Tulsa's RPI and NCAA Tourney chances

[Bill Lowery]

If we win out, looking at how we've played the end of the year should mean something to the committee. Especially the narrow losses.

Tulsa Basketball ‏@TUMBasketball
In February, Tulsa is 5-2: atTemple, Houston, atSMU, atUConn, Cincy, atUCF, Temple 2 losses were OT at TEM, and by 2 pts at UConn #ReignCane

 

[lawpoke87]

http://basketballpredictions.blogspot.com/2014/03/how-well-did-computers-predict-field.html
but the Selection Committee itself uses the RPI, because they are apparently clueless to the fact that it's a horrible metric that is 20 years out-of-date and easily gamed."

Odd comment considering the Chairman of the committee last year specifically said that's not true when asked about the omission of CSU and their mid 30s RPI.

 

[Tu Geo]

Didn't realize SMU's RPI was #53 when left out in 2014. that year the Ponies had a record of 23-9.

Go TU!!!!

 

[Tu Geo]

http://basketballpredictions.blogspot.com/2014/03/how-well-did-computers-predict-field.html

"In general, we see the same results every year. The RPI and Sagarin PURE_ELO, as measures of resume quality, are always pretty close to the final Selection Committee results. The Sagarin PREDICTOR and Pomeroy ratings, as measures of team quality, are pretty far off. As I've said to the point of boredom over the last week, this is the way the brackets are always made. You are rewarded for your resume, not how good you are.

For example, UMass is one of the five or six worst teams in as an at-large bid, but they earned their 6 seed (23rd in RPI, 28th in Sagarin PURE_ELO). And Utah was better than more than a dozen at-large teams, yet they in no way deserved an at-large bid (59th in Sagarin PURE_ELO, 80th in RPI).

Now, what about whether RPI or Sagarin PURE_ELO is more accurate? As always, those two metrics are pretty close. The Sagarin PURE_ELO is a much more accurate measure, but the Selection Committee itself uses the RPI, because they are apparently clueless to the fact that it's a horrible metric that is 20 years out-of-date and easily gamed."

Wow friggen NC State sucked in every metric and still got in. There is no question that the selection committee is biased towards the P5.

Hey Law explain that one.

Go TU!!!!

 

[ctt8410]

Odd comment considering the Chairman of the committee last year specifically said that's not true when asked about the omission of CSU and their mid 30s RPI.

Here's what the committee chair said:

""We look at RPI really as an organizer in terms of top 50 and 100, but we spent a lot of time, more time than I remember, in comparing (RPI) in the swings to metrics such as the Sagarin, Ken Pom, BPI," Barnes said. "Those are used in the room as well and talked about quite extensively. We review those metrics each week, and that carried over to this week."

It's my belief (and the numbers bear this out) that RPI sets the framework and they move things around from there. In obvious cases of RPI manipulation (Colorado State last year) they'll take a closer look and say "Wait a second. Why is this team's RPI so out of line with everything else (CSU was 29th in RPI and 68th in KenPom).

There's no fool-proof way to predict the field. If there was, we wouldn't have 100+ different bracketologists all offering different brackets. But, IMO, you still have to start with RPI and work from there.

 

[lawpoke87]

Wow friggen NC State sucked in every metric and still got in. There is no question that the selection committee is biased towards the P5.

Hey Law explain that one.

Go TU!!!!

Have I ever said that the committee wasn't biased toward P5s?

 

[ctt8410]

Wow friggen NC State sucked in every metric and still got in. There is no question that the selection committee is biased towards the P5.

Hey Law explain that one.

Go TU!!!!

I highly recommend reading every year's version of that article. He admits last year's selections were the worst for RPI. Maybe an outlier, maybe not.

2009

2010

2011

2012

2013

2014

2015

 

[Bill Lowery]

Goo

I highly recommend reading every year's version of that article. He admits last year's selections were the worst for RPI. Maybe an outlier, maybe not.

2009

2010

2011

2012

2013

2014

2015

Good stuff CTT. Thanks.

 

[Tu Geo]

I highly recommend reading every year's version of that article. He admits last year's selections were the worst for RPI. Maybe an outlier, maybe not.

2009

2010

2011

2012

2013

2014

2015

Good stuff ctt:

28 out of the last 35 RPI teams (last 5 at large bids each year) are from P5's (80%). Leaving seven from non P5's(20%)

The Average RPI for the 35 teams is #53.

