New Mexico State, and Butler (10 schools).

2 others from this list- TCU, SMU, Kent State, and Western Kentucky.

What do you think?

[goeagles]

Hello! First time poster..Winthrop Alum. here..I am just wondering how you can substantiate these theories of the NCAA splitting D1 into two levels. Realistically I just can't see a change coming to the way the NCAA chooses who is invited to the tournament anytime soon even though some of you guys may disagree. The way the system works now is not perfect but would it really be neccessary to split the D1 college ranks into a so called BSC and subdivision levels? Everyone realizes that the best 65 teams never make it into the field at the end of the year however, the small league teams that make it into the tournament are justified because they have an unlikely/nearly-impossible chance to make it through the first round given a 15 or 16 seed. More and more teams are gaining D1 status and yes its a bit ridiculous that there are what..335+ D1 teams that can potentially play their way into the field of 65 however, the snubs and the bubble and the conference-tourney runs by upset minded teams are what make Selection Sunday and Championship week so interesting. I also realize--like someone mentioned before--that scheduling a good OOC is extremely difficult because high schools look at playing mids and low-majors as lose/lose situations, therefore its so tough to get a good win on a neutral court or home court much less on the road. At least WSU is in the Valley and has earned enough respect along with the rest of your confernce mates to earn strong consideration for getting multiple teams into the field year in and year out. Although it's difficult for Mids and Lows to get the at-large bids because of several circumstances and of course some high major teams are left out every year I don't think this warrants a shakeup of such large implications as splitting into a BCS and subdivision

ok rant over!! good luck to you guys this offseason--looking forward to next year already!

[Snapshot9]

With the Big Dance, the NIT, and now the C??, it is getting a little overboard.

What I propose is very simple.

For all four 16 team brackets, have a play-in game for each one, that would make it 68 teams, and the 3 or 4 in question could then be included. Or, you could have a play-in game for all 13-16 seeds,and that would bring it to 80 teams total.

[ShockerFever]

With the Big Dance, the NIT, and now the C??, it is getting a little overboard.

What I propose is very simple.

For all four 16 team brackets, have a play-in game for each one, that would make it 68 teams, and the 3 or 4 in question could then be included. Or, you could have a play-in game for all 13-16 seeds,and that would bring it to 80 teams total.

Then you have teams 69-72 complaining about not being included. Making the tourney bigger does not solve the problem.

While I agree there's too many teams in Division I as is, don't ruin a good thing. The format right now is perfect. The only thing that's wrong is the bag of tools incorporating teams into the format. That's where the change needs to come at.

[Snapshot9]

Well, that would be true, no matter what number of teams got in, wouldn't it?

The reason, nothing can ever be nailed down is because:

RPI = won/lost plus strength of schedule; plus strength of schedule of opponents; strength of schedule of opponents opponents.

BUT:

Nothing is accounted for as to HOW BAD a team lost or won.
(Yes, I am aware of bad and good nights for a team, plus injuries
they may have had at the time)

There is a difference in playing a team with an RPI forty points above you, and losing by 4 points as opposed to losing by 16, 20, or 24 points.

It seems to be amazing that all these 'Basketball experts' can NOT develop a formula that accurately resembles real life in D-1 Basketball. It doesn't seem to me that the data has been truly analyzed properly to come up with a more accurate formula for ranking, or perhaps some data was eliminated simply because of the additional space that would be required to be kept on personal PC's.

I am aware of what this would entail because I worked in middle/high level of systems development for many years, and I had all the math courses available in undergrad/grad school through Operations Research.

[ISASO]

If human beings can figure out how to land a man on the moon using slide rules, this basketball tournament thing should be handled in one afternoon. However, it is extremely difficult for a programmer to develop an application when you have committee of people changing the rules underneath you all the time.

I deal with that every day, and it's no fun. Let all the "eye testers" have some input into the development process then get the math whizzes involved, determine your datapoints and whatever weighting is appropriate - then give the committee a small override factor if they want to "tweak" it a little - then let it go to town.

[RoyalShock]

Two changes need to be made to the RPI:

1) The road bonus/penalty needs to be changed from 1.4/.6 to 1.2/.8 or 1.1/.9. A home team is only 10% more likely to win at home than on a neutral court, yet the RPI applies a factor of 40%.