16 of the 35 teams had RPI's above #55. In otherwords an average of 2.28 teams from P5's got in with RPI's above #55 each year.

I guess we must account for the high RPI teams above #55, at least two get an at large every year on average and sometimes three.

Today #58 Michigan and #59 Syracuse fit the bill.

I would think this data bodes well for #53 Vandy and #50 Alabama as well.

At large bids are tough for non Power 5's.

GO TU!!!

 

[Gmoney4WW]

What would an estimate of our Elo Chess be if we won our last 2 games?

 

[TU_BLA]

I highly recommend reading every year's version of that article. He admits last year's selections were the worst for RPI. Maybe an outlier, maybe not.

2009

2010

2011

2012

2013

2014

2015

This is good stuff. Of course it also shows the bias towards P5 teams even with the computers. Florida was a below .500 team that rated highly with all of the computers and wasn't even eligible to garner an at-large bid or NIT invite.

 

[Barlass]

What's the point in punishing Boeheim with a suspension if those game aren't going to count against his team's resume?

 

[drboobay]

And how do the account for it? Throw cuse a couple of exta wins?

 

[Barlass]

And how do the account for it? Throw cuse a couple of exta wins?

Games in which teams are missing key players is already figured into the BPI's rankings. According to Castiglione's quote, that same sort of metric will be applied to Syracuse for being without their head coach.

 

[TU_BLA]

So, cheat, not allow the team to play postseason...suspend the coach as punishment but don't count that against the team. It's like threatening a 4 old that you're going to take away Doc McStuffins for not listening except it's all threats and no actual long term consequence. It's like saying we know you're 3 scholarship players short for the cheating but we will take that into account.

 

[drboobay]

Yes it is absurd. Completely illogical.

 

[Tu Geo]

Bracket Matrix now has Tulsa in as #11 seed.

GO TU!!!

http://www.bracketmatrix.com/

 

[rabidTU]

What's the point in punishing Boeheim with a suspension if those game aren't going to count against his team's resume?

This!

 

[Weatherdemon]

Tulsa a 12 seed play in vs Butler in Lunardi's latest.

http://espn.go.com/mens-college-basketball/bracketology

 

[TUBballJunkie]

I realize that getting in would be better than not getting in but I'm still irritated with second-to-last at-large.

And the seeding order of the 4 AAC teams. Can you guys help me understand UConn at 8? Beat SMU at home but swept by Cinci and Temple? Split with us?

 

[TU Hurricane Warning]

Didn't realize SMU's RPI was #53 when left out in 2014. that year the Ponies had a record of 23-9.

Go TU!!!!

Yes, and SMU played a very very soft schedule and paid for it by not getting in that year. They were pretty strong in conference and the conference was strong overall. Louserville was still in the league that year.

 

[ctt8410]

Can you guys help me understand UConn at 8? Beat SMU at home but swept by Cinci and Temple? Split with us?

Best non-conference win of the group (@Texas). Zero bad losses.

 

[I.I.]

Best non-conference win of the group (@Texas). Zero bad losses.

Amida Brimah is another reason that doesn't show statistically. He is getting stronger each week.

 

[TUBballJunkie]

Best non-conference win of the group (@Texas). Zero bad losses.

Texas is 4 spots ahead of WSU in the AP...that's good for 12-15 slots ahead of us in seedings?

 

[TUBballJunkie]

Oh I see.

The flipping BPI.

 

[Weatherdemon]

Bracket Matrix show Tulsa was in 64 of 94 brackets as of last night but showed Yahoo didn't have us in... but they do know with interesting layout and comments on seeds:

http://sports.yahoo.com/featured/ncaab/bigboard/

Most of the brackets that don't have us listed were last updated Tuesday before the Temple game.

 

[Weatherdemon]

There are 5 brackets that have been updated that have us in that are not recorded.
Only two are updated and don't have us in them.
The rest haven't been updated since Tuesday.

 

[JesseTU]

The play-in game is an excuse to "include" more "mid majors" while at the same time hampering their ability to really advance. A 16 win SEC team? Yes sir, go right in. A 22 win AAC team? Well sir, you can come too. But you have to play a 22 win Big East team for the privilege of playing the 16 win SEC team.

Then they can self fulfill their prophecy when the "mid-major" wears down sooner.

Anyway, I think the BPI is trying to rate "how good teams are" as opposed to assigning a value to their resume. That's not how the game is supposed to be played.

Worse yet, the accounting for injured players is not universally applied apparently. Rather, it is an excuse for "P5" teams when they are missing a player or coach. Just pretend those games didn't really happen and move on. Has anyone heard of a "mid-major" that is benefiting from some sort of record shifting in this regard?

My cynical sense feels like BPI and Lunardi look at a "P5" and set a bar and say "you're in if you can at least get to here." They set a different bar for everyone else and demand that a case be made that they "deserve it." Then they sit back and point to Summit League AQs for statistics on why "Mid-majors" don't deserve more bids.

Which leads me to my final rambling point - expanding the tournament is crap. Unless there is a rule that you have to be above 500 in your league to get in, this is more opportunity to screw over mid-majors. Sure, there might be numerically more mid-majors in a 96 team field... but as a percent of the total and as a piece of the pie it goes further down. Worse yet, it sets up a scenario where nearly every team in the Big 12 or ACC gets to go tot he tournament every year, instant recruiting advantage.

Once 16 seeds beat #1 seeds from time to time, lets expand. Otherwise, we are just throwing more turds in the punch bowl.

/ramble

 

[TU_BLA]

Michigan as a 9 seed is an absolute travesty. Providence being in is sort of a joke now...that team is free falling faster than Peggy Hill when her chute didn't open. Not only did they lose to Seton Hall last night, they lost by 20. Syracure (59) and Cincinnati (60) getting higher seeds than us (34) is pretty ridiculous. So...it no longer matters if you play a tough schedule, win tough games in and out of conference, you have to have a likable name and coach as well to get higher consideration.

 

[TU Hurricane Warning]

I honestly don't think they can consider Tulsa or AAC as "Mid Major". I think it is more of just not being very well respected at this point. It's not like we are going out collectively and setting the world on fire in our respective OOC games. To me more than anything is what has been hampering the leagues perception IMHO.

 

[Weatherdemon]

USA Today today.

 

[ctt8410]

We want SMU and Temple each to lose one more.

FWIW, KenPom is now predicting a 3-way tie at 13-5.

 

[hotterthanyourhusbandTU]

I haven't seen a single article, news feed, or sports editorial that refers to TU or the AAC as "mid-major".

 

[ctt8410]

I haven't seen a single article, news feed, or sports editorial that refers to TU or the AAC as "mid-major".

 

[Tu Geo]

Gosh didn't Bill Self call Tulsa a High Major 15 years ago.

I thought we put that argument to bed a long time ago.

Not Mid Majors:

Butler
Gonzaga
San Diego St.
Wichita St
Memphis
Cincy
Temple
VCU
Houston
BYU
Dayton
St Joe's
Geo'town
Marquette
Prov
St Johns
Villanova
Xavier
Valpo

I'm probably leaving some out.

These programs year and year out can compete in any conference any year.

GO TU!!!!

 

[Tulsa_]

Gosh didn't Bill Self call Tulsa a High Major 15 years ago.

I thought we put that argument to bed a long time ago.

Not Mid Majors:

Butler
Gonzaga
San Diego St.
Wichita St
Memphis
Cincy
Temple
VCU
Houston
BYU
Dayton
St Joe's
Geo'town
Marquette
Prov
St Johns
Villanova
Xavier
Valpo

I'm probably leaving some out.

These programs year and year out can compete in any conference any year.

GO TU!!!!

How about UCONN... so a Mid Major won the National Championship 2 years ago ?

 

[hotterthanyourhusbandTU]

Wow...the only comment I've seen referencing TU as a "mid-major" is from a local guy.

Thanks.

 

[TUBballJunkie]

Wow...the only comment I've seen referencing TU as a "mid-major" is from a local guy.

Thanks.

ESPN does ALL the time during Big XII games in the reverse...graphics and stats on the Power 5 as if it's relevant in basketball.

 

[TUBballJunkie]

FWIW, KenPom is now predicting a 3-way tie at 13-5.

That'd make for an interesting tie break for seeding.

Would the league name co-champions officially?

 

[Gmoney4WW]

That'd make for an interesting tie break for seeding.

Not really. It would be the same if SMU and/or Temple won out. SMU tied or not, isn't seeded. Temple wins the tie breaker w/ us based on their record against Cinci and/or Uconn, thus nothing changes, tie or no tie.