2) Factor in margin of victory.

Then the rating will be useful for more than just strength of schedule.

[CharlieHog]

2) Factor in margin of victory.

Margin of Victory is a dangerous way to go. It's only somewhat useful measure in football, I'm not sure it's at all useful in basketball. In the last minute a close game can become a blowout if the losing team fouls. Or a blowout can look close if the winning team empties their bench. Plus it only encourages teams to run the score up, that's why it was thrown out of the football equation.

[Snapshot9]

ISASO and ROYAL, good comments, valid points.

I still see margin of victory to be relevant in Basketball with levels established and weighted appropriately:

3 or less Victory
4-7
8-11
11-14
15-19
20 or more

Probably 80% of the games would fall in the first 3 levels.

[Awesome Sauce Malone]

I think the margin isnt the best thing.

How many times have we heard "Team A lost by 25 but it wasnt really that close"

[RoyalShock]

Margin of victory is the only thing that can distinguish between two good teams playing versus a good and bad team playing. And you cap the margin at, say 20 points, as Snapshot suggested.

If a team empties its bench it's going to be close to a 20-point lead anyway and too late to change it much. Just don't over-emphasize MOV in the equation.

I know Sagarin factors it in and I think Pomeroy does, too.

[7hottamales]

Two changes need to be made to the RPI:

1) The road bonus/penalty needs to be changed from 1.4/.6 to 1.2/.8 or 1.1/.9. A home team is only 10% more likely to win at home than on a neutral court, yet the RPI applies a factor of 40%.

2) Factor in margin of victory.

Then the rating will be useful for more than just strength of schedule.

It's not that complicated. Ken Pom, Sagarin, and others already include all the relevant data.

Their formulas incorporate the exact statistically significant advantage/disadvantage to home/road/neutral scenarios. There is no need to arbitrarily throw a blanket ratio over it.

Ken Pom's formula is nearly flawless. The best check for stuff like this is Vegas lines. Pom's formula nearly always matches a prediction very close to the spread and the over/under. They are obviously using very similar data and methods.

Like I've said in the past, the RPI is a political tool designed to reward desired behavior. That's fine, but it gets really annoying when fans try to attach validity to in in determining how good teams are.

[7hottamales]

The reason, nothing can ever be nailed down is because:

RPI = won/lost plus strength of schedule; plus strength of schedule of opponents; strength of schedule of opponents opponents.

One issue here...saying it measures strength of schedule in the actual formula is a little bit of a misconception. It measures nothing more than winning percentage of the three components. So if your opponents have a great winning percentage and you beat them, your RPI is going to sky rocket, even if they are beating North Florida, Presbyterian, & NJ Inst of Technology that won a combined 4 games this year...even if you beat them all by 1 point and especially if you managed to beat them at their place. :D

[7hottamales]

2) Factor in margin of victory.

Margin of Victory is a dangerous way to go. It's only somewhat useful measure in football, I'm not sure it's at all useful in basketball. In the last minute a close game can become a blowout if the losing team fouls. Or a blowout can look close if the winning team empties their bench. Plus it only encourages teams to run the score up, that's why it was thrown out of the football equation.

It's the opponents job to keep a team from "running up the score"...if they are unable to, isn't that somewhat valuable information when judging how good both teams are? Include all the data points and that effect will end up having the correct amount of relevance...but you can't tell me that the ability to enforce your will on someone or the inability to prevent it is useless info.

This is a competitive environment.

A compromise could be to limit the impact at 25 points margin.


Edit: It looks like I'm a little late to this party.

[martymoose]

Memphis is damn good.

I think its more like "Tulsa is terrible." Memphis will be out in the Sweet 16.

You sound like every KU fan I've talked to this year...too bad Memphis is actually backing it up with wins.

I just don't think Memphis is that great. They've beaten up on terrible teams all season. They've played 34 games and 6 of those are against what I call "quality opponents." Its just an opinion, no need to start KU bashing just because I don't think Memphis will make it past the 4 seed.


FYI: The RPI is overrated, imo, so bringing it out to "prove" things to me isn't going to work.

Change your mind yet????

When KU played in Texas, they lost...Memphis wins by 20. I know we're in the middle of Big 12 country, but maybe people will start to open their eyes.